Testing on GH CI#1076
Closed
taurus-forever wants to merge 15 commits into16/edgefrom
Closed
Conversation
taurus-forever
force-pushed
the
alutay/is_restart_pending_test_ci
branch
2 times, most recently
from
3b97355 to
3381fb3
Compare
Codecov Report❌ Patch coverage is
❌ Your project status has failed because the head coverage (62.66%) is below the target coverage (70.00%). You can increase the head coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## 16/edge #1076 +/- ##
===========================================
- Coverage 64.87% 62.66% -2.21%
===========================================
Files 17 17
Lines 4270 4272 +2
Branches 656 636 -20
===========================================
- Hits 2770 2677 -93
- Misses 1333 1440 +107
+ Partials 167 155 -12 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
taurus-forever
force-pushed
the
alutay/is_restart_pending_test_ci
branch
3 times, most recently
from
1837b9c to
809daa0
Compare
The previous is_restart_pending() waited for long due to the Patroni's loop_wait default value (10 seconds), which tells how much time Patroni will wait before checking the configuration file again to reload it. Instead of checking PostgreSQL pending_restart from pg_settings, let's check Patroni API pending_restart=True flag.
The current Patroni 3.2.2 has wired/flickering behaviour:
it temporary flag pending_restart=True on many changes to REST API,
which is gone within a second but long enough to be cougth by charm.
Sleepping a bit is a necessary evil, until Patroni 3.3.0 upgrade.
The previous code sleept for 15 seconds waiting for pg_settings update.
Also, the unnecessary restarts could be triggered by missmatch of
Patroni config file and in-memory changes coming from REST API,
e.g. the slots were undefined in yaml file but set as an empty JSON {} => None.
Updating the default template to match the default API PATCHes and avoid restarts.
On topology observer event, the primary unit used to loose Primarly label.
Also: * use commong logger everywhere * and add several useful log messaged (e.g. DB connection) * remove no longer necessary debug 'Init class PostgreSQL' * align Patroni API requests style everhywhere * add Patroni API duration to debug logs
The list of IPs were randomly sorted causing unnecessary Partroni configuration re-generation with following Patroni restart/reload.
Housekeeping cleanup.
…hanged Those defers are necessary to support scale-up/scale-down during the refresh, while they have significalty slowdown PostgreSQL 16 bootstrap (and other daily related mainteinance tasks, like re-scaling, full node reboot/recovery, etc). Muting them for now with the proper documentation record to forbid rescaling during the refresh, untli we minimise amount of defers in PG16. Throw and warning for us to recall this promiss.
The current PG16 logic relies on Juju update-status or on_topology_change observer events, while in some cases we start Patroni without the Observer, causing a long waiting story till the next update-status arrives.
It is hard (impossible?) to catch the Juju Primary label manipulations from Juju debug-log. Logging it simplifyies troubleshooting.
We had to wait 30 seconds in case of lack of connection which is unnecessary long. Also, add details for the reason of failed connection Retry/CannotConnect.
It speedups the sinble unit app deployments.
taurus-forever
force-pushed
the
alutay/is_restart_pending_test_ci
branch
from
809daa0 to
269ed18
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Testing on GH CI...