-
Notifications
You must be signed in to change notification settings - Fork 96
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
feature(syslogng): move all default filters to client side #7041
Conversation
b0b71cd
to
df978f4
Compare
sdcm/sct_events/setup.py
Outdated
@@ -59,38 +56,6 @@ def start_events_device(log_dir: Optional[Union[str, Path]] = None, | |||
time.sleep(EVENTS_SUBSCRIBERS_START_DELAY) | |||
|
|||
# Default filters. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
please add some reference here that we also do filtering in syslog-ng clients
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I left some of the filter in place, since they are not db log or k8s specific (which works without syslog-ng, at least now)
df978f4
to
e76da56
Compare
00a329c
to
65a2dbf
Compare
5ecab51
to
7d482b8
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
but how about K8s where we don't use syslog-ng?
how about K8s where we don't use syslog-ng? |
there wouldn't be any filters. and we might get some of those issues surfacing regardless we have issue with logging on k8s, which we might end using syslog-ng or any maybe vector for it. let run some of the k8s provision test, and we'll see how bad it would be with those filters, and maybe we'll introduce them back only for k8s |
7d482b8
to
2fc0117
Compare
doing one more run as we are using in #7128 so it would be comparable |
do you know why the k8s-local-kind-aws doesn't work ?
|
The problem appeared using different image just trying to download it:
It is docker problem. Then, several days ago was released new docker version on ubuntu - It was around for a day then appeared new, fixed, version - So, I re-triggered the CI jobs for this PR assuming that fixed docker version will be used. |
2fc0117
to
a2f2fe8
Compare
since we have more and more filters on the log we apply by default trying to move them into client side, and not send them out at all, and not waste time apply those filter while reading the logs
so far those weren't print out in SCT log, now they wouldn't be printed and would be handled by the event system, to stop wasting our time on those line we are not importent enough for us.
taking all old events that are generating warning/normal events and filter on the db nodes, while removing those events. this would help improve the performence at the test end, and lower the amount of regexes used on test runner, and lowe amount of warning event generated
A new severity level, that would help us stop generating some events we need those events for skipping some log line, so we never generate error or critical events from those. until now we generated an event for each one, and now we'll be able to skip them since for some cases there are too many of those warnings.
a2f2fe8
to
11139d2
Compare
So, is it worth merging? |
looks like it's not really helpful, closing this for now |
since we have more and more filters on the log we apply by default trying to move them into client side, and not send them out at all, and not waste time apply those filter while reading the logs
Testing
PR pre-checks (self review)
backport
labelsReminders
sdcm/sct_config.py
)unit-test/
folder)