-
Notifications
You must be signed in to change notification settings - Fork 726
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
pd client panics with "[tso] timestamp fallback" #8836
Comments
maybe #8781? Depending on the version of the pd server, it is possible to happen. |
First time became leader
But meet Quickly became follower
The second time it becomes a leader, but it doesn't immediately synchronize tso
|
After discussion with @rleungx , we should not move So we should call |
It happens again: https://tcms.pingcap.net/dashboard/executions/case/12789276
|
At 23:49:44.543, we initiated tso, but the leader failed to create the raft cluster which resulted in tso not being reset.
At 23:51:24.234, it becomes leader and skips sync timestamp from etcd. TiDB report tso fallback
|
tracked by #8889 |
Bug Report
Test: https://tcms.pingcap.net/dashboard/executions/plan/7527309
The issue was found by @SpadeA-Tang, thank you!
What did you do?
Run jepsen tests.
What did you expect to see?
No panic.
What version of PD are you using (
pd-server -V
)?TiDB: pingcap/tidb@5448f9f
PD: 60b960a
The text was updated successfully, but these errors were encountered: