investigating where to ignore poststop task in alloc health tracker #9548
Conversation
drewbailey
force-pushed
the
drew/investigate-poststop-health
branch
from
75eb7c9
to
d272803
Compare
client/allochealth/tracker.go
Outdated
| if state.State == structs.TaskStatePending { | ||
| // Ignore poststop since it will be pending until the main job exists | ||
| if (state.State == structs.TaskStatePending) && | ||
| t.lifecycleTasks[taskName] != structs.TaskLifecycleHookPoststop { |
There was a problem hiding this comment.
Do we ever want to consider poststop hooks as part of a deployment? I don't think so in which case we should probably if poststop { continue } at the top of this loop? Otherwise I think if this code gets hit after poststop has started its started time will get taken into consideration?
There was a problem hiding this comment.
good question! @jazzyfresh @notnoop, I'm very unfamiliar with lifecycle hooks :D
There was a problem hiding this comment.
Agreed @schmichael. In both service & batch cases, deployments should watch for everything but the poststop tasks
drewbailey
force-pushed
the
drew/investigate-poststop-health
branch
from
0342656
to
a9436ba
Compare
vercel
bot
temporarily deployed
to
Preview – nomad-storybook-and-ui
vercel
bot
temporarily deployed
to
Preview – nomad-storybook-and-ui
* investigating where to ignore poststop task in alloc health tracker * ignore poststop when setting latest start time for allocation * clean up logic * lifecycle: isolate mocks for poststop deployment test * lifecycle: update comments in tracker Co-authored-by: Jasmine Dahilig <jasmine@dahilig.com>
|
I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions. |
Ignores poststop jobs pending state when calculating the lastest start time for an allocation
resolves #9361