feat(health): add Logstash resource customization #22028
Conversation
Signed-off-by: Francis Provencher <fprovencher@drw.com>
❗ Preview Environment undeploy from Bunnyshell failedSee: Environment Details | Pipeline Logs Available commands (reply to this comment):
|
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #22028 +/- ##
==========================================
- Coverage 60.03% 60.01% -0.03%
==========================================
Files 344 344
Lines 57781 57787 +6
==========================================
- Hits 34689 34680 -9
- Misses 20333 20341 +8
- Partials 2759 2766 +7 ☔ View full report in Codecov by Sentry. |
| @@ -0,0 +1,22 @@ | |||
| local hs = {} | |||
| if obj.status ~= nil then | |||
| if obj.status.health ~= nil then | |||
There was a problem hiding this comment.
Can you consolidate in one if using and?
Signed-off-by: Francis Provencher <fprovencher@drw.com>
| end | ||
|
|
||
| hs.status = "Unknown" | ||
| hs.message = "Logstash status is unknown. Ensure your ArgoCD is current and then check for/file a bug report: https://github.com/argoproj/argo-cd/issues" |
There was a problem hiding this comment.
Usually, if the resource does not have the status populated yet, the default behavior is to be progressing and waiting for the controller to populate it.
I think you could have an unknown status if obj.status.health is of an unknown colour. Then the message can say which colour it is.
hs.status = "Progressing"
hs.message = "Waiting for Logstash"
There was a problem hiding this comment.
Done.
As per the source of Logstash, if health exist, then a colour will be defined, so I left the rest of the code as-is.
Signed-off-by: Francis Provencher <fprovencher@drw.com>
|
@agaudreault any remaining concerns, or good to go? |
Signed-off-by: Francis Provencher <fprovencher@drw.com> Signed-off-by: dsuhinin <suhinin.dmitriy@gmail.com>
Signed-off-by: Francis Provencher <fprovencher@drw.com> Signed-off-by: Jonathan Ogilvie <jonathan.ogilvie@sumologic.com>
Signed-off-by: Francis Provencher <fprovencher@drw.com> Signed-off-by: enneitex <etienne.divet@gmail.com>
Adding custom health check logic for logstash.k8s.elastic.co/Logstash objects.
This is based on the health output that is defined here:
https://github.com/elastic/cloud-on-k8s/blob/872ca20e68404e740c879edced3933995e202316/pkg/controller/logstash/health.go#L13
Which is called from:
https://github.com/elastic/cloud-on-k8s/blob/872ca20e68404e740c879edced3933995e202316/pkg/controller/agent/reconcile.go#L217
I didn't want to reproduce the logic from the app (which could change), which is why I kept it simple.
Checklist: