feat(crons): Limit clock ticks to the slowest partition #58003
Merged
Conversation
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
github-actions
bot
added
the
Scope: Backend
Codecov Report
@@ Coverage Diff @@
## master #58003 +/- ##
=======================================
Coverage 79.03% 79.03%
=======================================
Files 5130 5129 -1
Lines 223024 223059 +35
Branches 37559 37567 +8
=======================================
+ Hits 176256 176304 +48
+ Misses 41131 41120 -11
+ Partials 5637 5635 -2
|
evanpurkhiser
force-pushed
the
evanpurkhiser/feat-crons-limit-clock-ticks-to-the-slowest-partition
branch
from
a0ea82d to
d9f8f22
Compare
This function keeps a global clock driven by the monitor ingest topic to trigger the monitor tasks once per minute. This change updates this function to track the slowest partition within the topic. This will help to avoid the clock being moved forward when a single partition has a large number of check-ins read out of it (in a backlog situation), causing check-ins to be marked missed since they were not read before the clock ticked. This change does NOT yet use the slowest partition timestamp as the driver of the global clock, but simply logs the timestamp so we can validate in production that it is still accurately moving forward.
evanpurkhiser
force-pushed
the
evanpurkhiser/feat-crons-limit-clock-ticks-to-the-slowest-partition
branch
from
d9f8f22 to
bed3e48
Compare
wedamija
approved these changes
Oct 12, 2023
Comment on lines
+121
to
+126
| slowest_partitions = redis_client.zrange( | ||
| name=MONITOR_TASKS_PARTITION_CLOCKS, | ||
| withscores=True, | ||
| start=0, | ||
| end=0, | ||
| ) |
There was a problem hiding this comment.
Lol so weird that start/end both 0 is how you fetch one item
There was a problem hiding this comment.
yeah inclusive, starts with one and ends with one
evanpurkhiser
deleted the
evanpurkhiser/feat-crons-limit-clock-ticks-to-the-slowest-partition
branch
evanpurkhiser
added a commit
that referenced
this pull request
Oct 12, 2023
This is a follow up after GH-58003. The clock which dispatches tasks will now only tick forward once all partitions have been read up to to the synchronized time
evanpurkhiser
added a commit
that referenced
this pull request
Oct 13, 2023
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Part of #55821
This function keeps a global clock driven by the monitor ingest topic to trigger the monitor tasks once per minute.
This change updates this function to track the slowest partition within the topic. This will help to avoid the clock being moved forward when a single partition has a large number of check-ins read out of it (in a backlog situation), causing check-ins to be marked missed since they were not read before the clock ticked.
This change does NOT yet use the slowest partition timestamp as the driver of the global clock, but simply logs the timestamp so we can validate in production that it is still accurately moving forward.
In a future PR I will switch this function to use the slowest partition timestamp as the
reference_tsand add tests to validate this works