Bug #3240 MySQL process manager sometimes fails due to binding problem

• Submitted: 19 Mar 2004 1:44
• Modified: 25 Mar 2004 10:40
• Reporter: Marek Les
• Status: Can't repeat
• Category: MySQL Server
• Severity: S2 (Serious)
• Version: 4.0.18-standard
• OS: Linux (Linux)
• CPU Architecture: Any

[19 Mar 2004 1:44] Marek Les

Description:
When MySQL decides to restart itself, it sometimes doesn't wait for unbinding itself from the 3306 port. It goes like this:

Number of processes running now: 1
mysqld process hanging, pid 22313 - killed
040318 22:35:12  mysqld restarted
040318 22:35:12  Can't start server: Bind on TCP/IP port: Address already in use
040318 22:35:12  Do you already have another mysqld server running on port: 3306 ?
040318 22:35:12  Aborting

040318 22:35:12  /www/mysql/bin/mysqld: Shutdown Complete

How to repeat:
Don't know.

Suggested fix:
I think it's rather easy to implement some kind of self-check or (configurable?) delay.

[25 Mar 2004 10:40] MySQL Verification Team

Thank you for writting this report.

This bug report forum is only for repeatable bug reports. For bug
reports like this one you should use the MySQL mailing list or buy
commercial support from MySQL AB.  

What we need is fully repeatable test case.

This test case should consist of the series of SQL commands or actaions that will always lead to the problem, in your case to the crash.