Bug 1756920: fluentd pods do not process kubernetes events#1766
Conversation
|
@richm: An error was encountered searching for bug 1756920 on the Bugzilla server at https://bugzilla.redhat.com:
DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
| json_fields "#{ENV['JSON_FIELDS'] || 'log,MESSAGE'}" | ||
| </filter> | ||
|
|
||
| <filter kubernetes.var.log.containers.eventrouter-** kubernetes.var.log.containers.cluster-logging-eventrouter-**> |
There was a problem hiding this comment.
cluster-logging-eventrouter is to workaround a "bug" in the current documentation - https://docs.openshift.com/container-platform/4.1/logging/efk-logging-eventrouter.html - it really should be eventrouter
3dbf294 to
ebfc2fd
Compare
|
I wonder if we should consider waiting until openshift/cluster-logging-operator#231 merges or if its urgent enough using this PR against a 4.2 release branch |
ebfc2fd to
a28e9b8
Compare
|
/retest |
|
/lgtm |
|
/retest |
|
A version error? Maybe, most 4.2 is okay, except the last one? |
Nothing is going to pass until #1765 merges, which is dependent on openshift/elasticsearch-operator#191 |
|
/retest Please review the full test history for this PR and help us cut down flakes. |
a28e9b8 to
25037a7
Compare
|
please re-review |
25037a7 to
7f6cdb4
Compare
|
/retest |
ce45207 to
0a87591
Compare
|
now passing - tests fixed - please review |
|
/retest |
|
problems in CI framework - not sure if related to 3.11 problems |
https://bugzilla.redhat.com/show_bug.cgi?id=1756920 This tries to keep the functionality the same in 4.x as it was in 3.x. In 3.x, you had to: - deploy the eventrouter with a pod name like "logging-eventrouter-*" - deploy the eventrouter in the default namespace - set `TRANSFORM_EVENTS=true` - set `MERGE_JSON_LOG=true` (which was the default in 3.x) In 4.x, the pod name changes to "eventrouter-*" to follow the convention of our other logging pods. The eventrouter defaults to running in "openshift-logging" - it doesn't really matter, as long as it is running in an "infra" namespace. This change also enables `MERGE_JSON_LOG=true` for eventrouter pods. The biggest problem is that you have to set the cluster to unmanaged in order to set `TRANSFORM_EVENTS=true`. We could workaround that by making `TRANSFORM_EVENTS=true` the default value. That would cause every kubernetes record to be checked to see if it looks like and event, and process it as such if it matches. I'm not sure what the performance implications would be. Also ports eventrouter test to 4.x Also adds a test for eventrouter Info type support update eventrouter to pick up fix for Bug 1701495
0a87591 to
c017949
Compare
|
flake - vpc limit reached |
|
multi-tenancy flake |
|
test-zzz-correct-index-names flake? ErrImagePull? |
|
ok - finally passed, no flakes - please review |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: nhosoi, richm The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
@richm: An error was encountered searching for bug 1756920 on the Bugzilla server at https://bugzilla.redhat.com:
DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
/cherrypick release-4.2 |
|
@richm: new pull request created: #1768 DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
https://bugzilla.redhat.com/show_bug.cgi?id=1756920
This tries to keep the functionality the same in 4.x as it was
in 3.x. In 3.x, you had to:
TRANSFORM_EVENTS=trueMERGE_JSON_LOG=true(which was the default in 3.x)In 4.x, the pod name changes to "eventrouter-*" to follow the convention
of our other logging pods. The eventrouter defaults to running
in "openshift-logging" - it doesn't really matter, as long as it is
running in an "infra" namespace. This change also enables
MERGE_JSON_LOG=truefor eventrouter pods.The biggest problem is that you have to set the cluster to unmanaged
in order to set
TRANSFORM_EVENTS=true. We could workaround thatby making
TRANSFORM_EVENTS=truethe default value. That wouldcause every kubernetes record to be checked to see if it looks like
and event, and process it as such if it matches. I'm not sure what
the performance implications would be.
Also ports eventrouter test to 4.x