| List: | Commits | « Previous MessageNext Message » | |
| From: | Davi Arnaut | Date: | November 27 2009 10:54am |
| Subject: | Re: bzr commit into mysql-5.1-bugteam branch (Dao-Gang.Qu:3192) Bug#42851 | ||
| View as plain text | |||
Hi He Zhenxing, On 11/27/09 2:04 AM, He Zhenxing wrote: > Hi Davi, > > Davi Arnaut wrote: >> Hi All, >> >> I think you guys need to stop for a moment, take a deep breath and >> reflect about what problem needs to be solved. The problem is that a >> user might trigger messages to be written to the error log, which is a >> concern since disk space for the error log is limited. The concern is >> about messages written to the error log.. any kind (be it a warning, a >> error or a note). This has nothing to do with push_warning nor it is >> exclusive to replication code -- the case reported on the bug is just >> one of the various that a user can trigger. >> > > The bug#42851 is complaining about lots of "Statement is not safe to > log..." warnings in the error log. And after checking the code, the > binlog unsafe warnings are the only source that might trigger this > problem. and this are the only warnings that goes both to SHOW WARNINGS Yes, but as I already explained multiple times by now, this is fixed! > and to error log. I think there was a misunderstanding of the > --log-warnings option when implementing this in replication code. No. The log_warnings "protection" was introduced in Bug#46265. > --log-warnings does not mean that warnings that go SHOW WARNINGS should > also go error log. It means that some extra warnings (connection > closing/aborting, threads quiting, slave reconnection, etc) will be > written into error log. One has nothing to do with the another. Please see mail from Sergei G. (sent to dev-public with subject "Re: BUG#42851: How to treat warnings?) >> Please, let's not complicate this further. >> > > On the contrary, I think we are making the thing simple, just remove the > code that writing the unsafe warnings to error log will solve the > problem. Unsafe warnings can be seen by SHOW WARNINGS, we do not need to > write it to the error log. > No. This is a more general problem, please see the mail by Sergei G. -- Regards, Davi Arnaut
