-
Notifications
You must be signed in to change notification settings - Fork 593
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
Sendconfig update improvements #1397
Conversation
in my testing, DBLESS and Postgres mode have both shown problems with repetitive sub-second updates. This is something we're tracking as a potential bug elsewhere, but for now increasing the interval seems to make the most sense for stability, as testing has shown it reduces/removes the timing problems which we've encountered with subsecond interval updates.
Recent updates have made our controller faster and having such a long Ingress timeout doesn't make sense any longer: it just increasing the time it takes to reach a timeout condition on certain kinds of failures.
Codecov Report
@@ Coverage Diff @@
## next #1397 +/- ##
==========================================
- Coverage 55.20% 55.12% -0.08%
==========================================
Files 42 42
Lines 3639 3646 +7
==========================================
+ Hits 2009 2010 +1
- Misses 1479 1485 +6
Partials 151 151
Continue to review full report at Codecov.
|
c60a87e
to
81e76be
Compare
The follow up issue for this PR (wherein we will further investigate after this mitigation) is: |
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.
LGTM.
In CI there have been several problems that appear to be due to the high rate at which KIC 2.0 has been sending updates to the Kong Admin API at a regular interval. This PR includes some improvements that seemed to function better with minimal/no end-user impact, while making the defaults less noisy in testing/CI. This also adds an update to
sendconfig.SendUpdate()
which will stifle repetitivenil
update warnings (you'll get one once for each update that would make no change now, rather than them repeating every few seconds).This helps to resolve #1390