Merge pull request #693 from puppetlabs/release_prep_3-8-0

Added a action to generate release notes

.github/workflows/auto_release_prep.yml

@@ -0,0 +1,12 @@
+name: Automated release prep
+
+on:
+  workflow_dispatch:
+
+jobs:
+  auto_release_prep:
+    uses: puppetlabs/release-engineering-repo-standards/.github/workflows/auto_release_prep.yml@v1
+    secrets: inherit
+    with:
+      project-type: ruby
+      version-file-path: lib/vmpooler/version.rb

.github/workflows/dependabot_merge.yml

@@ -0,0 +1,8 @@
+name: Dependabot auto-merge
+
+on: pull_request
+
+jobs:
+  dependabot_merge:
+    uses: puppetlabs/release-engineering-repo-standards/.github/workflows/dependabot_merge.yml@v1
+    secrets: inherit

.github/workflows/ensure_label.yml

@@ -0,0 +1,8 @@
+name: Ensure label
+
+on: pull_request
+
+jobs:
+  ensure_label:
+    uses: puppetlabs/release-engineering-repo-standards/.github/workflows/ensure_label.yml@v1
+    secrets: inherit

.github/workflows/release.yml

@@ -7,10 +7,10 @@ jobs:
     runs-on: ubuntu-latest
     if: github.repository == 'puppetlabs/vmpooler'
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: Get Current Version
-        uses: actions/github-script@v6
+        uses: actions/github-script@v7
         id: cv
         with:
           script: |
@@ -29,37 +29,6 @@ jobs:
           echo "version=$version" >> $GITHUB_OUTPUT
           echo "Found version $version from lib/vmpooler/version.rb"
 
-      - name: Generate Changelog
-        uses: docker://githubchangeloggenerator/github-changelog-generator:1.16.2
-        with:
-          args: >-
-            --future-release ${{ steps.nv.outputs.version }}
-        env:
-          CHANGELOG_GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-
-      - name: Validate Changelog
-        run : |
-          set -e
-          if [[ -n $(git status --porcelain) ]]; then
-            echo "Here is the current git status:"
-            git status
-            echo
-            echo "The following changes were detected:"
-            git --no-pager diff
-            echo "Uncommitted PRs found in the changelog. Please submit a release prep PR of changes after running `./update-changelog`"
-            exit 1
-          fi
-
-      - name: Generate Release Notes
-        uses: docker://githubchangeloggenerator/github-changelog-generator:1.16.2
-        with:
-          args: >-
-            --since-tag ${{ steps.cv.outputs.result }}
-            --future-release ${{ steps.nv.outputs.version }}
-            --output release-notes.md
-        env:
-          CHANGELOG_GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-
       - name: Tag Release
         uses: ncipollo/release-action@v1
         with:
@@ -70,10 +39,10 @@ jobs:
           prerelease: false
 
       # This step should closely match what is used in `docker/Dockerfile` in vmpooler-deployment
-      - name: Install Ruby jruby-9.4.3.0
+      - name: Install Ruby jruby-9.4.12.1
         uses: ruby/setup-ruby@v1
         with:
-          ruby-version: 'jruby-9.4.3.0'
+          ruby-version: 'jruby-9.4.12.1'
 
       - name: Build gem
         run: gem build *.gemspec

.github/workflows/security.yml

@@ -11,7 +11,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - name: checkout repo content
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
       with:
         fetch-depth: 1
     - name: setup ruby
@@ -22,7 +22,7 @@ jobs:
     - name: check lock
       run: '[ -f "Gemfile.lock" ] && echo "package lock file exists, skipping" || bundle lock'
     # install java
-    - uses: actions/setup-java@v3
+    - uses: actions/setup-java@v4
       with:
         distribution: 'temurin' # See 'Supported distributions' for available options
         java-version: '17'

.github/workflows/testing.yml

@@ -18,9 +18,9 @@ jobs:
     strategy:
       matrix:
         ruby-version:
-          - 'jruby-9.4.3.0'
+          - 'jruby-9.4.12.1'
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Set up Ruby
       uses: ruby/setup-ruby@v1
       with:
@@ -34,9 +34,9 @@ jobs:
     strategy:
       matrix:
         ruby-version:
-          - 'jruby-9.4.3.0'
+          - 'jruby-9.4.12.1'
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
     - name: Set up Ruby
       uses: ruby/setup-ruby@v1
       with:

.github_changelog_generator

@@ -1,3 +1,5 @@
 project=vmpooler
 user=puppetlabs
 exclude_labels=maintenance
+github-api=https://api.github.com
+release-branch=main

CHANGELOG.md

@@ -1,5 +1,91 @@
 # Changelog
 
+## [3.8.1](https://github.com/puppetlabs/vmpooler/tree/3.8.1) (2026-01-14)
+
+[Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.7.0...3.8.1)
+
+**Implemented enhancements:**
+
+- \(P4DEVOPS-9434\) Add rate limiting and input validation security enhancements [\#690](https://github.com/puppetlabs/vmpooler/pull/690) ([mahima-singh](https://github.com/mahima-singh))
+- \(P4DEVOPS-8570\) Add Phase 2 optimizations: status API caching and improved Redis pipelining [\#689](https://github.com/puppetlabs/vmpooler/pull/689) ([mahima-singh](https://github.com/mahima-singh))
+- \(P4DEVOPS-8567\) Add DLQ, auto-purge, and health checks for Redis queues [\#688](https://github.com/puppetlabs/vmpooler/pull/688) ([mahima-singh](https://github.com/mahima-singh))
+- Add retry logic for immediate clone failures [\#687](https://github.com/puppetlabs/vmpooler/pull/687) ([mahima-singh](https://github.com/mahima-singh))
+
+**Fixed bugs:**
+
+- \(P4DEVOPS-8567\) Prevent VM allocation for already-deleted request-ids [\#688](https://github.com/puppetlabs/vmpooler/pull/688) ([mahima-singh](https://github.com/mahima-singh))
+- Prevent re-queueing requests already marked as failed [\#687](https://github.com/puppetlabs/vmpooler/pull/687) ([mahima-singh](https://github.com/mahima-singh))
+
+## [3.7.0](https://github.com/puppetlabs/vmpooler/tree/3.7.0) (2025-06-04)
+
+[Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.6.0...3.7.0)
+
+**Implemented enhancements:**
+
+- \(P4DEVOPS-6096\) Include VMs that have been requested but not moved to pending when getting queue metrics [\#681](https://github.com/puppetlabs/vmpooler/pull/681) ([isaac-hammes](https://github.com/isaac-hammes))
+- Bump redis from 5.1.0 to 5.2.0 [\#675](https://github.com/puppetlabs/vmpooler/pull/675) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rake from 13.1.0 to 13.2.1 [\#673](https://github.com/puppetlabs/vmpooler/pull/673) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump redis from 5.0.8 to 5.1.0 [\#665](https://github.com/puppetlabs/vmpooler/pull/665) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rspec from 3.12.0 to 3.13.0 [\#664](https://github.com/puppetlabs/vmpooler/pull/664) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump opentelemetry-sdk from 1.3.1 to 1.4.0 [\#663](https://github.com/puppetlabs/vmpooler/pull/663) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump mock\_redis from 0.43.0 to 0.44.0 [\#662](https://github.com/puppetlabs/vmpooler/pull/662) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump mock\_redis from 0.41.0 to 0.43.0 [\#658](https://github.com/puppetlabs/vmpooler/pull/658) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump net-ldap from 0.18.0 to 0.19.0 [\#653](https://github.com/puppetlabs/vmpooler/pull/653) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump sinatra from 3.1.0 to 3.2.0 [\#652](https://github.com/puppetlabs/vmpooler/pull/652) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump mock\_redis from 0.40.0 to 0.41.0 [\#650](https://github.com/puppetlabs/vmpooler/pull/650) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump mock\_redis from 0.37.0 to 0.40.0 [\#643](https://github.com/puppetlabs/vmpooler/pull/643) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rake from 13.0.6 to 13.1.0 [\#638](https://github.com/puppetlabs/vmpooler/pull/638) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump thor from 1.2.2 to 1.3.0 [\#635](https://github.com/puppetlabs/vmpooler/pull/635) ([dependabot[bot]](https://github.com/apps/dependabot))
+
+**Fixed bugs:**
+
+- Bump opentelemetry-sdk from 1.4.0 to 1.4.1 [\#672](https://github.com/puppetlabs/vmpooler/pull/672) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rack from 2.2.8.1 to 2.2.9 [\#671](https://github.com/puppetlabs/vmpooler/pull/671) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump thor from 1.3.0 to 1.3.1 [\#668](https://github.com/puppetlabs/vmpooler/pull/668) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rack from 2.2.8 to 2.2.8.1 [\#666](https://github.com/puppetlabs/vmpooler/pull/666) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump concurrent-ruby from 1.2.2 to 1.2.3 [\#660](https://github.com/puppetlabs/vmpooler/pull/660) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump puma from 6.4.1 to 6.4.2 [\#655](https://github.com/puppetlabs/vmpooler/pull/655) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump puma from 6.4.0 to 6.4.1 [\#654](https://github.com/puppetlabs/vmpooler/pull/654) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Update opentelemetry-instrumentation-http\_client requirement from = 0.22.2 to = 0.22.3 [\#646](https://github.com/puppetlabs/vmpooler/pull/646) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Update opentelemetry-instrumentation-concurrent\_ruby requirement from = 0.21.1 to = 0.21.2 [\#645](https://github.com/puppetlabs/vmpooler/pull/645) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump opentelemetry-sdk from 1.3.0 to 1.3.1 [\#642](https://github.com/puppetlabs/vmpooler/pull/642) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump prometheus-client from 4.2.1 to 4.2.2 [\#641](https://github.com/puppetlabs/vmpooler/pull/641) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump redis from 5.0.7 to 5.0.8 [\#637](https://github.com/puppetlabs/vmpooler/pull/637) ([dependabot[bot]](https://github.com/apps/dependabot))
+- \(RE-15817\) Reword fail warning and get error from redis before generating message [\#633](https://github.com/puppetlabs/vmpooler/pull/633) ([isaac-hammes](https://github.com/isaac-hammes))
+
+**Merged pull requests:**
+
+- \(P4DEVOPS-6096\) Fix gems to prevent warnings in logs [\#685](https://github.com/puppetlabs/vmpooler/pull/685) ([isaac-hammes](https://github.com/isaac-hammes))
+- \(maint\) Revert gems to last release [\#683](https://github.com/puppetlabs/vmpooler/pull/683) ([isaac-hammes](https://github.com/isaac-hammes))
+- Bump actions/setup-java from 3 to 4 [\#648](https://github.com/puppetlabs/vmpooler/pull/648) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump actions/github-script from 6 to 7 [\#644](https://github.com/puppetlabs/vmpooler/pull/644) ([dependabot[bot]](https://github.com/apps/dependabot))
+
+## [3.6.0](https://github.com/puppetlabs/vmpooler/tree/3.6.0) (2023-10-05)
+
+[Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.5.1...3.6.0)
+
+**Fixed bugs:**
+
+- \(maint\) Fix message for timeout notification. [\#624](https://github.com/puppetlabs/vmpooler/pull/624) ([isaac-hammes](https://github.com/isaac-hammes))
+
+**Merged pull requests:**
+
+- Bump rubocop from 1.56.3 to 1.56.4 [\#631](https://github.com/puppetlabs/vmpooler/pull/631) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump puma from 6.3.1 to 6.4.0 [\#630](https://github.com/puppetlabs/vmpooler/pull/630) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rubocop from 1.56.2 to 1.56.3 [\#628](https://github.com/puppetlabs/vmpooler/pull/628) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump actions/checkout from 3 to 4 [\#627](https://github.com/puppetlabs/vmpooler/pull/627) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Update opentelemetry-resource\_detectors requirement from = 0.24.1 to = 0.24.2 [\#626](https://github.com/puppetlabs/vmpooler/pull/626) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rubocop from 1.56.1 to 1.56.2 [\#625](https://github.com/puppetlabs/vmpooler/pull/625) ([dependabot[bot]](https://github.com/apps/dependabot))
+- Bump rubocop from 1.56.0 to 1.56.1 [\#623](https://github.com/puppetlabs/vmpooler/pull/623) ([dependabot[bot]](https://github.com/apps/dependabot))
+
+## [3.5.1](https://github.com/puppetlabs/vmpooler/tree/3.5.1) (2023-08-24)
+
+[Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.5.0...3.5.1)
+
+**Fixed bugs:**
+
+- \(maint\) Fix bugs from redis and timeout notification updates. [\#621](https://github.com/puppetlabs/vmpooler/pull/621) ([isaac-hammes](https://github.com/isaac-hammes))
+
 ## [3.5.0](https://github.com/puppetlabs/vmpooler/tree/3.5.0) (2023-08-23)
 
 [Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.4.0...3.5.0)
@@ -160,6 +246,7 @@
 - \(maint\) Adding a provider method tag\_vm\_user [\#469](https://github.com/puppetlabs/vmpooler/pull/469) ([sbeaulie](https://github.com/sbeaulie))
 - Update testing.yml [\#468](https://github.com/puppetlabs/vmpooler/pull/468) ([sbeaulie](https://github.com/sbeaulie))
 - Move vsphere specific methods out of vmpooler [\#467](https://github.com/puppetlabs/vmpooler/pull/467) ([sbeaulie](https://github.com/sbeaulie))
+- Release prep for v2.0.0 [\#465](https://github.com/puppetlabs/vmpooler/pull/465) ([genebean](https://github.com/genebean))
 
 ## [2.0.0](https://github.com/puppetlabs/vmpooler/tree/2.0.0) (2021-12-08)
 
@@ -168,7 +255,6 @@
 **Merged pull requests:**
 
 - Use credentials file for Rubygems auth [\#466](https://github.com/puppetlabs/vmpooler/pull/466) ([genebean](https://github.com/genebean))
-- Release prep for v2.0.0 [\#465](https://github.com/puppetlabs/vmpooler/pull/465) ([genebean](https://github.com/genebean))
 - Add Gem release workflow [\#464](https://github.com/puppetlabs/vmpooler/pull/464) ([genebean](https://github.com/genebean))
 - Update icon in the readme to reference this repo [\#463](https://github.com/puppetlabs/vmpooler/pull/463) ([genebean](https://github.com/genebean))
 - \(DIO-2769\) Move vsphere provider to its own gem [\#462](https://github.com/puppetlabs/vmpooler/pull/462) ([genebean](https://github.com/genebean))
@@ -213,13 +299,17 @@
 **Merged pull requests:**
 
 - \(POOLER-176\) Add Operation Label to User Metric [\#455](https://github.com/puppetlabs/vmpooler/pull/455) ([yachub](https://github.com/yachub))
-- Update OTel gems to 0.15.0 [\#450](https://github.com/puppetlabs/vmpooler/pull/450) ([genebean](https://github.com/genebean))
-- Migrate testing to GH Actions from Travis [\#446](https://github.com/puppetlabs/vmpooler/pull/446) ([genebean](https://github.com/genebean))
 
 ## [1.1.0-rc.1](https://github.com/puppetlabs/vmpooler/tree/1.1.0-rc.1) (2021-08-11)
 
 [Full Changelog](https://github.com/puppetlabs/vmpooler/compare/1.0.0...1.1.0-rc.1)
 
+**Merged pull requests:**
+
+- \(POOLER-176\) Add Operation Label to User Metric [\#454](https://github.com/puppetlabs/vmpooler/pull/454) ([yachub](https://github.com/yachub))
+- Update OTel gems to 0.15.0 [\#450](https://github.com/puppetlabs/vmpooler/pull/450) ([genebean](https://github.com/genebean))
+- Migrate testing to GH Actions from Travis [\#446](https://github.com/puppetlabs/vmpooler/pull/446) ([genebean](https://github.com/genebean))
+
 ## [1.0.0](https://github.com/puppetlabs/vmpooler/tree/1.0.0) (2021-02-02)
 
 [Full Changelog](https://github.com/puppetlabs/vmpooler/compare/0.18.2...1.0.0)
@@ -728,13 +818,13 @@
 - Do not have a hardcoded list of VM providers [\#230](https://github.com/puppetlabs/vmpooler/issues/230)
 - Use a dynamic check\_pool period [\#226](https://github.com/puppetlabs/vmpooler/issues/226)
 - vmpooler doesn't seem to recognize ready VMs [\#218](https://github.com/puppetlabs/vmpooler/issues/218)
-- `find_vmdks` in `vsphere_helper` should not use `vmdk_datastore._connection` [\#213](https://github.com/puppetlabs/vmpooler/issues/213)
+- `find\_vmdks` in `vsphere\_helper` should not use `vmdk\_datastore.\_connection` [\#213](https://github.com/puppetlabs/vmpooler/issues/213)
-- `get_base_vm_container_from` in `vsphere_helper` ensures the wrong connection [\#212](https://github.com/puppetlabs/vmpooler/issues/212)
+- `get\_base\_vm\_container\_from` in `vsphere\_helper` ensures the wrong connection [\#212](https://github.com/puppetlabs/vmpooler/issues/212)
 - `close` in vsphere\_helper throws an error if a connection was never made [\#211](https://github.com/puppetlabs/vmpooler/issues/211)
-- `find_pool` in vsphere\_helper.rb has subtle errors [\#210](https://github.com/puppetlabs/vmpooler/issues/210)
+- `find\_pool` in vsphere\_helper.rb has subtle errors [\#210](https://github.com/puppetlabs/vmpooler/issues/210)
-- `find_pool` in vsphere\_helper tends to throw instead of returning nil for missing pools [\#209](https://github.com/puppetlabs/vmpooler/issues/209)
+- `find\_pool` in vsphere\_helper tends to throw instead of returning nil for missing pools [\#209](https://github.com/puppetlabs/vmpooler/issues/209)
 - Vsphere connections are always insecure \(Ignore cert errors\) [\#207](https://github.com/puppetlabs/vmpooler/issues/207)
-- `find_folder` in vsphere\_helper.rb has subtle errors [\#204](https://github.com/puppetlabs/vmpooler/issues/204)
+- `find\_folder` in vsphere\_helper.rb has subtle errors [\#204](https://github.com/puppetlabs/vmpooler/issues/204)
 - Should not use `abort` in vsphere\_helper [\#203](https://github.com/puppetlabs/vmpooler/issues/203)
 - No reason why get\_snapshot\_list is defined in vsphere\_helper [\#202](https://github.com/puppetlabs/vmpooler/issues/202)
 - Setting max\_tries in configuration results in vSphereHelper going into infinite loop [\#199](https://github.com/puppetlabs/vmpooler/issues/199)
@@ -796,7 +886,7 @@
 - \(POOLER-93\) Extend API endpoint to provide just what is needed [\#245](https://github.com/puppetlabs/vmpooler/pull/245) ([sbeaulie](https://github.com/sbeaulie))
 - \(POOLER-92\) Add the alias information in the API status page for each… [\#244](https://github.com/puppetlabs/vmpooler/pull/244) ([sbeaulie](https://github.com/sbeaulie))
 - \(QENG-5305\) Improve vmpooler host selection [\#242](https://github.com/puppetlabs/vmpooler/pull/242) ([mattkirby](https://github.com/mattkirby))
-- Allow user to specify a configuration file in VMPOOLER\_CONFIG\_FILE variable [\#241](https://github.com/puppetlabs/vmpooler/pull/241) ([adamdav](https://github.com/adamdav))
+- Allow user to specify a configuration file in VMPOOLER\_CONFIG\_FILE variable [\#241](https://github.com/puppetlabs/vmpooler/pull/241) ([amcdson](https://github.com/amcdson))
 - Fix no implicit conversion to rational from nil [\#239](https://github.com/puppetlabs/vmpooler/pull/239) ([sbeaulie](https://github.com/sbeaulie))
 - Updated Vagrant box and associated docs [\#237](https://github.com/puppetlabs/vmpooler/pull/237) ([genebean](https://github.com/genebean))
 - \(GH-226\) Respond quickly to VMs being consumed [\#236](https://github.com/puppetlabs/vmpooler/pull/236) ([glennsarti](https://github.com/glennsarti))
@@ -830,88 +920,7 @@
 - \(maint\) Add rubocop and allow failures in Travis CI [\#183](https://github.com/puppetlabs/vmpooler/pull/183) ([glennsarti](https://github.com/glennsarti))
 - \(POOLER-73\) Update unit tests prior to refactoring [\#182](https://github.com/puppetlabs/vmpooler/pull/182) ([glennsarti](https://github.com/glennsarti))
 - \(POOLER-71\) Add dummy authentication provider [\#180](https://github.com/puppetlabs/vmpooler/pull/180) ([glennsarti](https://github.com/glennsarti))
-- \(maint\) Remove Ruby 1.9.3 testing from Travis [\#178](https://github.com/puppetlabs/vmpooler/pull/178) ([glennsarti](https://github.com/glennsarti))
 - \(maint\) Enhance VM Pooler developer experience [\#177](https://github.com/puppetlabs/vmpooler/pull/177) ([glennsarti](https://github.com/glennsarti))
-- \(POOLER-47\) Send clone errors up [\#175](https://github.com/puppetlabs/vmpooler/pull/175) ([mattkirby](https://github.com/mattkirby))
-- \(POOLER-48\) Clear migrations at application start time [\#174](https://github.com/puppetlabs/vmpooler/pull/174) ([mattkirby](https://github.com/mattkirby))
-- Add retry logic with a delay for vsphere connections [\#173](https://github.com/puppetlabs/vmpooler/pull/173) ([mattkirby](https://github.com/mattkirby))
-- \(POOLER-44\) Fix vmpooler.migrate reference [\#172](https://github.com/puppetlabs/vmpooler/pull/172) ([mattkirby](https://github.com/mattkirby))
-- Add `puma` as required gem [\#171](https://github.com/puppetlabs/vmpooler/pull/171) ([sschneid](https://github.com/sschneid))
-- Fix JavaScript error on nil `weekly_data` [\#170](https://github.com/puppetlabs/vmpooler/pull/170) ([sschneid](https://github.com/sschneid))
-- Containerize vmpooler [\#169](https://github.com/puppetlabs/vmpooler/pull/169) ([sschneid](https://github.com/sschneid))
-- Add vagrant-vmpooler plugin to readme [\#168](https://github.com/puppetlabs/vmpooler/pull/168) ([briancain](https://github.com/briancain))
-- Improve vmpooler scheduling logic [\#167](https://github.com/puppetlabs/vmpooler/pull/167) ([mattkirby](https://github.com/mattkirby))
-- \[QENG-4181\] Add per-pool stats to `/status` API [\#162](https://github.com/puppetlabs/vmpooler/pull/162) ([rick](https://github.com/rick))
-- Merge CI.next into Master [\#161](https://github.com/puppetlabs/vmpooler/pull/161) ([shermdog](https://github.com/shermdog))
-- \(maint\) update README.md and LICENSE to reflect rebranding [\#157](https://github.com/puppetlabs/vmpooler/pull/157) ([erosa](https://github.com/erosa))
-- Add info about vmfloaty [\#156](https://github.com/puppetlabs/vmpooler/pull/156) ([briancain](https://github.com/briancain))
-- Added IP lookup functionality for /vm/hostname [\#154](https://github.com/puppetlabs/vmpooler/pull/154) ([frozenfoxx](https://github.com/frozenfoxx))
-- Improved tests for vmpooler [\#152](https://github.com/puppetlabs/vmpooler/pull/152) ([rick](https://github.com/rick))
-- Added prefix parameter to the vmpooler configuration [\#149](https://github.com/puppetlabs/vmpooler/pull/149) ([frozenfoxx](https://github.com/frozenfoxx))
-- Update license copyright [\#148](https://github.com/puppetlabs/vmpooler/pull/148) ([sschneid](https://github.com/sschneid))
-- Allow new disks to be added to running VMs via vmpooler API [\#147](https://github.com/puppetlabs/vmpooler/pull/147) ([sschneid](https://github.com/sschneid))
-- Updated YAML config variables in create\_template\_deltas.rb [\#145](https://github.com/puppetlabs/vmpooler/pull/145) ([frozenfoxx](https://github.com/frozenfoxx))
-- \(QA-2036\) Update README for Client Utility [\#143](https://github.com/puppetlabs/vmpooler/pull/143) ([cowofevil](https://github.com/cowofevil))
-- add guestinfo.hostname to VirtualMachineConfigSpecs [\#139](https://github.com/puppetlabs/vmpooler/pull/139) ([heathseals](https://github.com/heathseals))
-- \(QENG-2807\) Allow pool 'alias' names [\#138](https://github.com/puppetlabs/vmpooler/pull/138) ([sschneid](https://github.com/sschneid))
-- \(QENG-2995\) Display associated VMs in GET /token/:token endpoint [\#137](https://github.com/puppetlabs/vmpooler/pull/137) ([sschneid](https://github.com/sschneid))
-- Update API docs to include "domain" key for get vm requests [\#136](https://github.com/puppetlabs/vmpooler/pull/136) ([briancain](https://github.com/briancain))
-- \(MAINT\) Remove Ping Check on Running VMs [\#133](https://github.com/puppetlabs/vmpooler/pull/133) ([colinPL](https://github.com/colinPL))
-- \(maint\) Move VM Only When SSH Check Succeeds [\#131](https://github.com/puppetlabs/vmpooler/pull/131) ([colinPL](https://github.com/colinPL))
-- \(QENG-2952\) Check that SSH is available [\#130](https://github.com/puppetlabs/vmpooler/pull/130) ([sschneid](https://github.com/sschneid))
-- \(maint\) Update license copyright [\#128](https://github.com/puppetlabs/vmpooler/pull/128) ([sschneid](https://github.com/sschneid))
-- \(maint\) Remove duplicate \(nested\) "ok" responses [\#127](https://github.com/puppetlabs/vmpooler/pull/127) ([sschneid](https://github.com/sschneid))
-- \(maint\) Documentation updates [\#126](https://github.com/puppetlabs/vmpooler/pull/126) ([sschneid](https://github.com/sschneid))
-- Track token use times [\#125](https://github.com/puppetlabs/vmpooler/pull/125) ([sschneid](https://github.com/sschneid))
-- Docs update [\#124](https://github.com/puppetlabs/vmpooler/pull/124) ([sschneid](https://github.com/sschneid))
-- User token list [\#123](https://github.com/puppetlabs/vmpooler/pull/123) ([sschneid](https://github.com/sschneid))
-- \(maint\) Additional utility and reporting scripts [\#122](https://github.com/puppetlabs/vmpooler/pull/122) ([sschneid](https://github.com/sschneid))
-- \(maint\) Syntax fixup [\#121](https://github.com/puppetlabs/vmpooler/pull/121) ([sschneid](https://github.com/sschneid))
-- \(MAINT\) Reduce redis Calls in API [\#120](https://github.com/puppetlabs/vmpooler/pull/120) ([colinPL](https://github.com/colinPL))
-- \(maint\) Use expect\_json helper method for determining JSON response status [\#119](https://github.com/puppetlabs/vmpooler/pull/119) ([sschneid](https://github.com/sschneid))
-- \(QENG-1304\) vmpooler should require an auth key for VM destruction [\#118](https://github.com/puppetlabs/vmpooler/pull/118) ([sschneid](https://github.com/sschneid))
-- \(QENG-2636\) Host snapshots [\#117](https://github.com/puppetlabs/vmpooler/pull/117) ([sschneid](https://github.com/sschneid))
-- \(maint\) Use dep caching and containers [\#116](https://github.com/puppetlabs/vmpooler/pull/116) ([sschneid](https://github.com/sschneid))
-- \(maint\) Include travis-ci build status in README [\#115](https://github.com/puppetlabs/vmpooler/pull/115) ([sschneid](https://github.com/sschneid))
-- Show test contexts and names [\#114](https://github.com/puppetlabs/vmpooler/pull/114) ([sschneid](https://github.com/sschneid))
-- \(QENG-2246\) Add Default Rake Task [\#113](https://github.com/puppetlabs/vmpooler/pull/113) ([colinPL](https://github.com/colinPL))
-- Log empty pools [\#112](https://github.com/puppetlabs/vmpooler/pull/112) ([sschneid](https://github.com/sschneid))
-- \(QENG-2246\) Add Travis CI [\#111](https://github.com/puppetlabs/vmpooler/pull/111) ([colinPL](https://github.com/colinPL))
-- \(QENG-2388\) Tagging restrictions [\#110](https://github.com/puppetlabs/vmpooler/pull/110) ([sschneid](https://github.com/sschneid))
-- An updated dashboard [\#109](https://github.com/puppetlabs/vmpooler/pull/109) ([sschneid](https://github.com/sschneid))
-- API summary rework [\#108](https://github.com/puppetlabs/vmpooler/pull/108) ([sschneid](https://github.com/sschneid))
-- Only filter regex matches [\#106](https://github.com/puppetlabs/vmpooler/pull/106) ([sschneid](https://github.com/sschneid))
-- \(QENG-2518\) Tag-filtering [\#105](https://github.com/puppetlabs/vmpooler/pull/105) ([sschneid](https://github.com/sschneid))
-- \(QENG-2360\) check\_running\_vm Spec Tests [\#104](https://github.com/puppetlabs/vmpooler/pull/104) ([colinPL](https://github.com/colinPL))
-- \(QENG-2056\) Create daily tag indexes, report in /summary [\#102](https://github.com/puppetlabs/vmpooler/pull/102) ([sschneid](https://github.com/sschneid))
-- Store token metadata in vmpooler\_\_vm\_\_ Redis hash [\#101](https://github.com/puppetlabs/vmpooler/pull/101) ([sschneid](https://github.com/sschneid))
-- Display VM state in GET /vm/:hostname route [\#100](https://github.com/puppetlabs/vmpooler/pull/100) ([sschneid](https://github.com/sschneid))
-- Add basic auth token functionality [\#98](https://github.com/puppetlabs/vmpooler/pull/98) ([sschneid](https://github.com/sschneid))
-- Add basic HTTP authentication and /token routes [\#97](https://github.com/puppetlabs/vmpooler/pull/97) ([sschneid](https://github.com/sschneid))
-- \(QENG-2208\) Add more helper tests [\#95](https://github.com/puppetlabs/vmpooler/pull/95) ([colinPL](https://github.com/colinPL))
-- \(QENG-2208\) Move Sinatra Helpers to own file [\#94](https://github.com/puppetlabs/vmpooler/pull/94) ([colinPL](https://github.com/colinPL))
-- Fix rspec tests broken in f9de28236b726e37977123cea9b4f3a562bfdcdb [\#93](https://github.com/puppetlabs/vmpooler/pull/93) ([sschneid](https://github.com/sschneid))
-- Redirect / to /dashboard [\#92](https://github.com/puppetlabs/vmpooler/pull/92) ([sschneid](https://github.com/sschneid))
-- Ensure 'lifetime' val returned by GET /vm/:hostname is an int [\#91](https://github.com/puppetlabs/vmpooler/pull/91) ([sschneid](https://github.com/sschneid))
-- running-to-lifetime comparison should be 'greater than or equal to' [\#90](https://github.com/puppetlabs/vmpooler/pull/90) ([sschneid](https://github.com/sschneid))
-- Auto-expire Redis metadata key via Redis EXPIRE [\#89](https://github.com/puppetlabs/vmpooler/pull/89) ([sschneid](https://github.com/sschneid))
-- \(QENG-1906\) Add specs for Dashboard and root API class [\#88](https://github.com/puppetlabs/vmpooler/pull/88) ([colinPL](https://github.com/colinPL))
-- \(maint\) Fix bad redis reference [\#87](https://github.com/puppetlabs/vmpooler/pull/87) ([colinPL](https://github.com/colinPL))
-- \(QENG-1906\) Break apart check\_pending\_vm and add spec tests [\#86](https://github.com/puppetlabs/vmpooler/pull/86) ([colinPL](https://github.com/colinPL))
-- Remove defined? when checking configuration for graphite server. [\#85](https://github.com/puppetlabs/vmpooler/pull/85) ([colinPL](https://github.com/colinPL))
-- \(QENG-1906\) Add spec tests for Janitor [\#78](https://github.com/puppetlabs/vmpooler/pull/78) ([colinPL](https://github.com/colinPL))
-- \(QENG-1906\) Refactor initialize to allow config passing [\#77](https://github.com/puppetlabs/vmpooler/pull/77) ([colinPL](https://github.com/colinPL))
-- Use 'checkout' time to calculate 'running' time [\#75](https://github.com/puppetlabs/vmpooler/pull/75) ([sschneid](https://github.com/sschneid))
-- Catch improperly-formatted data payloads [\#73](https://github.com/puppetlabs/vmpooler/pull/73) ([sschneid](https://github.com/sschneid))
-- \(QENG-1905\) Adding VM-tagging support via PUT /vm/:hostname endpoint [\#72](https://github.com/puppetlabs/vmpooler/pull/72) ([sschneid](https://github.com/sschneid))
-- \(QENG-2057\) Historic Redis VM metadata [\#71](https://github.com/puppetlabs/vmpooler/pull/71) ([sschneid](https://github.com/sschneid))
-- \(QENG-1899\) Add documentation for /summary [\#67](https://github.com/puppetlabs/vmpooler/pull/67) ([colinPL](https://github.com/colinPL))
-- Use $redis.hgetall rather than hget in a loop [\#66](https://github.com/puppetlabs/vmpooler/pull/66) ([sschneid](https://github.com/sschneid))
-- /summary per-pool metrics [\#65](https://github.com/puppetlabs/vmpooler/pull/65) ([sschneid](https://github.com/sschneid))
-- Show boot metrics in /status and /summary endpoints [\#64](https://github.com/puppetlabs/vmpooler/pull/64) ([sschneid](https://github.com/sschneid))
-- \(maint\) Fixing spacing [\#63](https://github.com/puppetlabs/vmpooler/pull/63) ([sschneid](https://github.com/sschneid))
-- Metric calc via helpers [\#62](https://github.com/puppetlabs/vmpooler/pull/62) ([sschneid](https://github.com/sschneid))
-- More granular metrics [\#61](https://github.com/puppetlabs/vmpooler/pull/61) ([sschneid](https://github.com/sschneid))
 
 
 

Gemfile

@@ -3,11 +3,11 @@ source ENV['GEM_SOURCE'] || 'https://rubygems.org'
 gemspec
 
 # Evaluate Gemfile.local if it exists
-if File.exists? "#{__FILE__}.local"
+if File.exist? "#{__FILE__}.local"
   instance_eval(File.read("#{__FILE__}.local"))
 end
 
 # Evaluate ~/.gemfile if it exists
-if File.exists?(File.join(Dir.home, '.gemfile'))
+if File.exist?(File.join(Dir.home, '.gemfile'))
   instance_eval(File.read(File.join(Dir.home, '.gemfile')))
 end

