-
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
set vreplication net read and net write timeout session vars to high values #14203
set vreplication net read and net write timeout session vars to high values #14203
Conversation
…values Signed-off-by: Olga Shestopalova <[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
|
Signed-off-by: Olga Shestopalova <[email protected]>
@@ -60,6 +60,8 @@ const ( | |||
getAutoIncrementStep = "select @@session.auto_increment_increment" | |||
setSessionTZ = "set @@session.time_zone = '+00:00'" | |||
setNames = "set names 'binary'" | |||
setNetReadTimeout = "set @@session.net_read_timeout = 300" |
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.
We were discussing this yesterday and the consensus was that we should not hardcode values since the actual value might vary based on the cluster environment.
We can add flags for this as common vttablet flags similar to the query timeout PR. Your values seem fine as defaults to me.
Signed-off-by: Olga Shestopalova <[email protected]>
Signed-off-by: Olga Shestopalova <[email protected]>
Signed-off-by: Olga Shestopalova <[email protected]>
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
…values (vitessio#14203) Signed-off-by: Olga Shestopalova <[email protected]> Co-authored-by: Olga Shestopalova <[email protected]>
…values (vitessio#14203) Signed-off-by: Olga Shestopalova <[email protected]> Co-authored-by: Olga Shestopalova <[email protected]>
Description
On some particularly busy tables, we cannot run vitess onlineddl schema changes because it fails with
unexpected EOF\nio.ReadFull(packet body of length 287) failed (errno 2013) (sqlstate HY000)
. Note that ghost also fails on these tables, but we have higher hopes for vitess.In several slack threads, the recommendation was to increase
net_read_timeout
(default is 30, recommended for vreplication is 300) andnet_write_timeout
(default is 60, recommended for vreplication is 600). However, we don't want to modify these values globally, would be nice if only vreplication's session had these higher values - and this PR does just thatUpdate: with these changes, we were able to successfully run onlineddl schema changes on an incredibly busy table that was previously unable to run any kind of migrations
Related Issue(s)
Fixes #14202
Checklist
Deployment Notes