Skip to:

  1. Skip to Jira Navigation
  2. Skip to Side Navigation
  3. Skip to Main Content

"no such a transition COMMITTING -> ROLLED_BACK" with LOCK BINLOG FOR BACKUP after fix for PXC-2103

Description

It looks like fix for PXC-2103 is not full. For example, if acquiring backup lock protection fails for any other reason than conflicting transaction instance still can crash with [ERROR] WSREP: FSM: no such a transition COMMITTING -> ROLLED_BACK

How to Repeat:

Run attached test case in galera suite:

Test case is pretty simple:

Suggested Fix:

Don't crash, make proper error handling for all situations when commit may fail.

Environment

None

AFFECTED CS IDs

239702

Attachments

1

Smart Checklist

Activity

Julia Vural March 4, 2025 at 9:29 PM

It appears that this issue is no longer being worked on, so we are closing it for housekeeping purposes. If you believe the issue still exists, please open a new ticket after confirming it's present in the latest release.

Sveta Smirnova December 5, 2018 at 3:04 PM

Same effect happens if a connection gets disconnected forcefully by, say, application or buggy connetion pool. Error is the same: timeout. Reason is this code in sql/mdl.cc

Maybe would be possible to acquire backup lock before wsrep sends data to remote nodes to avoid this crash?

Won't Do

Details

Assignee

Reporter

Time tracking

1d 4h logged

Affects versions

Priority

Smart Checklist

Created November 22, 2018 at 10:09 PM
Updated March 4, 2025 at 9:29 PM
Resolved March 4, 2025 at 9:29 PM