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    vmpooler (3.5.0)
+    vmpooler (3.8.1)
       concurrent-ruby (~> 1.1)
       connection_pool (~> 2.4)
       deep_merge (~> 1.2)
@@ -9,10 +9,11 @@ PATH
       opentelemetry-exporter-jaeger (= 0.23.0)
       opentelemetry-instrumentation-concurrent_ruby (= 0.21.1)
       opentelemetry-instrumentation-http_client (= 0.22.2)
+      opentelemetry-instrumentation-rack (= 0.23.4)
       opentelemetry-instrumentation-redis (= 0.25.3)
       opentelemetry-instrumentation-sinatra (= 0.23.2)
-      opentelemetry-resource_detectors (= 0.24.1)
+      opentelemetry-resource_detectors (= 0.24.2)
-      opentelemetry-sdk (~> 1.3, >= 1.3.0)
+      opentelemetry-sdk (~> 1.8)
       pickup (~> 0.0.11)
       prometheus-client (>= 2, < 5)
       puma (>= 5.0.4, < 7)
@@ -26,36 +27,41 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    ast (2.4.2)
+    ast (2.4.3)
-    base64 (0.1.1)
+    base64 (0.1.2)
-    bindata (2.4.15)
+    bindata (2.5.1)
-    builder (3.2.4)
+    builder (3.3.0)
     climate_control (1.2.0)
     coderay (1.1.3)
-    concurrent-ruby (1.2.2)
+    concurrent-ruby (1.3.5)
-    connection_pool (2.4.1)
+    connection_pool (2.5.3)
     deep_merge (1.2.2)
-    diff-lcs (1.5.0)
+    diff-lcs (1.6.2)
-    docile (1.4.0)
+    docile (1.4.1)
-    faraday (2.7.10)
+    faraday (2.13.1)
-      faraday-net_http (>= 2.0, < 3.1)
+      faraday-net_http (>= 2.0, < 3.5)
-      ruby2_keywords (>= 0.0.4)
+      json
-    faraday-net_http (3.0.2)
+      logger
-    ffi (1.15.5-java)
+    faraday-net_http (3.4.0)
-    google-cloud-env (1.6.0)
+      net-http (>= 0.5.0)
-      faraday (>= 0.17.3, < 3.0)
+    ffi (1.17.2-java)
-    json (2.6.3)
+    google-cloud-env (2.2.1)
-    json (2.6.3-java)
+      faraday (>= 1.0, < 3.a)
-    language_server-protocol (3.17.0.3)
+    json (2.12.2)
-    method_source (1.0.0)
+    json (2.12.2-java)
+    language_server-protocol (3.17.0.5)
+    logger (1.7.0)
+    method_source (1.1.0)
     mock_redis (0.37.0)
-    mustermann (3.0.0)
+    mustermann (3.0.3)
       ruby2_keywords (~> 0.0.1)
-    net-ldap (0.18.0)
+    net-http (0.6.0)
-    nio4r (2.5.9)
+      uri
-    nio4r (2.5.9-java)
+    net-ldap (0.19.0)
-    opentelemetry-api (1.2.1)
+    nio4r (2.7.4)
-    opentelemetry-common (0.20.0)
+    nio4r (2.7.4-java)
+    opentelemetry-api (1.5.0)
+    opentelemetry-common (0.20.1)
       opentelemetry-api (~> 1.0)
     opentelemetry-exporter-jaeger (0.23.0)
       opentelemetry-api (~> 1.1)
@@ -63,7 +69,7 @@ GEM
       opentelemetry-sdk (~> 1.2)
       opentelemetry-semantic_conventions
       thrift
-    opentelemetry-instrumentation-base (0.22.2)
+    opentelemetry-instrumentation-base (0.22.3)
       opentelemetry-api (~> 1.0)
       opentelemetry-registry (~> 0.1)
     opentelemetry-instrumentation-concurrent_ruby (0.21.1)
@@ -86,64 +92,67 @@ GEM
       opentelemetry-common (~> 0.20.0)
       opentelemetry-instrumentation-base (~> 0.22.1)
       opentelemetry-instrumentation-rack (~> 0.21)
-    opentelemetry-registry (0.3.0)
+    opentelemetry-registry (0.4.0)
       opentelemetry-api (~> 1.1)
-    opentelemetry-resource_detectors (0.24.1)
+    opentelemetry-resource_detectors (0.24.2)
       google-cloud-env
       opentelemetry-sdk (~> 1.0)
-    opentelemetry-sdk (1.3.0)
+    opentelemetry-sdk (1.8.0)
       opentelemetry-api (~> 1.1)
       opentelemetry-common (~> 0.20)
       opentelemetry-registry (~> 0.2)
       opentelemetry-semantic_conventions
-    opentelemetry-semantic_conventions (1.10.0)
+    opentelemetry-semantic_conventions (1.11.0)
       opentelemetry-api (~> 1.0)
-    parallel (1.23.0)
+    parallel (1.27.0)
-    parser (3.2.2.3)
+    parser (3.3.8.0)
       ast (~> 2.4.1)
       racc
     pickup (0.0.11)
-    prometheus-client (4.2.1)
+    prism (1.4.0)
-    pry (0.14.2)
+    prometheus-client (4.2.4)
+      base64
+    pry (0.15.2)
       coderay (~> 1.1)
       method_source (~> 1.0)
-    pry (0.14.2-java)
+    pry (0.15.2-java)
       coderay (~> 1.1)
       method_source (~> 1.0)
       spoon (~> 0.0)
-    puma (6.3.1)
+    puma (6.6.0)
       nio4r (~> 2.0)
-    puma (6.3.1-java)
+    puma (6.6.0-java)
       nio4r (~> 2.0)
-    racc (1.7.1)
+    racc (1.8.1)
-    racc (1.7.1-java)
+    racc (1.8.1-java)
-    rack (2.2.8)
+    rack (2.2.17)
-    rack-protection (3.1.0)
+    rack-protection (3.2.0)
+      base64 (>= 0.1.0)
       rack (~> 2.2, >= 2.2.4)
-    rack-test (2.1.0)
+    rack-test (2.2.0)
       rack (>= 1.3)
     rainbow (3.1.1)
-    rake (13.0.6)
+    rake (13.3.0)
-    redis (5.0.7)
+    redis (5.4.0)
-      redis-client (>= 0.9.0)
+      redis-client (>= 0.22.0)
-    redis-client (0.15.0)
+    redis-client (0.24.0)
       connection_pool
-    regexp_parser (2.8.1)
+    regexp_parser (2.10.0)
-    rexml (3.2.6)
+    rexml (3.4.1)
-    rspec (3.12.0)
+    rspec (3.13.1)
-      rspec-core (~> 3.12.0)
+      rspec-core (~> 3.13.0)
-      rspec-expectations (~> 3.12.0)
+      rspec-expectations (~> 3.13.0)
-      rspec-mocks (~> 3.12.0)
+      rspec-mocks (~> 3.13.0)
-    rspec-core (3.12.2)
+    rspec-core (3.13.4)
-      rspec-support (~> 3.12.0)
+      rspec-support (~> 3.13.0)
-    rspec-expectations (3.12.3)
+    rspec-expectations (3.13.5)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.12.0)
+      rspec-support (~> 3.13.0)
-    rspec-mocks (3.12.6)
+    rspec-mocks (3.13.5)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.12.0)
+      rspec-support (~> 3.13.0)
-    rspec-support (3.12.1)
+    rspec-support (3.13.4)
-    rubocop (1.56.0)
+    rubocop (1.56.4)
       base64 (~> 0.1.1)
       json (~> 2.3)
       language_server-protocol (>= 3.17.0)
@@ -155,43 +164,48 @@ GEM
       rubocop-ast (>= 1.28.1, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 3.0)
-    rubocop-ast (1.29.0)
+    rubocop-ast (1.44.1)
-      parser (>= 3.2.1.0)
+      parser (>= 3.3.7.2)
+      prism (~> 1.4)
     ruby-progressbar (1.13.0)
     ruby2_keywords (0.0.5)
     simplecov (0.22.0)
       docile (~> 1.1)
       simplecov-html (~> 0.11)
       simplecov_json_formatter (~> 0.1)
-    simplecov-html (0.12.3)
+    simplecov-html (0.13.1)
     simplecov_json_formatter (0.1.4)
-    sinatra (3.1.0)
+    sinatra (3.2.0)
       mustermann (~> 3.0)
       rack (~> 2.2, >= 2.2.4)
-      rack-protection (= 3.1.0)
+      rack-protection (= 3.2.0)
       tilt (~> 2.0)
     spicy-proton (2.1.15)
       bindata (~> 2.3)
     spoon (0.0.6)
       ffi
     statsd-ruby (1.5.0)
-    thor (1.2.2)
+    thor (1.3.2)
-    thrift (0.18.1)
+    thrift (0.22.0)
-    tilt (2.2.0)
+    tilt (2.6.0)
-    unicode-display_width (2.4.2)
+    unicode-display_width (2.6.0)
+    uri (1.0.3)
     yarjuf (2.0.0)
       builder
       rspec (~> 3)
 
 PLATFORMS
   arm64-darwin-22
+  arm64-darwin-23
+  arm64-darwin-25
   universal-java-11
+  universal-java-17
   x86_64-darwin-22
   x86_64-linux
 
 DEPENDENCIES
   climate_control (>= 0.2.0)
-  mock_redis (>= 0.17.0)
+  mock_redis (= 0.37.0)
   pry
   rack-test (>= 0.6)
   rspec (>= 3.2)

lib/vmpooler/api/helpers.rb

@@ -1,10 +1,13 @@
 # frozen_string_literal: true
 
+require 'vmpooler/api/input_validator'
+
 module Vmpooler
 
   class API
 
     module Helpers
+      include InputValidator
 
       def tracer
         @tracer ||= OpenTelemetry.tracer_provider.tracer('api', Vmpooler::VERSION)
@@ -289,6 +292,7 @@ module Vmpooler
       def get_queue_metrics(pools, backend)
         tracer.in_span("Vmpooler::API::Helpers.#{__method__}") do
           queue = {
+              requested: 0,
               pending: 0,
               cloning: 0,
               booting: 0,
@@ -298,15 +302,35 @@ module Vmpooler
               total: 0
           }
 
-          queue[:pending]   = get_total_across_pools_redis_scard(pools, 'vmpooler__pending__', backend)
+          # Use a single pipeline to fetch all queue counts at once for better performance
-          queue[:ready]     = get_total_across_pools_redis_scard(pools, 'vmpooler__ready__', backend)
+          results = backend.pipelined do |pipeline|
-          queue[:running]   = get_total_across_pools_redis_scard(pools, 'vmpooler__running__', backend)
+            # Order matters - we'll use indices to extract values
-          queue[:completed] = get_total_across_pools_redis_scard(pools, 'vmpooler__completed__', backend)
+            pools.each do |pool|
+              pipeline.scard("vmpooler__provisioning__request#{pool['name']}")    # 0..n-1
+              pipeline.scard("vmpooler__provisioning__processing#{pool['name']}") # n..2n-1
+              pipeline.scard("vmpooler__odcreate__task#{pool['name']}")           # 2n..3n-1
+              pipeline.scard("vmpooler__pending__#{pool['name']}")                # 3n..4n-1
+              pipeline.scard("vmpooler__ready__#{pool['name']}")                  # 4n..5n-1
+              pipeline.scard("vmpooler__running__#{pool['name']}")                # 5n..6n-1
+              pipeline.scard("vmpooler__completed__#{pool['name']}")              # 6n..7n-1
+            end
+            pipeline.get('vmpooler__tasks__clone')           # 7n
+            pipeline.get('vmpooler__tasks__ondemandclone')   # 7n+1
+          end
 
-          queue[:cloning] = backend.get('vmpooler__tasks__clone').to_i + backend.get('vmpooler__tasks__ondemandclone').to_i
+          n = pools.length
-          queue[:booting] = queue[:pending].to_i - queue[:cloning].to_i
+          # Safely extract results with default to empty array if slice returns nil
-          queue[:booting] = 0 if queue[:booting] < 0
+          queue[:requested] = (results[0...n] || []).sum(&:to_i) +
-          queue[:total]   = queue[:pending].to_i + queue[:ready].to_i + queue[:running].to_i + queue[:completed].to_i
+                              (results[n...(2 * n)] || []).sum(&:to_i) +
+                              (results[(2 * n)...(3 * n)] || []).sum(&:to_i)
+          queue[:pending]   = (results[(3 * n)...(4 * n)] || []).sum(&:to_i)
+          queue[:ready]     = (results[(4 * n)...(5 * n)] || []).sum(&:to_i)
+          queue[:running]   = (results[(5 * n)...(6 * n)] || []).sum(&:to_i)
+          queue[:completed] = (results[(6 * n)...(7 * n)] || []).sum(&:to_i)
+          queue[:cloning]   = (results[7 * n] || 0).to_i + (results[7 * n + 1] || 0).to_i
+          queue[:booting]   = queue[:pending].to_i - queue[:cloning].to_i
+          queue[:booting]   = 0 if queue[:booting] < 0
+          queue[:total]     = queue[:requested] + queue[:pending].to_i + queue[:ready].to_i + queue[:running].to_i + queue[:completed].to_i
 
           queue
         end

lib/vmpooler/api/input_validator.rb

@@ -0,0 +1,159 @@
+# frozen_string_literal: true
+
+module Vmpooler
+  class API
+    # Input validation helpers to enhance security
+    module InputValidator
+      # Maximum lengths to prevent abuse
+      MAX_HOSTNAME_LENGTH = 253
+      MAX_TAG_KEY_LENGTH = 50
+      MAX_TAG_VALUE_LENGTH = 255
+      MAX_REASON_LENGTH = 500
+      MAX_POOL_NAME_LENGTH = 100
+      MAX_TOKEN_LENGTH = 64
+
+      # Valid patterns
+      HOSTNAME_PATTERN = /\A[a-z0-9]([a-z0-9-]{0,61}[a-z0-9])?(\.[a-z0-9]([a-z0-9-]{0,61}[a-z0-9])?)* \z/ix.freeze
+      POOL_NAME_PATTERN = /\A[a-zA-Z0-9_-]+\z/.freeze
+      TAG_KEY_PATTERN = /\A[a-zA-Z0-9_\-.]+\z/.freeze
+      TOKEN_PATTERN = /\A[a-zA-Z0-9\-_]+\z/.freeze
+      INTEGER_PATTERN = /\A\d+\z/.freeze
+
+      class ValidationError < StandardError; end
+
+      # Validate hostname format and length
+      def validate_hostname(hostname)
+        return error_response('Hostname is required') if hostname.nil? || hostname.empty?
+        return error_response('Hostname too long') if hostname.length > MAX_HOSTNAME_LENGTH
+        return error_response('Invalid hostname format') unless hostname.match?(HOSTNAME_PATTERN)
+
+        true
+      end
+
+      # Validate pool/template name
+      def validate_pool_name(pool_name)
+        return error_response('Pool name is required') if pool_name.nil? || pool_name.empty?
+        return error_response('Pool name too long') if pool_name.length > MAX_POOL_NAME_LENGTH
+        return error_response('Invalid pool name format') unless pool_name.match?(POOL_NAME_PATTERN)
+
+        true
+      end
+
+      # Validate tag key and value
+      def validate_tag(key, value)
+        return error_response('Tag key is required') if key.nil? || key.empty?
+        return error_response('Tag key too long') if key.length > MAX_TAG_KEY_LENGTH
+        return error_response('Invalid tag key format') unless key.match?(TAG_KEY_PATTERN)
+
+        if value
+          return error_response('Tag value too long') if value.length > MAX_TAG_VALUE_LENGTH
+
+          # Sanitize value to prevent injection attacks
+          sanitized_value = value.gsub(/[^\w\s\-.@:\/]/, '')
+          return error_response('Tag value contains invalid characters') if sanitized_value != value
+        end
+
+        true
+      end
+
+      # Validate token format
+      def validate_token_format(token)
+        return error_response('Token is required') if token.nil? || token.empty?
+        return error_response('Token too long') if token.length > MAX_TOKEN_LENGTH
+        return error_response('Invalid token format') unless token.match?(TOKEN_PATTERN)
+
+        true
+      end
+
+      # Validate integer parameter
+      def validate_integer(value, name = 'value', min: nil, max: nil)
+        return error_response("#{name} is required") if value.nil?
+
+        value_str = value.to_s
+        return error_response("#{name} must be a valid integer") unless value_str.match?(INTEGER_PATTERN)
+
+        int_value = value.to_i
+        return error_response("#{name} must be at least #{min}") if min && int_value < min
+        return error_response("#{name} must be at most #{max}") if max && int_value > max
+
+        int_value
+      end
+
+      # Validate VM request count
+      def validate_vm_count(count)
+        validated = validate_integer(count, 'VM count', min: 1, max: 100)
+        return validated if validated.is_a?(Hash) # error response
+
+        validated
+      end
+
+      # Validate disk size
+      def validate_disk_size(size)
+        validated = validate_integer(size, 'Disk size', min: 1, max: 2048)
+        return validated if validated.is_a?(Hash) # error response
+
+        validated
+      end
+
+      # Validate lifetime (TTL) in hours
+      def validate_lifetime(lifetime)
+        validated = validate_integer(lifetime, 'Lifetime', min: 1, max: 168) # max 1 week
+        return validated if validated.is_a?(Hash) # error response
+
+        validated
+      end
+
+      # Validate reason text
+      def validate_reason(reason)
+        return true if reason.nil? || reason.empty?
+        return error_response('Reason too long') if reason.length > MAX_REASON_LENGTH
+
+        # Sanitize to prevent XSS/injection
+        sanitized = reason.gsub(/[<>"']/, '')
+        return error_response('Reason contains invalid characters') if sanitized != reason
+
+        true
+      end
+
+      # Sanitize JSON body to prevent injection
+      def sanitize_json_body(body)
+        return {} if body.nil? || body.empty?
+
+        begin
+          parsed = JSON.parse(body)
+          return error_response('Request body must be a JSON object') unless parsed.is_a?(Hash)
+
+          # Limit depth and size to prevent DoS
+          return error_response('Request body too complex') if json_depth(parsed) > 5
+          return error_response('Request body too large') if body.length > 10_240 # 10KB max
+
+          parsed
+        rescue JSON::ParserError => e
+          error_response("Invalid JSON: #{e.message}")
+        end
+      end
+
+      # Check if validation result is an error
+      def validation_error?(result)
+        result.is_a?(Hash) && result['ok'] == false
+      end
+
+      private
+
+      def error_response(message)
+        { 'ok' => false, 'error' => message }
+      end
+
+      def json_depth(obj, depth = 0)
+        return depth unless obj.is_a?(Hash) || obj.is_a?(Array)
+        return depth + 1 if obj.empty?
+
+        if obj.is_a?(Hash)
+          depth + 1 + obj.values.map { |v| json_depth(v, 0) }.max
+        else
+          depth + 1 + obj.map { |v| json_depth(v, 0) }.max
+        end
+      end
+    end
+  end
+end

lib/vmpooler/api/rate_limiter.rb

@@ -0,0 +1,116 @@
+# frozen_string_literal: true
+
+module Vmpooler
+  class API
+    # Rate limiter middleware to protect against abuse
+    # Uses Redis to track request counts per IP and token
+    class RateLimiter
+      DEFAULT_LIMITS = {
+        global_per_ip: { limit: 100, period: 60 }, # 100 requests per minute per IP
+        authenticated: { limit: 500, period: 60 }, # 500 requests per minute with token
+        vm_creation: { limit: 20, period: 60 },    # 20 VM creations per minute
+        vm_deletion: { limit: 50, period: 60 }     # 50 VM deletions per minute
+      }.freeze
+
+      def initialize(app, redis, config = {})
+        @app = app
+        @redis = redis
+        @config = DEFAULT_LIMITS.merge(config[:rate_limits] || {})
+        @enabled = config.fetch(:rate_limiting_enabled, true)
+      end
+
+      def call(env)
+        return @app.call(env) unless @enabled
+
+        request = Rack::Request.new(env)
+        client_id = identify_client(request)
+        endpoint_type = classify_endpoint(request)
+
+        # Check rate limits
+        return rate_limit_response(client_id, endpoint_type) if rate_limit_exceeded?(client_id, endpoint_type, request)
+
+        # Track the request
+        increment_request_count(client_id, endpoint_type)
+
+        @app.call(env)
+      end
+
+      private
+
+      def identify_client(request)
+        # Prioritize token-based identification for authenticated requests
+        token = request.env['HTTP_X_AUTH_TOKEN']
+        return "token:#{token}" if token && !token.empty?
+
+        # Fall back to IP address
+        ip = request.ip || request.env['REMOTE_ADDR'] || 'unknown'
+        "ip:#{ip}"
+      end
+
+      def classify_endpoint(request)
+        path = request.path
+        method = request.request_method
+
+        return :vm_creation if method == 'POST' && path.include?('/vm')
+        return :vm_deletion if method == 'DELETE' && path.include?('/vm')
+        return :authenticated if request.env['HTTP_X_AUTH_TOKEN']
+
+        :global_per_ip
+      end
+
+      def rate_limit_exceeded?(client_id, endpoint_type, _request)
+        limit_config = @config[endpoint_type] || @config[:global_per_ip]
+        key = "vmpooler__ratelimit__#{endpoint_type}__#{client_id}"
+
+        current_count = @redis.get(key).to_i
+        current_count >= limit_config[:limit]
+      rescue StandardError => e
+        # If Redis fails, allow the request through (fail open)
+        warn "Rate limiter Redis error: #{e.message}"
+        false
+      end
+
+      def increment_request_count(client_id, endpoint_type)
+        limit_config = @config[endpoint_type] || @config[:global_per_ip]
+        key = "vmpooler__ratelimit__#{endpoint_type}__#{client_id}"
+
+        @redis.pipelined do |pipeline|
+          pipeline.incr(key)
+          pipeline.expire(key, limit_config[:period])
+        end
+      rescue StandardError => e
+        # Log error but don't fail the request
+        warn "Rate limiter increment error: #{e.message}"
+      end
+
+      def rate_limit_response(client_id, endpoint_type)
+        limit_config = @config[endpoint_type] || @config[:global_per_ip]
+        key = "vmpooler__ratelimit__#{endpoint_type}__#{client_id}"
+
+        begin
+          ttl = @redis.ttl(key)
+        rescue StandardError
+          ttl = limit_config[:period]
+        end
+
+        headers = {
+          'Content-Type' => 'application/json',
+          'X-RateLimit-Limit' => limit_config[:limit].to_s,
+          'X-RateLimit-Remaining' => '0',
+          'X-RateLimit-Reset' => (Time.now.to_i + ttl).to_s,
+          'Retry-After' => ttl.to_s
+        }
+
+        body = JSON.pretty_generate({
+                                      'ok' => false,
+                                      'error' => 'Rate limit exceeded',
+                                      'limit' => limit_config[:limit],
+                                      'period' => limit_config[:period],
+                                      'retry_after' => ttl
+                                    })
+
+        [429, headers, [body]]
+      end
+    end
+  end
+end

lib/vmpooler/api/v3.rb

@@ -9,6 +9,20 @@ module Vmpooler
       api_version = '3'
       api_prefix  = "/api/v#{api_version}"
 
+      # Simple in-memory cache for status endpoint
+      # rubocop:disable Style/ClassVars
+      @@status_cache = {}
+      @@status_cache_mutex = Mutex.new
+      # rubocop:enable Style/ClassVars
+      STATUS_CACHE_TTL = 30 # seconds
+
+      # Clear cache (useful for testing)
+      def self.clear_status_cache
+        @@status_cache_mutex.synchronize do
+          @@status_cache.clear
+        end
+      end
+
       helpers do
         include Vmpooler::API::Helpers
       end
@@ -283,11 +297,9 @@ module Vmpooler
       def update_user_metrics(operation, vmname)
         tracer.in_span("Vmpooler::API::V3.#{__method__}") do |span|
           begin
-            backend.multi
+            jenkins_build_url = backend.hget("vmpooler__vm__#{vmname}", 'tag:jenkins_build_url')
-            backend.hget("vmpooler__vm__#{vmname}", 'tag:jenkins_build_url')
+            user = backend.hget("vmpooler__vm__#{vmname}", 'token:user')
-            backend.hget("vmpooler__vm__#{vmname}", 'token:user')
+            poolname = backend.hget("vmpooler__vm__#{vmname}", 'template')
-            backend.hget("vmpooler__vm__#{vmname}", 'template')
-            jenkins_build_url, user, poolname = backend.exec
             poolname = poolname.gsub('.', '_')
 
             if user
@@ -466,6 +478,32 @@ module Vmpooler
         end
       end
 
+      # Cache helper methods for status endpoint
+      def get_cached_status(cache_key)
+        @@status_cache_mutex.synchronize do
+          cached = @@status_cache[cache_key]
+          if cached && (Time.now - cached[:timestamp]) < STATUS_CACHE_TTL
+            return cached[:data]
+          end
+
+          nil
+        end
+      end
+
+      def set_cached_status(cache_key, data)
+        @@status_cache_mutex.synchronize do
+          @@status_cache[cache_key] = {
+            data: data,
+            timestamp: Time.now
+          }
+          # Cleanup old cache entries (keep only last 10 unique view combinations)
+          if @@status_cache.size > 10
+            oldest = @@status_cache.min_by { |_k, v| v[:timestamp] }
+            @@status_cache.delete(oldest[0])
+          end
+        end
+      end
+
       def sync_pool_templates
         tracer.in_span("Vmpooler::API::V3.#{__method__}") do
           pool_index = pool_index(pools)
@@ -648,6 +686,13 @@ module Vmpooler
       get "#{api_prefix}/status/?" do
         content_type :json
 
+        # Create cache key based on view parameters
+        cache_key = params[:view] ? "status_#{params[:view]}" : "status_all"
+
+        # Try to get cached response
+        cached_response = get_cached_status(cache_key)
+        return cached_response if cached_response
+
         if params[:view]
           views = params[:view].split(",")
         end
@@ -708,7 +753,12 @@ module Vmpooler
 
         result[:status][:uptime] = (Time.now - Vmpooler::API.settings.config[:uptime]).round(1) if Vmpooler::API.settings.config[:uptime]
 
-        JSON.pretty_generate(Hash[result.sort_by { |k, _v| k }])
+        response = JSON.pretty_generate(Hash[result.sort_by { |k, _v| k }])
+
+        # Cache the response
+        set_cached_status(cache_key, response)
+
+        response
       end
 
       # request statistics for specific pools by passing parameter 'pool'
@@ -1087,9 +1137,29 @@ module Vmpooler
         result = { 'ok' => false }
         metrics.increment('http_requests_vm_total.post.vm.checkout')
 
-        payload = JSON.parse(request.body.read)
+        # Validate and sanitize JSON body
+        payload = sanitize_json_body(request.body.read)
+        if validation_error?(payload)
+          status 400
+          return JSON.pretty_generate(payload)
+        end
 
-        if payload
+        # Validate each template and count
+        payload.each do |template, count|
+          validation = validate_pool_name(template)
+          if validation_error?(validation)
+            status 400
+            return JSON.pretty_generate(validation)
+          end
+
+          validated_count = validate_vm_count(count)
+          if validation_error?(validated_count)
+            status 400
+            return JSON.pretty_generate(validated_count)
+          end
+        end
+
+        if payload && !payload.empty?
           invalid = invalid_templates(payload)
           if invalid.empty?
             result = atomically_allocate_vms(payload)
@@ -1208,6 +1278,7 @@ module Vmpooler
         result = { 'ok' => false }
         metrics.increment('http_requests_vm_total.get.vm.template')
 
+        # Template can contain multiple pools separated by +, so validate after parsing
         payload = extract_templates_from_query_params(params[:template])
 
         if payload
@@ -1237,6 +1308,13 @@ module Vmpooler
         status 404
         result['ok'] = false
 
+        # Validate hostname
+        validation = validate_hostname(params[:hostname])
+        if validation_error?(validation)
+          status 400
+          return JSON.pretty_generate(validation)
+        end
+
         params[:hostname] = hostname_shorten(params[:hostname])
 
         rdata = backend.hgetall("vmpooler__vm__#{params[:hostname]}")
@@ -1375,6 +1453,13 @@ module Vmpooler
         status 404
         result['ok'] = false
 
+        # Validate hostname
+        validation = validate_hostname(params[:hostname])
+        if validation_error?(validation)
+          status 400
+          return JSON.pretty_generate(validation)
+        end
+
         params[:hostname] = hostname_shorten(params[:hostname])
 
         rdata = backend.hgetall("vmpooler__vm__#{params[:hostname]}")
@@ -1405,16 +1490,21 @@ module Vmpooler
 
         failure = []
 
+        # Validate hostname
+        validation = validate_hostname(params[:hostname])
+        if validation_error?(validation)
+          status 400
+          return JSON.pretty_generate(validation)
+        end
+
         params[:hostname] = hostname_shorten(params[:hostname])
 
         if backend.exists?("vmpooler__vm__#{params[:hostname]}")
-          begin
+          # Validate and sanitize JSON body
-            jdata = JSON.parse(request.body.read)
+          jdata = sanitize_json_body(request.body.read)
-          rescue StandardError => e
+          if validation_error?(jdata)
-            span = OpenTelemetry::Trace.current_span
+            status 400
-            span.record_exception(e)
+            return JSON.pretty_generate(jdata)
-            span.status = OpenTelemetry::Trace::Status.error(e.to_s)
-            halt 400, JSON.pretty_generate(result)
           end
 
           # Validate data payload
@@ -1423,6 +1513,13 @@ module Vmpooler
               when 'lifetime'
                 need_token! if Vmpooler::API.settings.config[:auth]
 
+                # Validate lifetime is a positive integer
+                lifetime_int = arg.to_i
+                if lifetime_int <= 0
+                  failure.push("Lifetime must be a positive integer (got #{arg})")
+                  next
+                end
+
                 # in hours, defaults to one week
                 max_lifetime_upper_limit = config['max_lifetime_upper_limit']
                 if max_lifetime_upper_limit
@@ -1432,13 +1529,17 @@ module Vmpooler
                   end
                 end
 
-                # validate lifetime is within boundaries
-                unless arg.to_i > 0
-                  failure.push("You provided a lifetime (#{arg}) but you must provide a positive number.")
-                end
-
               when 'tags'
                 failure.push("You provided tags (#{arg}) as something other than a hash.") unless arg.is_a?(Hash)
+
+                # Validate each tag key and value
+                arg.each do |key, value|
+                  tag_validation = validate_tag(key, value)
+                  if validation_error?(tag_validation)
+                    failure.push(tag_validation['error'])
+                  end
+                end
+
                 failure.push("You provided unsuppored tags (#{arg}).") if config['allowed_tags'] && !(arg.keys - config['allowed_tags']).empty?
               else
                 failure.push("Unknown argument #{arg}.")
@@ -1480,9 +1581,23 @@ module Vmpooler
         status 404
         result = { 'ok' => false }
 
+        # Validate hostname
+        validation = validate_hostname(params[:hostname])
+        if validation_error?(validation)
+          status 400
+          return JSON.pretty_generate(validation)
+        end
+
+        # Validate disk size
+        validated_size = validate_disk_size(params[:size])
+        if validation_error?(validated_size)
+          status 400
+          return JSON.pretty_generate(validated_size)
+        end
+
         params[:hostname] = hostname_shorten(params[:hostname])
 
-        if ((params[:size].to_i > 0 )and (backend.exists?("vmpooler__vm__#{params[:hostname]}")))
+        if backend.exists?("vmpooler__vm__#{params[:hostname]}")
           result[params[:hostname]] = {}
           result[params[:hostname]]['disk'] = "+#{params[:size]}gb"
 

lib/vmpooler/metrics/promstats.rb

@@ -329,6 +329,30 @@ module Vmpooler
             buckets: REDIS_CONNECT_BUCKETS,
             docstring: 'vmpooler redis connection wait time',
             param_labels: %i[type provider]
+          },
+          vmpooler_health: {
+            mtype: M_GAUGE,
+            torun: %i[manager],
+            docstring: 'vmpooler health check metrics',
+            param_labels: %i[metric_path]
+          },
+          vmpooler_purge: {
+            mtype: M_GAUGE,
+            torun: %i[manager],
+            docstring: 'vmpooler purge metrics',
+            param_labels: %i[metric_path]
+          },
+          vmpooler_destroy: {
+            mtype: M_GAUGE,
+            torun: %i[manager],
+            docstring: 'vmpooler destroy metrics',
+            param_labels: %i[poolname]
+          },
+          vmpooler_clone: {
+            mtype: M_GAUGE,
+            torun: %i[manager],
+            docstring: 'vmpooler clone metrics',
+            param_labels: %i[poolname]
           }
         }
       end

lib/vmpooler/pool_manager.rb

@@ -129,6 +129,8 @@ module Vmpooler
       already_timed_out = time_since_clone > timeout
       timing_out_soon = time_since_clone > timeout_notification && !redis.hget("vmpooler__vm__#{vm}", 'timeout_notification')
 
+      return true if !already_timed_out && !timing_out_soon
+
       if already_timed_out
         unless exists
           remove_nonexistent_vm(vm, pool, redis)
@@ -142,7 +144,9 @@ module Vmpooler
       nonexist_warning = if already_timed_out
                            "[!] [#{pool}] '#{vm}' marked as 'failed' after #{timeout} minutes with error: #{open_socket_error}"
                          elsif timing_out_soon
-                           "[!] [#{pool}] '#{vm}' no longer exists when attempting to send notification of impending failure"
+                           time_remaining = timeout - timeout_notification
+                           open_socket_error = redis.hget("vmpooler__vm__#{vm}", 'open_socket_error')
+                           "[!] [#{pool}] '#{vm}' impending failure in #{time_remaining} minutes with error: #{open_socket_error}"
                          else
                            "[!] [#{pool}] '#{vm}' This error is wholly unexpected"
                          end
@@ -157,16 +161,80 @@ module Vmpooler
       request_id = redis.hget("vmpooler__vm__#{vm}", 'request_id')
       pool_alias = redis.hget("vmpooler__vm__#{vm}", 'pool_alias') if request_id
       open_socket_error = redis.hget("vmpooler__vm__#{vm}", 'open_socket_error')
+      retry_count = redis.hget("vmpooler__odrequest__#{request_id}", 'retry_count').to_i if request_id
+
+      # Move to DLQ before moving to completed queue
+      move_to_dlq(vm, pool, 'pending', 'Timeout',
+                  open_socket_error || 'VM timed out during pending phase',
+                  redis, request_id: request_id, pool_alias: pool_alias, retry_count: retry_count)
+
+      clone_error = redis.hget("vmpooler__vm__#{vm}", 'clone_error')
+      clone_error_class = redis.hget("vmpooler__vm__#{vm}", 'clone_error_class')
       redis.smove("vmpooler__pending__#{pool}", "vmpooler__completed__#{pool}", vm)
+
       if request_id
         ondemandrequest_hash = redis.hgetall("vmpooler__odrequest__#{request_id}")
         if ondemandrequest_hash && ondemandrequest_hash['status'] != 'failed' && ondemandrequest_hash['status'] != 'deleted'
-          # will retry a VM that did not come up as vm_ready? only if it has not been market failed or deleted
+          # Check retry count and max retry limit before retrying
-          redis.zadd('vmpooler__odcreate__task', 1, "#{pool_alias}:#{pool}:1:#{request_id}")
+          retry_count = (redis.hget("vmpooler__odrequest__#{request_id}", 'retry_count') || '0').to_i
+          max_retries = $config[:config]['max_vm_retries'] || 3
+
+          $logger.log('s', "[!] [#{pool}] '#{vm}' checking retry logic: error='#{clone_error}', error_class='#{clone_error_class}', retry_count=#{retry_count}, max_retries=#{max_retries}")
+
+          # Determine if error is likely permanent (configuration issues)
+          permanent_error = permanent_error?(clone_error, clone_error_class)
+          $logger.log('s', "[!] [#{pool}] '#{vm}' permanent_error check result: #{permanent_error}")
+
+          if retry_count < max_retries && !permanent_error
+            # Increment retry count and retry VM creation
+            redis.hset("vmpooler__odrequest__#{request_id}", 'retry_count', retry_count + 1)
+            redis.zadd('vmpooler__odcreate__task', 1, "#{pool_alias}:#{pool}:1:#{request_id}")
+            $logger.log('s', "[!] [#{pool}] '#{vm}' failed, retrying (attempt #{retry_count + 1}/#{max_retries})")
+          else
+            # Max retries exceeded or permanent error, mark request as permanently failed
+            failure_reason = if permanent_error
+                               "Configuration error: #{clone_error}"
+                             else
+                               'Max retry attempts exceeded'
+                             end
+            redis.hset("vmpooler__odrequest__#{request_id}", 'status', 'failed')
+            redis.hset("vmpooler__odrequest__#{request_id}", 'failure_reason', failure_reason)
+            $logger.log('s', "[!] [#{pool}] '#{vm}' permanently failed: #{failure_reason}")
+            $metrics.increment("vmpooler_errors.permanently_failed.#{pool}")
+          end
         end
       end
-      $metrics.increment("errors.markedasfailed.#{pool}")
+      $metrics.increment("vmpooler_errors.markedasfailed.#{pool}")
-      open_socket_error
+      open_socket_error || clone_error
+    end
+
+    # Determine if an error is likely permanent (configuration issue) vs transient
+    def permanent_error?(error_message, error_class)
+      return false if error_message.nil? || error_class.nil?
+
+      permanent_error_patterns = [
+        /template.*not found/i,
+        /template.*does not exist/i,
+        /invalid.*path/i,
+        /folder.*not found/i,
+        /datastore.*not found/i,
+        /resource pool.*not found/i,
+        /permission.*denied/i,
+        /authentication.*failed/i,
+        /invalid.*credentials/i,
+        /configuration.*error/i
+      ]
+
+      permanent_error_classes = [
+        'ArgumentError',
+        'NoMethodError',
+        'NameError'
+      ]
+
+      # Check error message patterns
+      permanent_error_patterns.any? { |pattern| error_message.match?(pattern) } ||
+        # Check error class types
+        permanent_error_classes.include?(error_class)
     end
 
     def move_pending_vm_to_ready(vm, pool, redis, request_id = nil)
@@ -219,8 +287,16 @@ module Vmpooler
       return true if provider.vm_ready?(pool_name, vm_name, redis)
 
       raise("VM #{vm_name} is not ready")
-    rescue StandardError
+    rescue StandardError => e
       open_socket_error = redis.hget("vmpooler__vm__#{vm_name}", 'open_socket_error')
+      request_id = redis.hget("vmpooler__vm__#{vm_name}", 'request_id')
+      pool_alias = redis.hget("vmpooler__vm__#{vm_name}", 'pool_alias')
+
+      # Move to DLQ before moving to completed queue
+      move_to_dlq(vm_name, pool_name, 'ready', e.class.name,
+                  open_socket_error || 'VM became unreachable in ready queue',
+                  redis, request_id: request_id, pool_alias: pool_alias)
+
       move_vm_queue(pool_name, vm_name, 'ready', 'completed', redis, "removed from 'ready' queue. vm unreachable with error: #{open_socket_error}")
     end
 
@@ -353,6 +429,60 @@ module Vmpooler
       $logger.log('d', "[!] [#{pool}] '#{vm}' #{msg}") if msg
     end
 
