This commit moves the migrate_vm logic to the vsphere provider. Without
this change migrate_vm has lots of vsphere specific logic in
pool_manager migrate_vm method.
* Fix no implicit conversion to rational from nil
Before this change if the boottime was nil, the check_ready
loop would exit on Time.now - host['boottime'] with a TypeError
in jruby. The boottime is nil when the power is Off so moving that check
earlier should catch that bug.
* set test data properly
Previously in commit 9b0e55f959 the looping period was changed from a static
number to a dynamic one depending on load, however this meant that the operation
to refill a pool was slowed down somewhat. While not a problem under normal
loads, when a pool was quickly consumed, the pool manager may not respond
quickly enough to refill the pool. This commit:
- Changes the sleep method, to us a helper sleep method that will wakeup
periodically and evaluate other wakeup events. This could be used later to
exist sleep loops when pooler is shutting down to stop blocking threads
- By default the wakeup_period is set to the minimum pool check loop time, thus
emulating the behaviour prior to commit 9b0e55f959
- Adds tests for the behaviour
Previously, if inventory failed for some reason, it would return an incomplete
set of VMs which could then cause the pool to perform off behaviours such as
fill the pool high than it should, or remove VMs which exist. Also, if the
redis cache of VMs in a pool had a VM but it did not actually exist in the
inventory it would never be removed.
This commit:
- Immediately exits the check_pool if an error occurs during inventory
collection
- Will mark a VM as completed if it exists in Redis, but does not exist in
inventory
- Adds tests for these behaviours
Sometimes this test would fail if the computer running the tests was under a
bit of load. This commit changes the expected output to be up to 1.99 seconds
instead of the previous 0.99 seconds.
Previously the check_pool would always check the pool every 5 seconds, however
with a large number of pools, this can cause resource issues inside the
providers. This commit:
- Introduces a dynamic check_pool period which increases during stability and
decreases when the pool is being change in an important way
- Surfaces the settings as global config defaults but can also be set on a per
pool basis
- Adds defaults to emulate the current behaviour
- Unit tests for the new behaviour
This commit updates find_least_used_compatible_host method to specify
the pool name when evaluating a VM for migration. Without this change VM
migration fails with a wrong number of arguments error. Pool_manager
test references are updated to reflect the change.
Refactoring the vmpooler.yaml format to support multiple providers.
The second level key under :providers: is a unique key name that
represents a provider that can be refered in the pool's parameter
called provider. The code is still backward compatible to support
the :vsphere: and :dummy: keys but in reality if you have more than
one vsphere configuration you would give them a different name. For
example :vsphere-pdx: and :vsphere-bfs: and the actual provider
class would be specified as a parameter called 'provider_class'.
See tests and examples for more information.
This commit modifies execute! to create the VM Providers on VMPooler startup
instead of check_pool creating a provider per pool. This commit also adds
legacy support for old configuration files:
- Setting the default provider for pools to be vsphere
- Copying VSphere connection settings in the configuration file from the legacy
location in the root, to under :providers/:vsphere which is new location for
all provider configuration
This commit adds a public function to access the internal variable holding the
VMPooler configuration. This is required for later commits for the execute!
function testing.
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the pool_manager to use the VM provider methods instead
- Removes the MockFindFolder class as it is no longer required
- Minor update for rubocop violations
Previously the Pool Manager would use a single VM provider per Pool. This
commit changes Pool Manager to use a single provider that services multiple
pools.
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the migrate_vm_and_record_timing method to use VM and Pool names
instead of VM and Pool objects.
Previously the Pool Manager would use vSphere objects directly. This commit
removes get_vm_host_info as this functionality is now in the vSphere VM
Provider.
Previously there was no simple way to calculate which pool a VM was a member of.
This commit adds a helper method which queries redis for the pool name for a
given VM.
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the pool_manager to use the VM provider methods instead
- Splits the destroy_vm function into two. One function spawns the thread
while the other actually does the work. This makes testing much easier.
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the pool_manager to use the VM provider methods instead
- Splits the check_ready_vm function into two. One function spawns the thread
while the other actually does the work. This makes testing much easier.
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the pool_manager to use the VM provider methods instead
- Modified to return true or false to indicate that the VM was failed
Previously the Pool Manager would use vSphere objects directly. This commit
- Modifies the pool_manager to use the VM provider methods instead
- Removes the open_socket method and tests as it is only required in the vSphere
VM provider
This commit creates a VM Provider test fixture for spec tests that merely uses
the Base Provider class with a name of mock_provider. This will then be used
by unit tests in further commits.
VM provisioning will be handled by VM Providers. This commit renames the use of
vsphere to provider where appropriate and changes the per-pool helper from
vsphere to providers to more accurately represent it's intended use.
Previously, the clone_vm method took various VSphere specific parameters e.g.
template folder. However in order make VMPooler less VSphere specific this
method should just take the pool configuration and then it can determine the
appropriate settings itself. This commit also moves the threading to a clone_vm
while the actual method which does the work is now _clone_vm as per all other
multithread worker methods in pool_manager. This commit also updates the spec
tests appropriately.
Previously in check_ready_vm, if the VM is powered off, the VM is moved in
redis however the function doesn't return there, and instead then checks if the
hostname is the same, and then if TCP socket 22 is open. This is unnecessary as
we already know the VM is turned off so of course the hostname is wrong and TCP
22 is unavailable. The same applies for the VM hostname.
This commit instead returns after it is found a VM is no longer ready. This
commit also amends the spec tests for the correct behaviour.
Add spec tests for check_snapshot_queue
Previously the check_snapshot_queue method would execute the loop indefinitely
as it did not have a terminating condition. This made it impossible to test.
This commit modifies the check_snapshot_queue method so that it can take a
maxloop and delay parameter so that it can be tested.
Add spec tests for check_disk_queue
Previously the check_disk_queue method would execute the loop indefinitely as it
did not have a terminating condition. This made it impossible to test. This
commit modifies the check_disk_queue method so that it can take a maxloop and
delay parameter so that it can be tested.
