-
Notifications
You must be signed in to change notification settings - Fork 2.1k
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
VReplication: Support reversing read-only traffic in vtctldclient #16920
VReplication: Support reversing read-only traffic in vtctldclient #16920
Conversation
Signed-off-by: Matt Lord <[email protected]>
Review ChecklistHello reviewers! 👋 Please follow this checklist when reviewing this Pull Request. General
Tests
Documentation
New flags
If a workflow is added or modified:
Backward compatibility
|
1ab47a7
to
6557d28
Compare
Signed-off-by: Matt Lord <[email protected]>
6557d28
to
a48daf6
Compare
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #16920 +/- ##
==========================================
- Coverage 69.35% 67.15% -2.20%
==========================================
Files 1571 1571
Lines 204149 251785 +47636
==========================================
+ Hits 141583 169087 +27504
- Misses 62566 82698 +20132 ☔ View full report in Codecov by Sentry. |
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
So that we can see e.g. that we're doing a forward switch for the reverse workflow. Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Meaning that we continue to use 0 replica and 0 rdonly tablets. Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
…cancel Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
@rohit-nayak-ps thanks again for the help with the new tests! That was a life saver. I thought about what else to test, now that we have the new and updated e2e and unit tests and nothing obvious came to mind. If you have general suggestions please let me know and I'll work on adding/updating tests for those conditions. In the meantime I'll think about what gaps might still be left myself (haven't come up with anything yet). |
Signed-off-by: Matt Lord <[email protected]>
d8e959d
to
64f8799
Compare
Signed-off-by: Matt Lord <[email protected]>
64f8799
to
8f241e8
Compare
…cancel Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
… checked for the routed keyspace. Also it not actually failing if route was incorrect. Signed-off-by: Rohit Nayak <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Description
This PR fixes a bug in
vtctldclient
, in that it did not support a needed feature that the legacyvtctlclient
did (see the client transition docs). Specifically, support for usingReverseTraffic
when traffic for read-only tablet types (REPLICA, RDONLY) have been switched BUT writes (PRIMARY tablet type) have not ever been before.h/t to @rohit-nayak-ps for the additional tests that were a huge help! 🙇
Using the manual test from below, this fails with
vtctldclient
but works withvtctlclient
:I think that it makes sense to backport this at least to v21 so that this is addressed in a minimum of 1 GA branches/releases. For those on older releases, the workaround is to use the equivalent
vtctlclient
command. In the end, what this really did is closer align the two implementations around how this work is done.Manual test
With the client command output on the PR branch being:
Related Issue(s)
ReverseTraffic
for onlyreplica
andrdonly
tablet types #16918Checklist