+    # Dead-Letter Queue (DLQ) helper methods
+    def dlq_enabled?
+      $config[:config] && $config[:config]['dlq_enabled'] == true
+    end
+
+    def dlq_ttl
+      ($config[:config] && $config[:config]['dlq_ttl']) || 168 # default 7 days in hours
+    end
+
+    def dlq_max_entries
+      ($config[:config] && $config[:config]['dlq_max_entries']) || 10_000
+    end
+
+    def move_to_dlq(vm, pool, queue_type, error_class, error_message, redis, request_id: nil, pool_alias: nil, retry_count: 0, skip_metrics: false)
+      return unless dlq_enabled?
+
+      dlq_key = "vmpooler__dlq__#{queue_type}"
+      timestamp = Time.now.to_i
+
+      # Build DLQ entry
+      dlq_entry = {
+        'vm' => vm,
+        'pool' => pool,
+        'queue_from' => queue_type,
+        'error_class' => error_class.to_s,
+        'error_message' => error_message.to_s,
+        'failed_at' => Time.now.iso8601,
+        'retry_count' => retry_count,
+        'request_id' => request_id,
+        'pool_alias' => pool_alias
+      }.compact
+
+      # Use sorted set with timestamp as score for easy age-based queries and TTL
+      dlq_entry_json = dlq_entry.to_json
+      redis.zadd(dlq_key, timestamp, "#{vm}:#{timestamp}:#{dlq_entry_json}")
+
+      # Enforce max entries limit by removing oldest entries
+      current_size = redis.zcard(dlq_key)
+      if current_size > dlq_max_entries
+        remove_count = current_size - dlq_max_entries
+        redis.zremrangebyrank(dlq_key, 0, remove_count - 1)
+        $logger.log('d', "[!] [dlq] Trimmed #{remove_count} oldest entries from #{dlq_key}")
+      end
+
+      # Set expiration on the entire DLQ (will be refreshed on next write)
+      ttl_seconds = dlq_ttl * 3600
+      redis.expire(dlq_key, ttl_seconds)
+
+      $metrics.increment("vmpooler_dlq.#{queue_type}.count") unless skip_metrics
+      $logger.log('d', "[!] [dlq] Moved '#{vm}' from '#{queue_type}' queue to DLQ: #{error_message}")
+    rescue StandardError => e
+      $logger.log('s', "[!] [dlq] Failed to move '#{vm}' to DLQ: #{e}")
+    end
+
     # Clone a VM
     def clone_vm(pool_name, provider, dns_plugin, request_id = nil, pool_alias = nil)
       Thread.new do
@@ -362,7 +492,13 @@ module Vmpooler
           if request_id
             $logger.log('s', "[!] [#{pool_name}] failed while cloning VM for request #{request_id} with an error: #{e}")
             @redis.with_metrics do |redis|
-              redis.zadd('vmpooler__odcreate__task', 1, "#{pool_alias}:#{pool_name}:1:#{request_id}")
+              # Only re-queue if the request wasn't already marked as failed (e.g., by permanent error detection)
+              request_status = redis.hget("vmpooler__odrequest__#{request_id}", 'status')
+              if request_status != 'failed'
+                redis.zadd('vmpooler__odcreate__task', 1, "#{pool_alias}:#{pool_name}:1:#{request_id}")
+              else
+                $logger.log('s', "[!] [#{pool_name}] Request #{request_id} already marked as failed, not re-queueing")
+              end
             end
           else
             $logger.log('s', "[!] [#{pool_name}] failed while cloning VM with an error: #{e}")
@@ -415,10 +551,10 @@ module Vmpooler
         hostname_retries += 1
 
         if !hostname_available
-          $metrics.increment("errors.duplicatehostname.#{pool_name}")
+          $metrics.increment("vmpooler_errors.duplicatehostname.#{pool_name}")
           $logger.log('s', "[!] [#{pool_name}] Generated hostname #{fqdn} was not unique (attempt \##{hostname_retries} of #{max_hostname_retries})")
         elsif !dns_available
-          $metrics.increment("errors.staledns.#{pool_name}")
+          $metrics.increment("vmpooler_errors.staledns.#{pool_name}")
           $logger.log('s', "[!] [#{pool_name}] Generated hostname #{fqdn} already exists in DNS records (#{dns_ip}), stale DNS")
         end
       end
@@ -464,7 +600,7 @@ module Vmpooler
           provider.create_vm(pool_name, new_vmname)
           finish = format('%<time>.2f', time: Time.now - start)
           $logger.log('s', "[+] [#{pool_name}] '#{new_vmname}' cloned in #{finish} seconds")
-          $metrics.timing("clone.#{pool_name}", finish)
+          $metrics.gauge("vmpooler_clone.#{pool_name}", finish)
 
           $logger.log('d', "[ ] [#{pool_name}] Obtaining IP for '#{new_vmname}'")
           ip_start = Time.now
@@ -485,14 +621,50 @@ module Vmpooler
 
           dns_plugin_class_name = get_dns_plugin_class_name_for_pool(pool_name)
           dns_plugin.create_or_replace_record(new_vmname) unless dns_plugin_class_name == 'dynamic-dns'
-        rescue StandardError
+        rescue StandardError => e
+          # Store error details for retry decision making
           @redis.with_metrics do |redis|
+            # Get retry count before moving to DLQ
+            retry_count = 0
+            if request_id
+              ondemandrequest_hash = redis.hgetall("vmpooler__odrequest__#{request_id}")
+              retry_count = ondemandrequest_hash['retry_count'].to_i if ondemandrequest_hash
+            end
+
+            # Move to DLQ before removing from pending queue
+            move_to_dlq(new_vmname, pool_name, 'clone', e.class.name, e.message,
+                        redis, request_id: request_id, pool_alias: pool_alias, retry_count: retry_count)
+
             redis.pipelined do |pipeline|
               pipeline.srem("vmpooler__pending__#{pool_name}", new_vmname)
+              pipeline.hset("vmpooler__vm__#{new_vmname}", 'clone_error', e.message)
+              pipeline.hset("vmpooler__vm__#{new_vmname}", 'clone_error_class', e.class.name)
               expiration_ttl = $config[:redis]['data_ttl'].to_i * 60 * 60
               pipeline.expire("vmpooler__vm__#{new_vmname}", expiration_ttl)
             end
+
+            # Handle retry logic for on-demand requests
+            if request_id
+              retry_count = (redis.hget("vmpooler__odrequest__#{request_id}", 'retry_count') || '0').to_i
+              max_retries = $config[:config]['max_vm_retries'] || 3
+              is_permanent = permanent_error?(e.message, e.class.name)
+
+              $logger.log('s', "[!] [#{pool_name}] '#{new_vmname}' checking immediate failure retry: error='#{e.message}', error_class='#{e.class.name}', retry_count=#{retry_count}, max_retries=#{max_retries}, permanent_error=#{is_permanent}")
+
+              if is_permanent || retry_count >= max_retries
+                reason = is_permanent ? 'permanent error detected' : 'max retries exceeded'
+                $logger.log('s', "[!] [#{pool_name}] Cancelling request #{request_id} due to #{reason}")
+                redis.hset("vmpooler__odrequest__#{request_id}", 'status', 'failed')
+                redis.zadd('vmpooler__odcreate__task', 0, "#{pool_alias}:#{pool_name}:0:#{request_id}")
+              else
+                # Increment retry count and re-queue for retry
+                redis.hincrby("vmpooler__odrequest__#{request_id}", 'retry_count', 1)
+                $logger.log('s', "[+] [#{pool_name}] Request #{request_id} will be retried (attempt #{retry_count + 1}/#{max_retries})")
+                redis.zadd('vmpooler__odcreate__task', 1, "#{pool_alias}:#{pool_name}:1:#{request_id}")
+              end
+            end
           end
+          $logger.log('s', "[!] [#{pool_name}] '#{new_vmname}' clone failed: #{e.class}: #{e.message}")
           raise
         ensure
           @redis.with_metrics do |redis|
@@ -542,7 +714,7 @@ module Vmpooler
 
           finish = format('%<time>.2f', time: Time.now - start)
           $logger.log('s', "[-] [#{pool}] '#{vm}' destroyed in #{finish} seconds")
-          $metrics.timing("destroy.#{pool}", finish)
+          $metrics.gauge("vmpooler_destroy.#{pool}", finish)
         end
       end
       dereference_mutex(vm)
@@ -578,6 +750,543 @@ module Vmpooler
       provider.purge_unconfigured_resources(allowlist)
     end
 
+    # Auto-purge stale queue entries
+    def purge_enabled?
+      $config[:config] && $config[:config]['purge_enabled'] == true
+    end
+
+    def purge_dry_run?
+      $config[:config] && $config[:config]['purge_dry_run'] == true
+    end
+
+    def max_pending_age
+      ($config[:config] && $config[:config]['max_pending_age']) || 7200 # default 2 hours in seconds
+    end
+
+    def max_ready_age
+      ($config[:config] && $config[:config]['max_ready_age']) || 86_400 # default 24 hours in seconds
+    end
+
+    def max_completed_age
+      ($config[:config] && $config[:config]['max_completed_age']) || 3600 # default 1 hour in seconds
+    end
+
+    def max_orphaned_age
+      ($config[:config] && $config[:config]['max_orphaned_age']) || 86_400 # default 24 hours in seconds
+    end
+
+    def purge_stale_queue_entries
+      return unless purge_enabled?
+
+      Thread.new do
+        begin
+          $logger.log('d', '[*] [purge] Starting stale queue entry purge cycle')
+          purge_start = Time.now
+
+          @redis.with_metrics do |redis|
+            total_purged = 0
+
+            # Purge stale entries from each pool
+            $config[:pools].each do |pool|
+              pool_name = pool['name']
+
+              # Purge pending queue
+              purged_pending = purge_pending_queue(pool_name, redis)
+              total_purged += purged_pending
+
+              # Purge ready queue
+              purged_ready = purge_ready_queue(pool_name, redis)
+              total_purged += purged_ready
+
+              # Purge completed queue
+              purged_completed = purge_completed_queue(pool_name, redis)
+              total_purged += purged_completed
+            end
+
+            # Purge orphaned VM metadata
+            purged_orphaned = purge_orphaned_metadata(redis)
+            total_purged += purged_orphaned
+
+            purge_duration = Time.now - purge_start
+            $logger.log('s', "[*] [purge] Completed purge cycle in #{purge_duration.round(2)}s: #{total_purged} entries purged")
+            $metrics.gauge('vmpooler_purge.cycle.duration', purge_duration)
+            $metrics.gauge('vmpooler_purge.total.count', total_purged)
+          end
+        rescue StandardError => e
+          $logger.log('s', "[!] [purge] Failed during purge cycle: #{e}")
+        end
+      end
+    end
+
+    def purge_pending_queue(pool_name, redis)
+      queue_key = "vmpooler__pending__#{pool_name}"
+      vms = redis.smembers(queue_key)
+      purged_count = 0
+
+      vms.each do |vm|
+        begin
+          clone_time_str = redis.hget("vmpooler__vm__#{vm}", 'clone')
+          next unless clone_time_str
+
+          clone_time = Time.parse(clone_time_str)
+          age = Time.now - clone_time
+
+          if age > max_pending_age
+            request_id = redis.hget("vmpooler__vm__#{vm}", 'request_id')
+            pool_alias = redis.hget("vmpooler__vm__#{vm}", 'pool_alias')
+
+            purged_count += 1
+
+            if purge_dry_run?
+              $logger.log('d', "[*] [purge][dry-run] Would purge stale pending VM '#{vm}' (age: #{age.round(0)}s, max: #{max_pending_age}s)")
+            else
+              # Move to DLQ before removing (skip DLQ metric since we're tracking purge metric)
+              move_to_dlq(vm, pool_name, 'pending', 'Purge',
+                          "Stale pending VM (age: #{age.round(0)}s > max: #{max_pending_age}s)",
+                          redis, request_id: request_id, pool_alias: pool_alias, skip_metrics: true)
+
+              redis.srem(queue_key, vm)
+
+              # Set expiration on VM metadata if data_ttl is configured
+              if $config[:redis] && $config[:redis]['data_ttl']
+                expiration_ttl = $config[:redis]['data_ttl'].to_i * 60 * 60
+                redis.expire("vmpooler__vm__#{vm}", expiration_ttl)
+              end
+
+              $logger.log('d', "[!] [purge] Purged stale pending VM '#{vm}' from '#{pool_name}' (age: #{age.round(0)}s)")
+              $metrics.increment("vmpooler_purge.pending.#{pool_name}.count")
+            end
+          end
+        rescue StandardError => e
+          $logger.log('d', "[!] [purge] Error checking pending VM '#{vm}': #{e}")
+        end
+      end
+
+      purged_count
+    end
+
+    def purge_ready_queue(pool_name, redis)
+      queue_key = "vmpooler__ready__#{pool_name}"
+      vms = redis.smembers(queue_key)
+      purged_count = 0
+
+      vms.each do |vm|
+        begin
+          ready_time_str = redis.hget("vmpooler__vm__#{vm}", 'ready')
+          next unless ready_time_str
+
+          ready_time = Time.parse(ready_time_str)
+          age = Time.now - ready_time
+
+          if age > max_ready_age
+            if purge_dry_run?
+              $logger.log('d', "[*] [purge][dry-run] Would purge stale ready VM '#{vm}' (age: #{age.round(0)}s, max: #{max_ready_age}s)")
+            else
+              redis.smove(queue_key, "vmpooler__completed__#{pool_name}", vm)
+              $logger.log('d', "[!] [purge] Moved stale ready VM '#{vm}' from '#{pool_name}' to completed (age: #{age.round(0)}s)")
+              $metrics.increment("vmpooler_purge.ready.#{pool_name}.count")
+            end
+            purged_count += 1
+          end
+        rescue StandardError => e
+          $logger.log('d', "[!] [purge] Error checking ready VM '#{vm}': #{e}")
+        end
+      end
+
+      purged_count
+    end
+
+    def purge_completed_queue(pool_name, redis)
+      queue_key = "vmpooler__completed__#{pool_name}"
+      vms = redis.smembers(queue_key)
+      purged_count = 0
+
+      vms.each do |vm|
+        begin
+          # Check destroy time or last activity time
+          destroy_time_str = redis.hget("vmpooler__vm__#{vm}", 'destroy')
+          checkout_time_str = redis.hget("vmpooler__vm__#{vm}", 'checkout')
+
+          # Use the most recent timestamp
+          timestamp_str = destroy_time_str || checkout_time_str
+          next unless timestamp_str
+
+          timestamp = Time.parse(timestamp_str)
+          age = Time.now - timestamp
+
+          if age > max_completed_age
+            if purge_dry_run?
+              $logger.log('d', "[*] [purge][dry-run] Would purge stale completed VM '#{vm}' (age: #{age.round(0)}s, max: #{max_completed_age}s)")
+            else
+              redis.srem(queue_key, vm)
+              $logger.log('d', "[!] [purge] Removed stale completed VM '#{vm}' from '#{pool_name}' (age: #{age.round(0)}s)")
+              $metrics.increment("vmpooler_purge.completed.#{pool_name}.count")
+            end
+            purged_count += 1
+          end
+        rescue StandardError => e
+          $logger.log('d', "[!] [purge] Error checking completed VM '#{vm}': #{e}")
+        end
+      end
+
+      purged_count
+    end
+
+    def purge_orphaned_metadata(redis)
+      # Find VM metadata that doesn't belong to any queue
+      all_vm_keys = redis.keys('vmpooler__vm__*')
+      purged_count = 0
+
+      all_vm_keys.each do |vm_key|
+        begin
+          vm = vm_key.sub('vmpooler__vm__', '')
+
+          # Check if VM exists in any queue
+          pool_name = redis.hget(vm_key, 'pool')
+          next unless pool_name
+
+          in_pending = redis.sismember("vmpooler__pending__#{pool_name}", vm)
+          in_ready = redis.sismember("vmpooler__ready__#{pool_name}", vm)
+          in_running = redis.sismember("vmpooler__running__#{pool_name}", vm)
+          in_completed = redis.sismember("vmpooler__completed__#{pool_name}", vm)
+          in_discovered = redis.sismember("vmpooler__discovered__#{pool_name}", vm)
+          in_migrating = redis.sismember("vmpooler__migrating__#{pool_name}", vm)
+
+          # VM is orphaned if not in any queue
+          unless in_pending || in_ready || in_running || in_completed || in_discovered || in_migrating
+            # Check age
+            clone_time_str = redis.hget(vm_key, 'clone')
+            next unless clone_time_str
+
+            clone_time = Time.parse(clone_time_str)
+            age = Time.now - clone_time
+
+            if age > max_orphaned_age
+              if purge_dry_run?
+                $logger.log('d', "[*] [purge][dry-run] Would purge orphaned metadata for '#{vm}' (age: #{age.round(0)}s, max: #{max_orphaned_age}s)")
+              else
+                expiration_ttl = 3600 # 1 hour
+                redis.expire(vm_key, expiration_ttl)
+                $logger.log('d', "[!] [purge] Set expiration on orphaned metadata for '#{vm}' (age: #{age.round(0)}s)")
+                $metrics.increment('vmpooler_purge.orphaned.count')
+              end
+              purged_count += 1
+            end
+          end
+        rescue StandardError => e
+          $logger.log('d', "[!] [purge] Error checking orphaned metadata '#{vm_key}': #{e}")
+        end
+      end
+
+      purged_count
+    end
+
+    # Health checks for Redis queues
+    def health_check_enabled?
+      $config[:config] && $config[:config]['health_check_enabled'] == true
+    end
+
+    def health_thresholds
+      defaults = {
+        'pending_queue_max' => 100,
+        'ready_queue_max' => 500,
+        'dlq_max_warning' => 100,
+        'dlq_max_critical' => 1000,
+        'stuck_vm_age_threshold' => 7200, # 2 hours
+        'stuck_vm_max_warning' => 10,
+        'stuck_vm_max_critical' => 50
+      }
+
+      if $config[:config] && $config[:config]['health_thresholds']
+        defaults.merge($config[:config]['health_thresholds'])
+      else
+        defaults
+      end
+    end
+
+    def check_queue_health
+      return unless health_check_enabled?
+
+      Thread.new do
+        begin
+          $logger.log('d', '[*] [health] Running queue health check')
+          health_start = Time.now
+
+          @redis.with_metrics do |redis|
+            health_metrics = calculate_health_metrics(redis)
+            health_status = determine_health_status(health_metrics)
+
+            # Store health metrics in Redis for API consumption
+            # Convert nested hash to JSON for storage
+            require 'json'
+            redis.hset('vmpooler__health', 'metrics', health_metrics.to_json)
+            redis.hset('vmpooler__health', 'status', health_status)
+            redis.hset('vmpooler__health', 'last_check', Time.now.iso8601)
+            redis.expire('vmpooler__health', 3600) # Expire after 1 hour
+
+            # Log health summary
+            log_health_summary(health_metrics, health_status)
+
+            # Push metrics
+            push_health_metrics(health_metrics, health_status)
+
+            health_duration = Time.now - health_start
+            $metrics.gauge('vmpooler_health.check.duration', health_duration)
+          end
+        rescue StandardError => e
+          $logger.log('s', "[!] [health] Failed during health check: #{e}")
+        end
+      end
+    end
+
+    def calculate_health_metrics(redis)
+      metrics = {
+        'queues' => {},
+        'tasks' => {},
+        'errors' => {}
+      }
+
+      total_stuck_vms = 0
+      total_dlq_size = 0
+      thresholds = health_thresholds
+
+      # Check each pool's queues
+      $config[:pools].each do |pool|
+        pool_name = pool['name']
+        metrics['queues'][pool_name] = {}
+
+        # Pending queue metrics
+        pending_key = "vmpooler__pending__#{pool_name}"
+        pending_vms = redis.smembers(pending_key)
+        pending_ages = calculate_queue_ages(pending_vms, 'clone', redis)
+        stuck_pending = pending_ages.count { |age| age > thresholds['stuck_vm_age_threshold'] }
+        total_stuck_vms += stuck_pending
+
+        metrics['queues'][pool_name]['pending'] = {
+          'size' => pending_vms.size,
+          'oldest_age' => pending_ages.max || 0,
+          'avg_age' => pending_ages.empty? ? 0 : (pending_ages.sum / pending_ages.size).round(0),
+          'stuck_count' => stuck_pending
+        }
+
+        # Ready queue metrics
+        ready_key = "vmpooler__ready__#{pool_name}"
+        ready_vms = redis.smembers(ready_key)
+        ready_ages = calculate_queue_ages(ready_vms, 'ready', redis)
+
+        metrics['queues'][pool_name]['ready'] = {
+          'size' => ready_vms.size,
+          'oldest_age' => ready_ages.max || 0,
+          'avg_age' => ready_ages.empty? ? 0 : (ready_ages.sum / ready_ages.size).round(0)
+        }
+
+        # Completed queue metrics
+        completed_key = "vmpooler__completed__#{pool_name}"
+        completed_size = redis.scard(completed_key)
+        metrics['queues'][pool_name]['completed'] = { 'size' => completed_size }
+      end
+
+      # Task queue metrics
+      clone_active = redis.get('vmpooler__tasks__clone').to_i
+      ondemand_active = redis.get('vmpooler__tasks__ondemandclone').to_i
+      odcreate_pending = redis.zcard('vmpooler__odcreate__task')
+
+      metrics['tasks']['clone'] = { 'active' => clone_active }
+      metrics['tasks']['ondemand'] = { 'active' => ondemand_active, 'pending' => odcreate_pending }
+
+      # DLQ metrics
+      if dlq_enabled?
+        dlq_keys = redis.keys('vmpooler__dlq__*')
+        dlq_keys.each do |dlq_key|
+          queue_type = dlq_key.sub('vmpooler__dlq__', '')
+          dlq_size = redis.zcard(dlq_key)
+          total_dlq_size += dlq_size
+          metrics['queues']['dlq'] ||= {}
+          metrics['queues']['dlq'][queue_type] = { 'size' => dlq_size }
+        end
+      end
+
+      # Error metrics
+      metrics['errors']['dlq_total_size'] = total_dlq_size
+      metrics['errors']['stuck_vm_count'] = total_stuck_vms
+
+      # Orphaned metadata count
+      orphaned_count = count_orphaned_metadata(redis)
+      metrics['errors']['orphaned_metadata_count'] = orphaned_count
+
+      metrics
+    end
+
+    def calculate_queue_ages(vms, timestamp_field, redis)
+      ages = []
+      vms.each do |vm|
+        begin
+          timestamp_str = redis.hget("vmpooler__vm__#{vm}", timestamp_field)
+          next unless timestamp_str
+
+          timestamp = Time.parse(timestamp_str)
+          age = (Time.now - timestamp).to_i
+          ages << age
+        rescue StandardError
+          # Skip VMs with invalid timestamps
+        end
+      end
+      ages
+    end
+
+    def count_orphaned_metadata(redis)
+      all_vm_keys = redis.keys('vmpooler__vm__*')
+      orphaned_count = 0
+
+      all_vm_keys.each do |vm_key|
+        begin
+          vm = vm_key.sub('vmpooler__vm__', '')
+          pool_name = redis.hget(vm_key, 'pool')
+          next unless pool_name
+
+          in_any_queue = redis.sismember("vmpooler__pending__#{pool_name}", vm) ||
+                         redis.sismember("vmpooler__ready__#{pool_name}", vm) ||
+                         redis.sismember("vmpooler__running__#{pool_name}", vm) ||
+                         redis.sismember("vmpooler__completed__#{pool_name}", vm) ||
+                         redis.sismember("vmpooler__discovered__#{pool_name}", vm) ||
+                         redis.sismember("vmpooler__migrating__#{pool_name}", vm)
+
+          orphaned_count += 1 unless in_any_queue
+        rescue StandardError
+          # Skip on error
+        end
+      end
+
+      orphaned_count
+    end
+
+    def determine_health_status(metrics)
+      thresholds = health_thresholds
+
+      # Check DLQ size
+      dlq_size = metrics['errors']['dlq_total_size']
+      return 'unhealthy' if dlq_size > thresholds['dlq_max_critical']
+
+      # Check stuck VM count
+      stuck_count = metrics['errors']['stuck_vm_count']
+      return 'unhealthy' if stuck_count > thresholds['stuck_vm_max_critical']
+
+      # Check queue sizes
+      metrics['queues'].each do |pool_name, queues|
+        next if pool_name == 'dlq'
+
+        pending_size = begin
+          queues['pending']['size']
+        rescue StandardError
+          0
+        end
+        ready_size = begin
+          queues['ready']['size']
+        rescue StandardError
+          0
+        end
+
+        return 'unhealthy' if pending_size > thresholds['pending_queue_max'] * 2
+        return 'unhealthy' if ready_size > thresholds['ready_queue_max'] * 2
+      end
+
+      # Check for degraded conditions
+      return 'degraded' if dlq_size > thresholds['dlq_max_warning']
+      return 'degraded' if stuck_count > thresholds['stuck_vm_max_warning']
+
+      metrics['queues'].each do |pool_name, queues|
+        next if pool_name == 'dlq'
+
+        pending_size = begin
+          queues['pending']['size']
+        rescue StandardError
+          0
+        end
+        ready_size = begin
+          queues['ready']['size']
+        rescue StandardError
+          0
+        end
+
+        return 'degraded' if pending_size > thresholds['pending_queue_max']
+        return 'degraded' if ready_size > thresholds['ready_queue_max']
+      end
+
+      'healthy'
+    end
+
+    def log_health_summary(metrics, status)
+      summary = "[*] [health] Status: #{status.upcase}"
+
+      # Queue summary
+      total_pending = 0
+      total_ready = 0
+      total_completed = 0
+
+      metrics['queues'].each do |pool_name, queues|
+        next if pool_name == 'dlq'
+
+        total_pending += begin
+          queues['pending']['size']
+        rescue StandardError
+          0
+        end
+        total_ready += begin
+          queues['ready']['size']
+        rescue StandardError
+          0
+        end
+        total_completed += begin
+          queues['completed']['size']
+        rescue StandardError
+          0
+        end
+      end
+
+      summary += " | Queues: P=#{total_pending} R=#{total_ready} C=#{total_completed}"
+      summary += " | DLQ=#{metrics['errors']['dlq_total_size']}"
+      summary += " | Stuck=#{metrics['errors']['stuck_vm_count']}"
+      summary += " | Orphaned=#{metrics['errors']['orphaned_metadata_count']}"
+
+      log_level = status == 'healthy' ? 's' : 'd'
+      $logger.log(log_level, summary)
+    end
+
+    def push_health_metrics(metrics, status)
+      # Push error metrics first
+      $metrics.gauge('vmpooler_health.dlq.total_size', metrics['errors']['dlq_total_size'])
+      $metrics.gauge('vmpooler_health.stuck_vms.count', metrics['errors']['stuck_vm_count'])
+      $metrics.gauge('vmpooler_health.orphaned_metadata.count', metrics['errors']['orphaned_metadata_count'])
+
+      # Push per-pool queue metrics
+      metrics['queues'].each do |pool_name, queues|
+        next if pool_name == 'dlq'
+
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.pending.size", queues['pending']['size'])
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.pending.oldest_age", queues['pending']['oldest_age'])
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.pending.stuck_count", queues['pending']['stuck_count'])
+
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.ready.size", queues['ready']['size'])
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.ready.oldest_age", queues['ready']['oldest_age'])
+
+        $metrics.gauge("vmpooler_health.queue.#{pool_name}.completed.size", queues['completed']['size'])
+      end
+
+      # Push DLQ metrics
+      metrics['queues']['dlq']&.each do |queue_type, dlq_metrics|
+        $metrics.gauge("vmpooler_health.dlq.#{queue_type}.size", dlq_metrics['size'])
+      end
+
+      # Push task metrics
+      $metrics.gauge('vmpooler_health.tasks.clone.active', metrics['tasks']['clone']['active'])
+      $metrics.gauge('vmpooler_health.tasks.ondemand.active', metrics['tasks']['ondemand']['active'])
+      $metrics.gauge('vmpooler_health.tasks.ondemand.pending', metrics['tasks']['ondemand']['pending'])
+
+      # Push status last (0=healthy, 1=degraded, 2=unhealthy)
+      status_value = { 'healthy' => 0, 'degraded' => 1, 'unhealthy' => 2 }[status] || 2
+      $metrics.gauge('vmpooler_health.status', status_value)
+    end
+
     def create_vm_disk(pool_name, vm, disk_size, provider)
       Thread.new do
         begin
@@ -978,7 +1687,12 @@ module Vmpooler
 
