-
Notifications
You must be signed in to change notification settings - Fork 8.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Ingest Manager] Assigning policy/config takes too long #77613
Comments
Pinging @elastic/ingest-management (Team:Ingest Management) |
are we sure the policy isn't applied? a lack of data may be related to a lack of activity on the host possibly? I don't think we're not seeing this in manual usage or other e2e tests. @mdelapenya has helped set up the Agent / Fleet e2e and he could help discuss our usage if desired. |
I do not thing the fleet checkin timeout is the issue here, the timeout is used in the checkin if there is no config change. If there is a config change the checkin will be immediate. If there is a config change the request will be the timeout duration. |
In our setup everything is in place - in this case sample access logs. Once the filebeat picks the config and starts processing files it is matter of seconds. |
as noted from @nchaulet the change should be immediate, I would be assuming the action_store has indeed the latest config. @michalpristas and @nchaulet can you pair on that issue. @mtojek what is the delay that you are seeing? |
Here is a trace from the latest failed test run:
17:51:19 ~ 17:53:21 - we assumed that 2 minutes should more than enough. For the second time:
17:55:50 ~ 17:57:49 - it passed just before timeout. Here is another successful run:
As you see this one it's relatively quick. |
@mtojek let me investigate a little more but looks like there is case I did not cover with config reassignment |
Hi @nchaulet, any progress on this issue? Just wondering since I'm about to merge a PR with a workaround in place: https://github.com/elastic/elastic-package/pull/111/files#diff-6d81fc25f6617cc18c88e703716754b9R23-R24. Thanks! |
@ycombinator I started on this but is a bigger change than what I thought #78493 |
Thanks for the update @nchaulet. I've subscribed to your PR now. |
Thanks for getting this fixed @nchaulet. Appreciate it! ❤️ |
Hi,
@ycombinator and I are working on system tests for integrations. While debugging a basic use case with Apache logs, we identified an issue around assigning policy to the Agent:
The reason behind this is the timeout problem on the Agent side:
As it's important for the system tests to be quick and simple, the workaround like
xpack.ingestManager.fleet.pollingRequestTimeout: 60000
doesn't seem to be the right choice. We appreciate any optimization on the path here to make the test execution shorter.The text was updated successfully, but these errors were encountered: