puppetlabs/vmpooler
2
0
Fork 0
mirror of https://github.com/puppetlabs/vmpooler.git synced 2026-01-26 10:08:40 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions
146 commits 27 branches 74 tags 5.6 MiB
Commit graph

82 commits

Author SHA1 Message Date
Scott Schneider
5e24d1e44e Send 'running' stats to graphite 2014-02-11 12:51:36 -08:00
Scott Schneider
c0b2a8366c New configuration parameter 'vm_lifetime' 
This allows for configurable TTLs for hosts in the 'running' queue.
 2014-02-06 11:20:15 -08:00
Scott Schneider
47e2d8ab1e Clean up the 'deploy_vm' method 
Removing duplicate lines of code better accomplished with a '||'
fallback to the 'find_vm_heavy' method.
 2014-02-05 12:41:54 -08:00
Scott Schneider
693abcd2ff Past-tense log messages 2014-02-05 11:29:57 -08:00
Scott Schneider
5d52bc95da Adding missing 'start' variable 2014-02-04 15:13:38 -08:00
Scott Schneider
55ada2d80e Store 'clone' timestamps in Redis database 
Using redis keys to store clone timestamps will allow for a global TTL
for VMs in the 'pending' pool; failed clones should be cleaned up and
retried after a set period of time (eg. 15m).
 2014-02-04 15:05:24 -08:00
Scott Schneider
1f76308f44 Remove non-inventory VMs sitting in 'completed' queue 2014-02-04 10:34:55 -08:00
Scott Schneider
1003afaf00 Rework vCenter VM retrieval method 
Previous, the propertyCollector method was used to find VMs within
vCenter.  This method was ineffecient, as it would retrieve a list of
the entire vCenter inventory and then parse for the specified VM.  This
has been replaced with a fetch via searchIndex.FindByDnsName, which
returns only the requested VM object.
 2014-02-04 09:42:28 -08:00
Scott Schneider
ed22ed6e08 Remove VMs not found in vCenter inventory from 'ready' queues 2014-02-03 15:57:35 -08:00
Scott Schneider
5823f099c1 Catch clonevm_task exceptions 2014-02-03 10:26:21 -08:00
Scott Schneider
4ab3203d5b Syntax fix 2014-01-23 13:38:59 -08:00
Scott Schneider
a67a6a1bfa Recover from failed clones/deletes 2014-01-23 13:38:10 -08:00
Scott Schneider
175dbca9ca Implement clone rate-limiting 2014-01-21 16:28:41 -08:00
Scott Schneider
977478b207 Check 2013-12-13 11:22:06 -08:00
Scott Schneider
3c42155c18 Fixup for long-running destruction 2013-12-13 11:19:37 -08:00
Scott Schneider
9675981927 Delete long-running VMs after 12 hours 2013-12-13 10:13:28 -08:00
Scott Schneider
e3017beba1 Start keeping track of how long hosts are running 2013-12-12 10:04:49 -08:00
Scott Schneider
bcfca89730 Catch ALL THE EXCEPTIONS 2013-12-12 10:02:29 -08:00
Scott Schneider
31df80aa7c Each pool gets it's own vSphere object 2013-12-06 14:16:56 -08:00
Scott Schneider
5ea20c71b9 Additional object-existance checks 2013-12-06 12:05:47 -08:00
Scott Schneider
4f3c2aedd5 Missing an 'and' 2013-12-05 11:46:58 -08:00
Scott Schneider
674a3eaf6f Replace srem/sadd with smove, check power state of 'ready' and 'running' VMs 2013-12-05 11:38:11 -08:00
Scott Schneider
5de7a8b244 Only check VMs found in inventory 2013-12-04 11:37:51 -08:00
Scott Schneider
f687641c98 Separate ready_ttl and running_ttl config params 2013-12-04 11:31:23 -08:00
Scott Schneider
14bd3e0084 Use pool['running_ttl'], or a default 2013-12-04 11:25:46 -08:00
Scott Schneider
009cf789ab Quick hack to auto-destroy long-running VMs 2013-12-04 11:15:03 -08:00
Scott Schneider
7b866f788e Clear pending queue on init 2013-11-25 22:31:08 -08:00
Scott Schneider
7ac06dc4d6 Why so sleepy? 2013-11-25 22:29:59 -08:00
Scott Schneider
9228abb5d1 Remove (commented-out) race condition 2013-11-25 13:46:47 -08:00
Scott Schneider
4f64355241 Only one pools.each required... 2013-11-21 10:47:25 -08:00
Scott Schneider
e8881d601c Check for dead threads and restart them 2013-11-18 10:27:35 -08:00
Scott Schneider
8aab996666 Move pool-checks into a function 2013-11-18 09:59:01 -08:00
Scott Schneider
29ad2b9e09 Expire hosts in 'ready' pool with uptimes exceeding TTL 2013-11-13 13:55:51 -08:00
Scott Schneider
0a0177e226 Rename 'check_vm' function to 'check_pending_vm' 2013-11-13 13:45:50 -08:00
Scott Schneider
c80cac8def Send pool availability to graphite 2013-11-13 11:58:44 -08:00
Scott Schneider
ea0a6786a4 Fixing variable 2013-11-12 15:16:02 -08:00
Scott Schneider
44637cef86 Send destroy times to graphite 2013-11-12 15:11:48 -08:00
Scott Schneider
77fe40b59b Send metrics to graphite (if configured) 2013-11-12 15:03:34 -08:00
Scott Schneider
57975a489f Log to a (configurable) file 2013-11-08 10:48:51 -08:00
Scott Schneider
2f4bdaf67a Per-pool timeout settings 2013-11-06 17:28:41 -08:00
Scott Schneider
c3b6215961 Fewer vSphere queries = better 2013-11-06 15:56:32 -08:00
Scott Schneider
bc20975fba Revert "Consolidate scripts" 
This reverts commit 8f18acfd2d91e1f35671f36552a669eb45754433.
 2013-11-06 09:47:24 -08:00
Scott Schneider
485e6a0e4b Consolidate scripts 2013-11-06 09:36:41 -08:00
Scott Schneider
dfc36a519e Don't re-add hosts in 'running' pool 2013-11-05 15:25:55 -08:00
Scott Schneider
d957ddd089 Fail and redeploy if VM not 'ready' after 10 minutes 2013-11-05 14:31:35 -08:00
Scott Schneider
71a9150631 Revert "Only correlate Redis inventory with vCenter on startup" 
This reverts commit 658496f776f684dfcc31d8423bbafff38ea94e27.
 2013-11-05 14:13:09 -08:00
Scott Schneider
b7f922fb11 Only correlate Redis inventory with vCenter on startup 2013-11-05 13:32:10 -08:00
Scott Schneider
deea66b0da Refactor #1 2013-11-05 12:54:30 -08:00
Scott Schneider
a7d6b27847 Refactoring this mess of code 2013-11-05 12:54:08 -08:00
Scott Schneider
aea6e32e3d destroy_vm should clear out Redis queue 2013-11-05 10:25:43 -08:00