           sync_pool_template(pool)
           loop do
+            start_time = Time.now
             result = _check_pool(pool, provider)
+            duration = Time.now - start_time
+
+            $metrics.gauge("vmpooler_performance.check_pool.#{pool['name']}", duration)
+            $logger.log('d', "[!] check_pool for #{pool['name']} took #{duration.round(2)}s") if duration > 5
 
             if result[:cloned_vms] > 0 || result[:checked_pending_vms] > 0 || result[:discovered_vms] > 0
               loop_delay = loop_delay_min
@@ -1537,6 +2251,15 @@ module Vmpooler
         redis.zrem('vmpooler__provisioning__request', request_id)
         return
       end
+
+      # Check if request was already marked as failed (e.g., by delete endpoint)
+      request_status = redis.hget("vmpooler__odrequest__#{request_id}", 'status')
+      if request_status == 'failed'
+        $logger.log('s', "Request '#{request_id}' already marked as failed, skipping VM creation")
+        redis.zrem('vmpooler__provisioning__request', request_id)
+        return
+      end
+
       score = redis.zscore('vmpooler__provisioning__request', request_id)
       requested = requested.split(',')
 
@@ -1760,6 +2483,48 @@ module Vmpooler
           check_ondemand_requests(check_loop_delay_min, check_loop_delay_max, check_loop_delay_decay)
         end
 
+        # Queue purge thread
+        if purge_enabled?
+          purge_interval = ($config[:config] && $config[:config]['purge_interval']) || 3600 # default 1 hour
+          if !$threads['queue_purge']
+            $threads['queue_purge'] = Thread.new do
+              loop do
+                purge_stale_queue_entries
+                sleep(purge_interval)
+              end
+            end
+          elsif !$threads['queue_purge'].alive?
+            $logger.log('d', '[!] [queue_purge] worker thread died, restarting')
+            $threads['queue_purge'] = Thread.new do
+              loop do
+                purge_stale_queue_entries
+                sleep(purge_interval)
+              end
+            end
+          end
+        end
+
+        # Health check thread
+        if health_check_enabled?
+          health_interval = ($config[:config] && $config[:config]['health_check_interval']) || 300 # default 5 minutes
+          if !$threads['health_check']
+            $threads['health_check'] = Thread.new do
+              loop do
+                check_queue_health
+                sleep(health_interval)
+              end
+            end
+          elsif !$threads['health_check'].alive?
+            $logger.log('d', '[!] [health_check] worker thread died, restarting')
+            $threads['health_check'] = Thread.new do
+              loop do
+                check_queue_health
+                sleep(health_interval)
+              end
+            end
+          end
+        end
+
         sleep(loop_delay)
 
         unless maxloop == 0

lib/vmpooler/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Vmpooler
-  VERSION = '3.5.0'
+  VERSION = '3.8.1'
 end

release-notes.md

@@ -0,0 +1,15 @@
+## [3.8.1](https://github.com/puppetlabs/vmpooler/tree/3.8.1) (2026-01-14)
+
+[Full Changelog](https://github.com/puppetlabs/vmpooler/compare/3.7.0...3.8.1)
+
+**Implemented enhancements:**
+
+- \(P4DEVOPS-9434\) Add rate limiting and input validation security enhancements [\#690](https://github.com/puppetlabs/vmpooler/pull/690) ([mahima-singh](https://github.com/mahima-singh))
+- \(P4DEVOPS-8570\) Add Phase 2 optimizations: status API caching and improved Redis pipelining [\#689](https://github.com/puppetlabs/vmpooler/pull/689) ([mahima-singh](https://github.com/mahima-singh))
+- \(P4DEVOPS-8567\) Add DLQ, auto-purge, and health checks for Redis queues [\#688](https://github.com/puppetlabs/vmpooler/pull/688) ([mahima-singh](https://github.com/mahima-singh))
+- Add retry logic for immediate clone failures [\#687](https://github.com/puppetlabs/vmpooler/pull/687) ([mahima-singh](https://github.com/mahima-singh))
+
+**Fixed bugs:**
+
+- \(P4DEVOPS-8567\) Prevent VM allocation for already-deleted request-ids [\#688](https://github.com/puppetlabs/vmpooler/pull/688) ([mahima-singh](https://github.com/mahima-singh))
+- Prevent re-queueing requests already marked as failed [\#687](https://github.com/puppetlabs/vmpooler/pull/687) ([mahima-singh](https://github.com/mahima-singh))

release-prep

@@ -3,13 +3,14 @@
 # The container tag should closely match what is used in `docker/Dockerfile` in vmpooler-deployment
 #
 # Update Gemfile.lock
-docker run -it --rm \
+docker run -t --rm \
   -v $(pwd):/app \
-  jruby:9.4.3.0-jdk11 \
+  jruby:9.4.12.1-jdk11 \
   /bin/bash -c 'apt-get update -qq && apt-get install -y --no-install-recommends git make netbase && cd /app && gem install bundler && bundle install --jobs 3; echo "LOCK_FILE_UPDATE_EXIT_CODE=$?"'
 
 # Update Changelog
-docker run -it --rm -e CHANGELOG_GITHUB_TOKEN -v $(pwd):/usr/local/src/your-app \
+docker run -t --rm -e CHANGELOG_GITHUB_TOKEN -v $(pwd):/usr/local/src/your-app \
-  githubchangeloggenerator/github-changelog-generator:1.16.2 \
+  githubchangeloggenerator/github-changelog-generator:1.16.4 \
-  github_changelog_generator --future-release $(grep VERSION lib/vmpooler/version.rb |rev |cut -d "'" -f2 |rev)
+  github_changelog_generator --future-release $(grep VERSION lib/vmpooler/version.rb |rev |cut -d "'" -f2 |rev) \
+  --token $CHANGELOG_GITHUB_TOKEN --release-branch main
 

spec/integration/api/v3/status_spec.rb

@@ -17,6 +17,8 @@ describe Vmpooler::API::V3 do
   # https://rubydoc.info/gems/sinatra/Sinatra/Base#reset!-class_method 
   before(:each) do
     app.reset!
+    # Clear status cache to prevent test interference
+    Vmpooler::API::V3.clear_status_cache
   end
 
   describe 'status and metrics endpoints' do

spec/unit/api/helpers_spec.rb

@@ -116,7 +116,7 @@ describe Vmpooler::API::Helpers do
       allow(redis).to receive(:pipelined).with(no_args).and_return [0]
       allow(redis).to receive(:get).and_return 0
 
-      expect(subject.get_queue_metrics([], redis)).to eq({pending: 0, cloning: 0, booting: 0, ready: 0, running: 0, completed: 0, total: 0})
+      expect(subject.get_queue_metrics([], redis)).to eq({requested: 0, pending: 0, cloning: 0, booting: 0, ready: 0, running: 0, completed: 0, total: 0})
     end
 
     it 'adds pool queues correctly' do
@@ -125,10 +125,14 @@ describe Vmpooler::API::Helpers do
           {'name' => 'p2'}
       ]
 
-      allow(redis).to receive(:pipelined).with(no_args).and_return [1,1]
+      # Mock returns 7*2 + 2 = 16 results (7 queue types for 2 pools + 2 global counters)
-      allow(redis).to receive(:get).and_return(1,0)
+      # For each pool: [request, processing, odcreate, pending, ready, running, completed]
+      # Plus 2 global counters: clone (1), ondemandclone (0)
+      # Results array: [1,1, 1,1, 1,1, 1,1, 1,1, 1,1, 1,1, 1, 0]
+      #                [req,  proc,  odc,   pend, rdy,  run,  comp, clone, odc]
+      allow(redis).to receive(:pipelined).with(no_args).and_return [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0]
 
-      expect(subject.get_queue_metrics(pools, redis)).to eq({pending: 2, cloning: 1, booting: 1, ready: 2, running: 2, completed: 2, total: 8})
+      expect(subject.get_queue_metrics(pools, redis)).to eq({requested: 6, pending: 2, cloning: 1, booting: 1, ready: 2, running: 2, completed: 2, total: 14})
     end
 
     it 'sets booting to 0 when negative calculation' do
@@ -137,10 +141,10 @@ describe Vmpooler::API::Helpers do
           {'name' => 'p2'}
       ]
 
-      allow(redis).to receive(:pipelined).with(no_args).and_return [1,1]
+      # Mock returns 7*2 + 2 = 16 results with clone=5 to cause negative booting
-      allow(redis).to receive(:get).and_return(5,0)
+      allow(redis).to receive(:pipelined).with(no_args).and_return [1,1,1,1,1,1,1,1,1,1,1,1,1,1,5,0]
 
