integrate netobserv to reliability tests #798
Conversation
openshift-ci
bot
added
the
do-not-merge/hold
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: memodi The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/cc @qiliRedHat |
Co-authored-by: Qiujie Li <81630527+qiliRedHat@users.noreply.github.com>
|
I can add lgtm after openshift-qe/ocp-qe-perfscale-ci#659 is merged and integration test is done |
There was a problem hiding this comment.
@qiliRedHat do you want this to be separate config? I was thinking to have these tasks under standard reliability tasks and we'd have these checks run always.
Of course that also means we'd have to change netobserv operator would be installed by default for all reliability
runs and optionally exclude it, current implementation in start.sh is inverse, wdyt? this would be the initial idea had discuss so that netobserv could piggyback off your standard reliability runs.
If we have netobserv install by default, we'd need to up standard instance types or somehow have configuration that would accommodate Loki resource requirements, couple of questions here:
- Do you typically have infra nodes set up for reliability runs? if so, what are instance types of infra nodes? I can look if you can bring up that environment and I can see if Loki is able to fit on infra nodes.
- would you be willing to up the reliability run instance types to m5.2xlarge or m5.4xlarge?
Thanks!
There was a problem hiding this comment.
@memodi Reliability test has many test profiles, consider of the cost, I don't want to have noo on all profiles by default. Usually there is no infra nodes in reliability test. But reliability has the option to configure it with -i in start.sh, default size should be same as the worker, size can be configured.
If test it once (7 days) a release, I think we can do m5.2xlarge on one of the profiles.
|
/unhold |
openshift-ci
bot
removed
the
do-not-merge/hold
| rc_return = 1 | ||
| elif rc == 1 and result == "": | ||
| self.logger.info(f"Flowcollector is Ready.") | ||
| rc_return = 0 |
There was a problem hiding this comment.
@memodi You can add an 'else' to cover the rest of the cases: rc == 1 and result != "", or rc !=1. And log the result to see what you can get.
| rc_return = 1 | ||
| elif rc == 1 and result == "": | ||
| self.logger.info(f"Pods in ns ${ns} are healthy.") | ||
| rc_return = 0 |
NETOBSERV-1874 integrate netobserv to reliability tests
dependent on openshift-qe/ocp-qe-perfscale-ci#659
Additionally, it adds new improvements and fixes some bugs: