ROW Level replication slow query logging

[Peter Zaitsev]

When Row level replication is enabled with log_slave_slow_statements only BEGIN statements are logged.

Suggestion: See how we can log it as a pseudo query. Something like WRITEROW <database>.<tablename> would be enough for sake of analyses.

[Baron Schwartz (baron-xaprb)]

mysqlbinlog has functionality to print out something from row-based events that looks query-ish. Maybe we can reuse that.

[Swany (greenlion)]

For consistency, can you please write it in the same format as:
mysqlbinlog -v --base64-output=decode-rows

That will make parsing the slow log with tools that can parse the
output of mysqlbinlog can read the slow log output without too much
additional work.

On Fri, Nov 5, 2010 at 11:27 AM, Oleg Tsarev <email address hidden> wrote:
> ** Changed in: percona-server
>       Status: Triaged => In Progress
>
> --
> ROW Level replication slow query logging
> https://bugs.launchpad.net/bugs/625022
> You received this bug notification because you are a member of Percona
> developers, which is the registrant for Percona Server.
>

[Peter Zaitsev (pz-percona)]

I would like to make sure this task is given proper priority.

Without being able to log row level replication traffic we can't analyze the complete workload on slave servers in many installations which is serious problem.

[Stewart Smith (stewart)]

For proper prioritization I think we should create a blueprint for it and better schedule work for it.

This will get done in my next phase of going through Percona Server related things (getting blueprints and milestones/series going).

[Laurynas Biveinis (laurynas-biveinis)]

binlog_rows_query_log_events could be used here.

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-2326