-      expect(subject.get_queue_metrics(pools, redis)).to eq({pending: 2, cloning: 5, booting: 0, ready: 2, running: 2, completed: 2, total: 8})
+      expect(subject.get_queue_metrics(pools, redis)).to eq({requested: 6, pending: 2, cloning: 5, booting: 0, ready: 2, running: 2, completed: 2, total: 14})
     end
   end
 

spec/unit/api/input_validator_spec.rb

@@ -0,0 +1,184 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+require 'rack/test'
+require 'vmpooler/api/input_validator'
+
+describe Vmpooler::API::InputValidator do
+  let(:test_class) do
+    Class.new do
+      include Vmpooler::API::InputValidator
+    end
+  end
+  let(:validator) { test_class.new }
+
+  describe '#validate_hostname' do
+    it 'accepts valid hostnames' do
+      expect(validator.validate_hostname('test-host.example.com')).to be true
+      expect(validator.validate_hostname('host123')).to be true
+    end
+
+    it 'rejects invalid hostnames' do
+      result = validator.validate_hostname('invalid_host!')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('Invalid hostname format')
+    end
+
+    it 'rejects hostnames that are too long' do
+      long_hostname = 'a' * 300
+      result = validator.validate_hostname(long_hostname)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('too long')
+    end
+
+    it 'rejects empty hostnames' do
+      result = validator.validate_hostname('')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('required')
+    end
+  end
+
+  describe '#validate_pool_name' do
+    it 'accepts valid pool names' do
+      expect(validator.validate_pool_name('centos-7-x86_64')).to be true
+      expect(validator.validate_pool_name('ubuntu-2204')).to be true
+    end
+
+    it 'rejects invalid pool names' do
+      result = validator.validate_pool_name('invalid pool!')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('Invalid pool name format')
+    end
+
+    it 'rejects pool names that are too long' do
+      result = validator.validate_pool_name('a' * 150)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('too long')
+    end
+  end
+
+  describe '#validate_tag' do
+    it 'accepts valid tags' do
+      expect(validator.validate_tag('project', 'test-123')).to be true
+      expect(validator.validate_tag('owner', 'user@example.com')).to be true
+    end
+
+    it 'rejects tags with invalid keys' do
+      result = validator.validate_tag('invalid key!', 'value')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('Invalid tag key format')
+    end
+
+    it 'rejects tags with invalid characters in value' do
+      result = validator.validate_tag('key', 'value<script>')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('invalid characters')
+    end
+
+    it 'rejects tags that are too long' do
+      result = validator.validate_tag('key', 'a' * 300)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('too long')
+    end
+  end
+
+  describe '#validate_vm_count' do
+    it 'accepts valid VM counts' do
+      expect(validator.validate_vm_count(5)).to eq(5)
+      expect(validator.validate_vm_count('10')).to eq(10)
+    end
+
+    it 'rejects counts less than 1' do
+      result = validator.validate_vm_count(0)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('at least 1')
+    end
+
+    it 'rejects counts greater than 100' do
+      result = validator.validate_vm_count(150)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('at most 100')
+    end
+
+    it 'rejects non-integer values' do
+      result = validator.validate_vm_count('abc')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('valid integer')
+    end
+  end
+
+  describe '#validate_disk_size' do
+    it 'accepts valid disk sizes' do
+      expect(validator.validate_disk_size(50)).to eq(50)
+      expect(validator.validate_disk_size('100')).to eq(100)
+    end
+
+    it 'rejects sizes less than 1' do
+      result = validator.validate_disk_size(0)
+      expect(result['ok']).to be false
+    end
+
+    it 'rejects sizes greater than 2048' do
+      result = validator.validate_disk_size(3000)
+      expect(result['ok']).to be false
+    end
+  end
+
+  describe '#validate_lifetime' do
+    it 'accepts valid lifetimes' do
+      expect(validator.validate_lifetime(24)).to eq(24)
+      expect(validator.validate_lifetime('48')).to eq(48)
+    end
+
+    it 'rejects lifetimes greater than 168 hours (1 week)' do
+      result = validator.validate_lifetime(200)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('at most 168')
+    end
+  end
+
+  describe '#sanitize_json_body' do
+    it 'parses valid JSON' do
+      result = validator.sanitize_json_body('{"key": "value"}')
+      expect(result).to eq('key' => 'value')
+    end
+
+    it 'rejects invalid JSON' do
+      result = validator.sanitize_json_body('{invalid}')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('Invalid JSON')
+    end
+
+    it 'rejects non-object JSON' do
+      result = validator.sanitize_json_body('["array"]')
+      expect(result['ok']).to be false
+      expect(result['error']).to include('must be a JSON object')
+    end
+
+    it 'rejects deeply nested JSON' do
+      deep_json = '{"a":{"b":{"c":{"d":{"e":{"f":"too deep"}}}}}}'
+      result = validator.sanitize_json_body(deep_json)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('too complex')
+    end
+
+    it 'rejects bodies that are too large' do
+      large_json = '{"data":"' + ('a' * 20000) + '"}'
+      result = validator.sanitize_json_body(large_json)
+      expect(result['ok']).to be false
+      expect(result['error']).to include('too large')
+    end
+  end
+
+  describe '#validation_error?' do
+    it 'returns true for error responses' do
+      error = { 'ok' => false, 'error' => 'test error' }
+      expect(validator.validation_error?(error)).to be true
+    end
+
+    it 'returns false for successful responses' do
+      expect(validator.validation_error?(true)).to be false
+      expect(validator.validation_error?(5)).to be false
+    end
+  end
+end

spec/unit/pool_manager_spec.rb

@@ -275,6 +275,7 @@ EOT
 
     it 'takes no action if VM is not cloning' do
       redis_connection_pool.with do |redis|
+        expect(logger).to_not receive(:log)
         expect(subject.fail_pending_vm(vm, pool, timeout, timeout_notification, redis)).to eq(true)
       end
     end
@@ -282,6 +283,7 @@ EOT
     it 'takes no action if VM is within timeout' do
       redis_connection_pool.with do |redis|
         redis.hset("vmpooler__vm__#{vm}", 'clone',Time.now.to_s)
+        expect(logger).to_not receive(:log)
         expect(subject.fail_pending_vm(vm, pool, timeout, timeout_notification, redis)).to eq(true)
         expect(redis.sismember("vmpooler__pending__#{pool}", vm)).to be(true)
       end
@@ -307,6 +309,7 @@ EOT
     it 'calls remove_nonexistent_vm if VM has exceeded timeout and does not exist' do
       redis_connection_pool.with do |redis|
         redis.hset("vmpooler__vm__#{vm}", 'clone',Date.new(2001,1,1).to_s)
+        expect(logger).to_not receive(:log)
         expect(subject).to receive(:remove_nonexistent_vm).with(vm, pool, redis)
         expect(subject.fail_pending_vm(vm, pool, timeout, timeout_notification, redis, exists: false)).to eq(true)
       end
@@ -342,6 +345,123 @@ EOT
     end
   end
 
+  describe '#handle_timed_out_vm' do
+    before do
+      expect(subject).not_to be_nil
+    end
+
+    before(:each) do
+      redis_connection_pool.with do |redis|
+        create_pending_vm(pool, vm, redis)
+        config[:config]['max_vm_retries'] = 3
+      end
+    end
+
+    context 'without request_id' do
+      it 'moves VM to completed queue and returns error' do
+        redis_connection_pool.with do |redis|
+          redis.hset("vmpooler__vm__#{vm}", 'open_socket_error', 'connection failed')
+          result = subject.handle_timed_out_vm(vm, pool, redis)
+          
+          expect(redis.sismember("vmpooler__pending__#{pool}", vm)).to be(false)
+          expect(redis.sismember("vmpooler__completed__#{pool}", vm)).to be(true)
+          expect(result).to eq('connection failed')
+        end
+      end
+    end
+
+    context 'with request_id and transient error' do
+      before(:each) do
+        redis_connection_pool.with do |redis|
+          redis.hset("vmpooler__vm__#{vm}", 'request_id', request_id)
+          redis.hset("vmpooler__vm__#{vm}", 'pool_alias', pool)
+          redis.hset("vmpooler__odrequest__#{request_id}", 'status', 'pending')
+          redis.hset("vmpooler__vm__#{vm}", 'clone_error', 'network timeout')
+          redis.hset("vmpooler__vm__#{vm}", 'clone_error_class', 'Timeout::Error')
+        end
+      end
+
+      it 'retries on first failure' do
+        redis_connection_pool.with do |redis|
+          subject.handle_timed_out_vm(vm, pool, redis)
+          
+          expect(redis.hget("vmpooler__odrequest__#{request_id}", 'retry_count')).to eq('1')
+          expect(redis.zrange('vmpooler__odcreate__task', 0, -1)).to include("#{pool}:#{pool}:1:#{request_id}")
+        end
+      end
+
+      it 'marks as failed after max retries' do
+        redis_connection_pool.with do |redis|
+          redis.hset("vmpooler__odrequest__#{request_id}", 'retry_count', '3')
+          
+          subject.handle_timed_out_vm(vm, pool, redis)
+          
+          expect(redis.hget("vmpooler__odrequest__#{request_id}", 'status')).to eq('failed')
+          expect(redis.hget("vmpooler__odrequest__#{request_id}", 'failure_reason')).to eq('Max retry attempts exceeded')
+          expect(redis.zrange('vmpooler__odcreate__task', 0, -1)).not_to include("#{pool}:#{pool}:1:#{request_id}")
+        end
+      end
+    end
+
+    context 'with request_id and permanent error' do
+      before(:each) do
+        redis_connection_pool.with do |redis|
+          redis.hset("vmpooler__vm__#{vm}", 'request_id', request_id)
+          redis.hset("vmpooler__vm__#{vm}", 'pool_alias', pool)
+          redis.hset("vmpooler__odrequest__#{request_id}", 'status', 'pending')
+          redis.hset("vmpooler__vm__#{vm}", 'clone_error', 'template not found')
+          redis.hset("vmpooler__vm__#{vm}", 'clone_error_class', 'RuntimeError')
+        end
+      end
+
+      it 'immediately marks as failed without retrying' do
+        redis_connection_pool.with do |redis|
+          subject.handle_timed_out_vm(vm, pool, redis)
+          
+          expect(redis.hget("vmpooler__odrequest__#{request_id}", 'status')).to eq('failed')
+          expect(redis.hget("vmpooler__odrequest__#{request_id}", 'failure_reason')).to include('Configuration error')
+          expect(redis.zrange('vmpooler__odcreate__task', 0, -1)).not_to include("#{pool}:#{pool}:1:#{request_id}")
+        end
+      end
+    end
+  end
+
+  describe '#permanent_error?' do
+    before do
+      expect(subject).not_to be_nil
+    end
+
+    it 'identifies template not found errors as permanent' do
+      expect(subject.permanent_error?('template not found', 'RuntimeError')).to be(true)
+    end
+
+    it 'identifies invalid path errors as permanent' do
+      expect(subject.permanent_error?('invalid path specified', 'ArgumentError')).to be(true)
+    end
+
+    it 'identifies permission denied errors as permanent' do
+      expect(subject.permanent_error?('permission denied', 'SecurityError')).to be(true)
+    end
+
+    it 'identifies ArgumentError class as permanent' do
+      expect(subject.permanent_error?('some argument error', 'ArgumentError')).to be(true)
+    end
+
+    it 'identifies network errors as transient' do
+      expect(subject.permanent_error?('connection timeout', 'Timeout::Error')).to be(false)
+    end
+
+    it 'identifies socket errors as transient' do
+      expect(subject.permanent_error?('connection refused', 'Errno::ECONNREFUSED')).to be(false)
+    end
+
+    it 'returns false for nil inputs' do
+      expect(subject.permanent_error?(nil, nil)).to be(false)
+      expect(subject.permanent_error?('error', nil)).to be(false)
+      expect(subject.permanent_error?(nil, 'Error')).to be(false)
+    end
+  end
+
   describe '#move_pending_vm_to_ready' do
     let(:host) { { 'hostname' => vm }}
 
@@ -987,7 +1107,8 @@ EOT
     context 'with no errors during cloning' do
       before(:each) do
         allow(metrics).to receive(:timing)
-        expect(metrics).to receive(:timing).with(/clone\./,/0/)
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with(/vmpooler_clone\./,/0/)
         expect(provider).to receive(:create_vm).with(pool, String)
         allow(provider).to receive(:get_vm_ip_address).and_return(1)
         allow(subject).to receive(:get_domain_for_pool).and_return('example.com')
@@ -1038,7 +1159,8 @@ EOT
     context 'with a failure to get ip address after cloning' do
       it 'should log a message that it completed being cloned' do
         allow(metrics).to receive(:timing)
-        expect(metrics).to receive(:timing).with(/clone\./,/0/)
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with(/vmpooler_clone\./,/0/)
         expect(provider).to receive(:create_vm).with(pool, String)
         allow(provider).to receive(:get_vm_ip_address).and_return(nil)
 
@@ -1097,7 +1219,8 @@ EOT
     context 'with request_id' do
       before(:each) do
         allow(metrics).to receive(:timing)
-        expect(metrics).to receive(:timing).with(/clone\./,/0/)
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with(/vmpooler_clone\./,/0/)
         expect(provider).to receive(:create_vm).with(pool, String)
         allow(provider).to receive(:get_vm_ip_address).with(vm,pool).and_return(1)
         allow(subject).to receive(:get_dns_plugin_class_name_for_pool).and_return(dns_plugin)
@@ -1135,7 +1258,7 @@ EOT
         resolv = class_double("Resolv").as_stubbed_const(:transfer_nested_constants => true)
         expect(subject).to receive(:generate_and_check_hostname).exactly(3).times.and_return([vm_name, true]) #skip this, make it available all times
         expect(resolv).to receive(:getaddress).exactly(3).times.and_return("1.2.3.4")
-        expect(metrics).to receive(:increment).with("errors.staledns.#{pool}").exactly(3).times
+        expect(metrics).to receive(:increment).with("vmpooler_errors.staledns.#{pool}").exactly(3).times
         expect{subject._clone_vm(pool,provider,dns_plugin)}.to raise_error(/Unable to generate a unique hostname after/)
       end
       it 'should be successful if DNS does not exist' do
@@ -1233,7 +1356,8 @@ EOT
       it 'should emit a timing metric' do
         allow(subject).to receive(:get_vm_usage_labels)
         allow(metrics).to receive(:timing)
-        expect(metrics).to receive(:timing).with("destroy.#{pool}", String)
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with("vmpooler_destroy.#{pool}", String)
 
         subject._destroy_vm(vm,pool,provider,dns_plugin)
       end
@@ -5054,6 +5178,44 @@ EOT
       end
     end
 
+    context 'when request is already marked as failed' do
+      let(:request_string) { "#{pool}:#{pool}:1" }
+      before(:each) do
+        redis_connection_pool.with do |redis|
+          create_ondemand_request_for_test(request_id, current_time.to_i, request_string, redis)
+          set_ondemand_request_status(request_id, 'failed', redis)
+        end
+      end
+
+      it 'logs that the request is already failed' do
+        redis_connection_pool.with do |redis|
+          expect(logger).to receive(:log).with('s', "Request '#{request_id}' already marked as failed, skipping VM creation")
+          subject.create_ondemand_vms(request_id, redis)
+        end
+      end
+
+      it 'removes the request from provisioning__request queue' do
+        redis_connection_pool.with do |redis|
+          subject.create_ondemand_vms(request_id, redis)
+          expect(redis.zscore('vmpooler__provisioning__request', request_id)).to be_nil
+        end
+      end
+
+      it 'does not create VM tasks' do
+        redis_connection_pool.with do |redis|
+          subject.create_ondemand_vms(request_id, redis)
+          expect(redis.zcard('vmpooler__odcreate__task')).to eq(0)
+        end
+      end
+
+      it 'does not add to provisioning__processing queue' do
+        redis_connection_pool.with do |redis|
+          subject.create_ondemand_vms(request_id, redis)
+          expect(redis.zscore('vmpooler__provisioning__processing', request_id)).to be_nil
+        end
+      end
+    end
+
     context 'with a request that has data' do
       let(:request_string) { "#{pool}:#{pool}:1" }
       before(:each) do

