All MySQL DBs limited to max 3 concurrent/idle connections #15 #931
Conversation
Actions/Workflow: upload artifact
Support a complete ALTER TABLE statement in --alter
handle driver timeout error
Updates from upstream 2020 10
Initial commit: towards setting up a test suite Signed-off-by: Shlomi Noach <[email protected]>
…original table, applying AUTO_INCREMENT value onto ghost table if applicable and user has not specified AUTO_INCREMENT in alter statement
Signed-off-by: Shlomi Noach <[email protected]>
Signed-off-by: Shlomi Noach <[email protected]>
Copying AUTO_INCREMENT value to ghost table
Signed-off-by: Shlomi Noach <[email protected]>
Signed-off-by: Shlomi Noach <[email protected]>
Generated column as part of UNIQUE (or PRIMARY) KEY
Cut-over should wait for heartbeat lag to be low enough to succeed
Signed-off-by: Shlomi Noach <[email protected]>
|
@shlomi-noach 👋 that's a good idea, thanks for the PR 👍 I'm seeing potentially unrelated changes vs openark#15 in this PR. Can you confirm? |
@timvaillancourt yeah, that's due to #911 and #919 not merged yet, I've continued on top of these. I can sanitize the relevant changes to this PR and re-submit (but hoping it would only be a matter of tim etill #911 and #919 are merged). |
@shlomi-noach ahh thanks. No problem, I can merge those PRs together eventually 👍 |
|
branch updated after recent merge of downstream contribution. |
|
This low-risk changed has passed many tests, merging it 👍 Thanks @shlomi-noach 🎉 |
Resubmission of openark#15 from downstream.
Issue reported privately by community user. In some scenario (user had semi-sync + long row data)
gh-ostconnections piled up toMAX_USER_CONNECTIONS. This PR puts a hard limit (constant3) on all pools. There's several of those (inspector, applier, throttler, and more), but each is limited to to 3 connections which at least does not cause a connection pileup on the MySQL server(s).script/cibuildreturns with no formatting errors, build errors or unit test errors.