spec/unit/queue_reliability_spec.rb

@@ -0,0 +1,497 @@
+# frozen_string_literal: true
+
+require 'spec_helper'
+require 'vmpooler/pool_manager'
+
+describe 'Vmpooler::PoolManager - Queue Reliability Features' do
+  let(:logger) { MockLogger.new }
+  let(:redis_connection_pool) { ConnectionPool.new(size: 1) { redis } }
+  let(:metrics) { Vmpooler::Metrics::DummyStatsd.new }
+  let(:config) { YAML.load(<<~EOT
+    ---
+    :config:
+      task_limit: 10
+      vm_checktime: 1
+      vm_lifetime: 12
+      prefix: 'pooler-'
+      dlq_enabled: true
+      dlq_ttl: 168
+      dlq_max_entries: 100
+      purge_enabled: true
+      purge_dry_run: false
+      max_pending_age: 7200
+      max_ready_age: 86400
+      max_completed_age: 3600
+      health_check_enabled: true
+      health_check_interval: 300
+      health_thresholds:
+        pending_queue_max: 100
+        ready_queue_max: 500
+        dlq_max_warning: 100
+        dlq_max_critical: 1000
+        stuck_vm_age_threshold: 7200
+    :providers:
+      :dummy: {}
+    :pools:
+      - name: 'test-pool'
+        size: 5
+        provider: 'dummy'
+    EOT
+    )
+  }
+
+  subject { Vmpooler::PoolManager.new(config, logger, redis_connection_pool, metrics) }
+
+  describe 'Dead-Letter Queue (DLQ)' do
+    let(:vm) { 'vm-abc123' }
+    let(:pool) { 'test-pool' }
+    let(:error_class) { 'StandardError' }
+    let(:error_message) { 'template does not exist' }
+    let(:request_id) { 'req-123' }
+    let(:pool_alias) { 'test-alias' }
+
+    before(:each) do
+      redis_connection_pool.with do |redis_connection|
+        allow(redis_connection).to receive(:zadd)
+        allow(redis_connection).to receive(:zcard).and_return(0)
+        allow(redis_connection).to receive(:expire)
+      end
+    end
+
+    describe '#dlq_enabled?' do
+      it 'returns true when dlq_enabled is true in config' do
+        expect(subject.dlq_enabled?).to be true
+      end
+
+      it 'returns false when dlq_enabled is false in config' do
+        config[:config]['dlq_enabled'] = false
+        expect(subject.dlq_enabled?).to be false
+      end
+    end
+
+    describe '#dlq_ttl' do
+      it 'returns configured TTL' do
+        expect(subject.dlq_ttl).to eq(168)
+      end
+
+      it 'returns default TTL when not configured' do
+        config[:config].delete('dlq_ttl')
+        expect(subject.dlq_ttl).to eq(168)
+      end
+    end
+
+    describe '#dlq_max_entries' do
+      it 'returns configured max entries' do
+        expect(subject.dlq_max_entries).to eq(100)
+      end
+
+      it 'returns default max entries when not configured' do
+        config[:config].delete('dlq_max_entries')
+        expect(subject.dlq_max_entries).to eq(10000)
+      end
+    end
+
+    describe '#move_to_dlq' do
+      context 'when DLQ is enabled' do
+        it 'adds entry to DLQ sorted set' do
+          redis_connection_pool.with do |redis_connection|
+            dlq_key = 'vmpooler__dlq__pending'
+            
+            expect(redis_connection).to receive(:zadd).with(dlq_key, anything, anything)
+            expect(redis_connection).to receive(:expire).with(dlq_key, anything)
+            
+            subject.move_to_dlq(vm, pool, 'pending', error_class, error_message, 
+                               redis_connection, request_id: request_id, pool_alias: pool_alias)
+          end
+        end
+
+        it 'includes error details in DLQ entry' do
+          redis_connection_pool.with do |redis_connection|
+            expect(redis_connection).to receive(:zadd) do |_key, _score, entry|
+              expect(entry).to include(vm)
+              expect(entry).to include(error_message)
+              expect(entry).to include(error_class)
+            end
+            
+            subject.move_to_dlq(vm, pool, 'pending', error_class, error_message, redis_connection)
+          end
+        end
+
+        it 'increments DLQ metrics' do
+          redis_connection_pool.with do |redis_connection|
+            expect(metrics).to receive(:increment).with('vmpooler_dlq.pending.count')
+            
+            subject.move_to_dlq(vm, pool, 'pending', error_class, error_message, redis_connection)
+          end
+        end
+
+        it 'enforces max entries limit' do
+          redis_connection_pool.with do |redis_connection|
+            allow(redis_connection).to receive(:zcard).and_return(150)
+            expect(redis_connection).to receive(:zremrangebyrank).with(anything, 0, 49)
+            
+            subject.move_to_dlq(vm, pool, 'pending', error_class, error_message, redis_connection)
+          end
+        end
+      end
+
+      context 'when DLQ is disabled' do
+        before { config[:config]['dlq_enabled'] = false }
+
+        it 'does not add entry to DLQ' do
+          redis_connection_pool.with do |redis_connection|
+            expect(redis_connection).not_to receive(:zadd)
+            
+            subject.move_to_dlq(vm, pool, 'pending', error_class, error_message, redis_connection)
+          end
+        end
+      end
+    end
+  end
+
+  describe 'Auto-Purge' do
+    describe '#purge_enabled?' do
+      it 'returns true when purge_enabled is true in config' do
+        expect(subject.purge_enabled?).to be true
+      end
+
+      it 'returns false when purge_enabled is false in config' do
+        config[:config]['purge_enabled'] = false
+        expect(subject.purge_enabled?).to be false
+      end
+    end
+
+    describe '#purge_dry_run?' do
+      it 'returns false when purge_dry_run is false in config' do
+        expect(subject.purge_dry_run?).to be false
+      end
+
+      it 'returns true when purge_dry_run is true in config' do
+        config[:config]['purge_dry_run'] = true
+        expect(subject.purge_dry_run?).to be true
+      end
+    end
+
+    describe '#max_pending_age' do
+      it 'returns configured max age' do
+        expect(subject.max_pending_age).to eq(7200)
+      end
+
+      it 'returns default max age when not configured' do
+        config[:config].delete('max_pending_age')
+        expect(subject.max_pending_age).to eq(7200)
+      end
+    end
+
+    describe '#purge_pending_queue' do
+      let(:pool) { 'test-pool' }
+      let(:old_vm) { 'vm-old' }
+      let(:new_vm) { 'vm-new' }
+
+      before(:each) do
+        redis_connection_pool.with do |redis_connection|
+          # Old VM (3 hours old, exceeds 2 hour threshold)
+          redis_connection.sadd("vmpooler__pending__#{pool}", old_vm)
+          redis_connection.hset("vmpooler__vm__#{old_vm}", 'clone', (Time.now - 10800).to_s)
+          
+          # New VM (30 minutes old, within threshold)
+          redis_connection.sadd("vmpooler__pending__#{pool}", new_vm)
+          redis_connection.hset("vmpooler__vm__#{new_vm}", 'clone', (Time.now - 1800).to_s)
+        end
+      end
+
+      context 'when not in dry-run mode' do
+        it 'purges stale pending VMs' do
+          redis_connection_pool.with do |redis_connection|
+            purged_count = subject.purge_pending_queue(pool, redis_connection)
+            
+            expect(purged_count).to eq(1)
+            expect(redis_connection.sismember("vmpooler__pending__#{pool}", old_vm)).to be false
+            expect(redis_connection.sismember("vmpooler__pending__#{pool}", new_vm)).to be true
+          end
+        end
+
+        it 'moves purged VMs to DLQ' do
+          redis_connection_pool.with do |redis_connection|
+            expect(subject).to receive(:move_to_dlq).with(
+              old_vm, pool, 'pending', 'Purge', anything, redis_connection, anything
+            )
+            
+            subject.purge_pending_queue(pool, redis_connection)
+          end
+        end
+
+        it 'increments purge metrics' do
+          redis_connection_pool.with do |redis_connection|
+            expect(metrics).to receive(:increment).with("vmpooler_purge.pending.#{pool}.count")
+            
+            subject.purge_pending_queue(pool, redis_connection)
+          end
+        end
+      end
+
+      context 'when in dry-run mode' do
+        before { config[:config]['purge_dry_run'] = true }
+
+        it 'detects but does not purge stale VMs' do
+          redis_connection_pool.with do |redis_connection|
+            purged_count = subject.purge_pending_queue(pool, redis_connection)
+            
+            expect(purged_count).to eq(1)
+            expect(redis_connection.sismember("vmpooler__pending__#{pool}", old_vm)).to be true
+          end
+        end
+
+        it 'does not move to DLQ' do
+          redis_connection_pool.with do |redis_connection|
+            expect(subject).not_to receive(:move_to_dlq)
+            
+            subject.purge_pending_queue(pool, redis_connection)
+          end
+        end
+      end
+    end
+
+    describe '#purge_ready_queue' do
+      let(:pool) { 'test-pool' }
+      let(:old_vm) { 'vm-old-ready' }
+      let(:new_vm) { 'vm-new-ready' }
+
+      before(:each) do
+        redis_connection_pool.with do |redis_connection|
+          # Old VM (25 hours old, exceeds 24 hour threshold)
+          redis_connection.sadd("vmpooler__ready__#{pool}", old_vm)
+          redis_connection.hset("vmpooler__vm__#{old_vm}", 'ready', (Time.now - 90000).to_s)
+          
+          # New VM (2 hours old, within threshold)
+          redis_connection.sadd("vmpooler__ready__#{pool}", new_vm)
+          redis_connection.hset("vmpooler__vm__#{new_vm}", 'ready', (Time.now - 7200).to_s)
+        end
+      end
+
+      it 'moves stale ready VMs to completed queue' do
+        redis_connection_pool.with do |redis_connection|
+          purged_count = subject.purge_ready_queue(pool, redis_connection)
+          
+          expect(purged_count).to eq(1)
+          expect(redis_connection.sismember("vmpooler__ready__#{pool}", old_vm)).to be false
+          expect(redis_connection.sismember("vmpooler__completed__#{pool}", old_vm)).to be true
+          expect(redis_connection.sismember("vmpooler__ready__#{pool}", new_vm)).to be true
+        end
+      end
+    end
+
+    describe '#purge_completed_queue' do
+      let(:pool) { 'test-pool' }
+      let(:old_vm) { 'vm-old-completed' }
+      let(:new_vm) { 'vm-new-completed' }
+
+      before(:each) do
+        redis_connection_pool.with do |redis_connection|
+          # Old VM (2 hours old, exceeds 1 hour threshold)
+          redis_connection.sadd("vmpooler__completed__#{pool}", old_vm)
+          redis_connection.hset("vmpooler__vm__#{old_vm}", 'destroy', (Time.now - 7200).to_s)
+          
+          # New VM (30 minutes old, within threshold)
+          redis_connection.sadd("vmpooler__completed__#{pool}", new_vm)
+          redis_connection.hset("vmpooler__vm__#{new_vm}", 'destroy', (Time.now - 1800).to_s)
+        end
+      end
+
+      it 'removes stale completed VMs' do
+        redis_connection_pool.with do |redis_connection|
+          purged_count = subject.purge_completed_queue(pool, redis_connection)
+          
+          expect(purged_count).to eq(1)
+          expect(redis_connection.sismember("vmpooler__completed__#{pool}", old_vm)).to be false
+          expect(redis_connection.sismember("vmpooler__completed__#{pool}", new_vm)).to be true
+        end
+      end
+    end
+  end
+
+  describe 'Health Checks' do
+    describe '#health_check_enabled?' do
+      it 'returns true when health_check_enabled is true in config' do
+        expect(subject.health_check_enabled?).to be true
+      end
+
+      it 'returns false when health_check_enabled is false in config' do
+        config[:config]['health_check_enabled'] = false
+        expect(subject.health_check_enabled?).to be false
+      end
+    end
+
+    describe '#health_thresholds' do
+      it 'returns configured thresholds' do
+        thresholds = subject.health_thresholds
+        expect(thresholds['pending_queue_max']).to eq(100)
+        expect(thresholds['stuck_vm_age_threshold']).to eq(7200)
+      end
+
+      it 'merges with defaults when partially configured' do
+        config[:config]['health_thresholds'] = { 'pending_queue_max' => 200 }
+        thresholds = subject.health_thresholds
+        
+        expect(thresholds['pending_queue_max']).to eq(200)
+        expect(thresholds['ready_queue_max']).to eq(500) # default
+      end
+    end
+
+    describe '#calculate_queue_ages' do
+      let(:pool) { 'test-pool' }
+      let(:vm1) { 'vm-1' }
+      let(:vm2) { 'vm-2' }
+      let(:vm3) { 'vm-3' }
+
+      before(:each) do
+        redis_connection_pool.with do |redis_connection|
+          redis_connection.hset("vmpooler__vm__#{vm1}", 'clone', (Time.now - 3600).to_s)
+          redis_connection.hset("vmpooler__vm__#{vm2}", 'clone', (Time.now - 7200).to_s)
+          redis_connection.hset("vmpooler__vm__#{vm3}", 'clone', (Time.now - 1800).to_s)
+        end
+      end
+
+      it 'calculates ages for all VMs' do
+        redis_connection_pool.with do |redis_connection|
+          vms = [vm1, vm2, vm3]
+          ages = subject.calculate_queue_ages(vms, 'clone', redis_connection)
+          
+          expect(ages.length).to eq(3)
+          expect(ages[0]).to be_within(5).of(3600)
+          expect(ages[1]).to be_within(5).of(7200)
+          expect(ages[2]).to be_within(5).of(1800)
+        end
+      end
+
+      it 'skips VMs with missing timestamps' do
+        redis_connection_pool.with do |redis_connection|
+          vms = [vm1, 'vm-nonexistent', vm3]
+          ages = subject.calculate_queue_ages(vms, 'clone', redis_connection)
+          
+          expect(ages.length).to eq(2)
+        end
+      end
+    end
+
+    describe '#determine_health_status' do
+      let(:base_metrics) do
+        {
+          'queues' => {
+            'test-pool' => {
+              'pending' => { 'size' => 10, 'stuck_count' => 2 },
+              'ready' => { 'size' => 50 }
+            }
+          },
+          'errors' => {
+            'dlq_total_size' => 50,
+            'stuck_vm_count' => 2
+          }
+        }
+      end
+
+      it 'returns healthy when all metrics are within thresholds' do
+        status = subject.determine_health_status(base_metrics)
+        expect(status).to eq('healthy')
+      end
+
+      it 'returns degraded when DLQ size exceeds warning threshold' do
+        metrics = base_metrics.dup
+        metrics['errors']['dlq_total_size'] = 150
+        
+        status = subject.determine_health_status(metrics)
+        expect(status).to eq('degraded')
+      end
+
+      it 'returns unhealthy when DLQ size exceeds critical threshold' do
+        metrics = base_metrics.dup
+        metrics['errors']['dlq_total_size'] = 1500
+        
+        status = subject.determine_health_status(metrics)
+        expect(status).to eq('unhealthy')
+      end
+
+      it 'returns degraded when pending queue exceeds warning threshold' do
+        metrics = base_metrics.dup
+        metrics['queues']['test-pool']['pending']['size'] = 120
+        
+        status = subject.determine_health_status(metrics)
+        expect(status).to eq('degraded')
+      end
+
+      it 'returns unhealthy when pending queue exceeds critical threshold' do
+        metrics = base_metrics.dup
+        metrics['queues']['test-pool']['pending']['size'] = 250
+        
+        status = subject.determine_health_status(metrics)
+        expect(status).to eq('unhealthy')
+      end
+
+      it 'returns unhealthy when stuck VM count exceeds critical threshold' do
+        metrics = base_metrics.dup
+        metrics['errors']['stuck_vm_count'] = 60
+        
+        status = subject.determine_health_status(metrics)
+        expect(status).to eq('unhealthy')
+      end
+    end
+
+    describe '#push_health_metrics' do
+      let(:metrics_data) do
+        {
+          'queues' => {
+            'test-pool' => {
+              'pending' => { 'size' => 10, 'oldest_age' => 3600, 'stuck_count' => 2 },
+              'ready' => { 'size' => 50, 'oldest_age' => 7200 },
+              'completed' => { 'size' => 5 }
+            }
+          },
+          'tasks' => {
+            'clone' => { 'active' => 3 },
+            'ondemand' => { 'active' => 2, 'pending' => 5 }
+          },
+          'errors' => {
+            'dlq_total_size' => 25,
+            'stuck_vm_count' => 2,
+            'orphaned_metadata_count' => 3
+          }
+        }
+      end
+
+      it 'pushes status metric' do
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.status', 0)
+        
+        subject.push_health_metrics(metrics_data, 'healthy')
+      end
+
+      it 'pushes error metrics' do
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.dlq.total_size', 25)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.stuck_vms.count', 2)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.orphaned_metadata.count', 3)
+        
+        subject.push_health_metrics(metrics_data, 'healthy')
+      end
+
+      it 'pushes per-pool queue metrics' do
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.queue.test-pool.pending.size', 10)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.queue.test-pool.pending.oldest_age', 3600)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.queue.test-pool.pending.stuck_count', 2)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.queue.test-pool.ready.size', 50)
+        
+        subject.push_health_metrics(metrics_data, 'healthy')
+      end
+
+      it 'pushes task metrics' do
+        allow(metrics).to receive(:gauge)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.tasks.clone.active', 3)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.tasks.ondemand.active', 2)
+        expect(metrics).to receive(:gauge).with('vmpooler_health.tasks.ondemand.pending', 5)
+        
+        subject.push_health_metrics(metrics_data, 'healthy')
+      end
+    end
+  end
+end

update-gemfile-lock

@@ -3,5 +3,5 @@
 # The container tag should closely match what is used in `docker/Dockerfile` in vmpooler-deployment
 docker run -it --rm \
   -v $(pwd):/app \
-  jruby:9.4.3.0-jdk11 \
+  jruby:9.4.12.1-jdk11 \
-  /bin/bash -c 'apt-get update -qq && apt-get install -y --no-install-recommends git make netbase && cd /app && gem install bundler && bundle install --jobs 3 && bundle update; echo "LOCK_FILE_UPDATE_EXIT_CODE=$?"'
+  /bin/bash -c 'apt-get update -qq && apt-get install -y --no-install-recommends git make netbase build-essential && cd /app && gem install bundler && bundle install --jobs 3 && bundle update; echo "LOCK_FILE_UPDATE_EXIT_CODE=$?"'

vmpooler.gemspec

@@ -23,10 +23,11 @@ Gem::Specification.new do |s|
   s.add_dependency 'opentelemetry-exporter-jaeger', '= 0.23.0'
   s.add_dependency 'opentelemetry-instrumentation-concurrent_ruby', '= 0.21.1'
   s.add_dependency 'opentelemetry-instrumentation-http_client', '= 0.22.2'
+  s.add_dependency 'opentelemetry-instrumentation-rack', '= 0.23.4'
   s.add_dependency 'opentelemetry-instrumentation-redis', '= 0.25.3'
   s.add_dependency 'opentelemetry-instrumentation-sinatra', '= 0.23.2'
-  s.add_dependency 'opentelemetry-resource_detectors', '= 0.24.1'
+  s.add_dependency 'opentelemetry-resource_detectors', '= 0.24.2'
-  s.add_dependency 'opentelemetry-sdk', '~> 1.3', '>= 1.3.0'
+  s.add_dependency 'opentelemetry-sdk', '~> 1.8'
   s.add_dependency 'pickup', '~> 0.0.11'
   s.add_dependency 'prometheus-client', '>= 2', '< 5'
   s.add_dependency 'puma', '>= 5.0.4', '< 7'
@@ -39,7 +40,7 @@ Gem::Specification.new do |s|
 
   # Testing dependencies
   s.add_development_dependency 'climate_control', '>= 0.2.0'
-  s.add_development_dependency 'mock_redis', '>= 0.17.0'
+  s.add_development_dependency 'mock_redis', '= 0.37.0'
   s.add_development_dependency 'pry'
   s.add_development_dependency 'rack-test', '>= 0.6'
   s.add_development_dependency 'rspec', '>= 3.2'

vmpooler.yaml.example

@@ -456,6 +456,12 @@
 #     How long (in minutes) before marking a clone in 'pending' queues as 'failed' and retrying.
 #     (default: 15)
 #
+#   - max_vm_retries
+#     Maximum number of times to retry VM creation for a failed request before marking it as permanently failed.
+#     This helps prevent infinite retry loops when there are configuration issues like invalid template paths.
+#     Permanent errors (like invalid template paths) are detected and will not be retried.
+#     (default: 3)
+#
 #   - vm_checktime
 #     How often (in minutes) to check the sanity of VMs in 'ready' queues.
 #     (default: 1)
@@ -619,6 +625,7 @@
   vm_checktime: 1
   vm_lifetime: 12
   vm_lifetime_auth: 24
+  max_vm_retries: 3
   allowed_tags:
     - 'created_by'
     - 'project'

vmpooler.yml.example

@@ -0,0 +1,92 @@
+---
+# VMPooler Configuration Example with Dead-Letter Queue, Auto-Purge, and Health Checks
+
+# Redis Configuration
+:redis:
+  server: 'localhost'
+  port: 6379
+  data_ttl: 168  # hours - how long to keep VM metadata in Redis
+  
+  # Dead-Letter Queue (DLQ) Configuration
+  dlq_enabled: true
+  dlq_ttl: 168  # hours (7 days) - how long to keep DLQ entries
+  dlq_max_entries: 10000  # maximum entries per DLQ queue before trimming
+
+# Application Configuration
+:config:
+  # ... other existing config ...
+  
+  # Dead-Letter Queue (DLQ) - Optional, defaults shown
+  dlq_enabled: false  # Set to true to enable DLQ
+  dlq_ttl: 168  # hours (7 days)
+  dlq_max_entries: 10000  # per DLQ queue
+  
+  # Auto-Purge Stale Queue Entries
+  purge_enabled: false  # Set to true to enable auto-purge
+  purge_interval: 3600  # seconds (1 hour) - how often to run purge cycle
+  purge_dry_run: false  # Set to true to log what would be purged without actually purging
+  
+  # Auto-Purge Age Thresholds (in seconds)
+  max_pending_age: 7200  # 2 hours - VMs stuck in pending
+  max_ready_age: 86400  # 24 hours - VMs idle in ready queue
+  max_completed_age: 3600  # 1 hour - VMs in completed queue
+  max_orphaned_age: 86400  # 24 hours - orphaned VM metadata
+  max_request_age: 86400  # 24 hours - stale on-demand requests
+  
+  # Health Checks
+  health_check_enabled: false  # Set to true to enable health checks
+  health_check_interval: 300  # seconds (5 minutes) - how often to run health checks
+  
+  # Health Check Thresholds
+  health_thresholds:
+    pending_queue_max: 100  # Warning threshold for pending queue size
+    ready_queue_max: 500  # Warning threshold for ready queue size
+    dlq_max_warning: 100  # Warning threshold for DLQ size
+    dlq_max_critical: 1000  # Critical threshold for DLQ size
+    stuck_vm_age_threshold: 7200  # 2 hours - age at which VM is considered "stuck"
+    stuck_vm_max_warning: 10  # Warning threshold for stuck VM count
+    stuck_vm_max_critical: 50  # Critical threshold for stuck VM count
+
+# Pool Configuration
+:pools:
+  - name: 'centos-7-x86_64'
+    size: 5
+    provider: 'vsphere'
+    # ... other pool settings ...
+
+# Provider Configuration
+:providers:
+  :vsphere:
+    server: 'vcenter.example.com'
+    username: 'vmpooler'
+    password: 'secret'
+    # ... other provider settings ...
+
+# Example: Production Configuration
+# For production use, you might want:
+# :config:
+#   dlq_enabled: true
+#   dlq_ttl: 168  # Keep failed VMs for a week
+#   
+#   purge_enabled: true
+#   purge_interval: 1800  # Run every 30 minutes
+#   purge_dry_run: false
+#   max_pending_age: 3600  # Purge pending VMs after 1 hour
+#   max_ready_age: 172800  # Purge ready VMs after 2 days
+#   
+#   health_check_enabled: true
+#   health_check_interval: 300  # Check every 5 minutes
+
+# Example: Development Configuration
+# For development/testing, you might want:
+# :config:
+#   dlq_enabled: true
+#   dlq_ttl: 24  # Keep failed VMs for a day
+#   
+#   purge_enabled: true
+#   purge_interval: 600  # Run every 10 minutes
+#   purge_dry_run: true  # Test mode - log but don't actually purge
+#   max_pending_age: 1800  # More aggressive - 30 minutes
+#   
+#   health_check_enabled: true
+#   health_check_interval: 60  # Check every minute