List:General Discussion« Previous MessageNext Message »
From:Prabhat Kumar Date:November 10 2011 5:13pm
Subject:Re: Relication Issues
View as plain text  
Hi,
It could be a hardware issue, as if you do not use RAID and you have bad
sectors or the likes. But the most common cause could be a server crash.

The INSERT or transaction was partially completed when server crash. When
the slave is then restarted, it will try to execute the
transaction/statement again *from the beginning*, and will fail since some
of the rows were already added to the table. The situation occurs also if
you are using MyISAM tables and a statement outside a transaction is
interrupted.

Thanks,
On Thu, Nov 10, 2011 at 6:08 AM, Steven Staples <sstaples@stripped> wrote:

> My Apologies,
>
> The tables are all MyISAM, most of the inserts/updates/deletes are done
> through stored procedures on the master.  There are about 2 stored
> procedure
> calls per second, consisting of a whole bunch of queries, updates, and
> inserts within them.
>
> The latest issue(s) we're having, are:
>
> Could not execute Write_rows event on table xxx.xxxxx; Duplicate entry
> '201111-00001016792' for key 'PRIMARY', Error_code: 1062; handler error
> HA_ERR_FOUND_DUPP_KEY; the event's mas
>
> Could not execute Write_rows event on table xxx.xxxxx; Duplicate entry
> '44870420' for key 'PRIMARY', Er
>
> Update_rows event on table xxx.xxxxx; Can't find record in 'xxxxx',
> Error_code: 1032; handler error HA_ERR_KEY
>
> (Where xxx is the databse, and xxxxx is the table)
>
> The replication broke yesterday morning for some reason, and the primary
> key
> index was at around 900,000.  When we found out it was down, and ended up
> getting to work on it, we tried to restart it and the next ID it was
> inserting for was 1,016,792.   So we're missing about 100,000 enteries, and
> they are not in the bin log (as far as I can see).
>
>
> Steve.
>
> > -----Original Message-----
> > From: Reindl Harald [mailto:h.reindl@stripped]
> > Sent: November 10, 2011 8:42 AM
> > To: mysql@stripped
> > Subject: Re: Relication Issues
> >
> > replication is buggy sometimes
> >
> > but without any outputs desribing your problem nobody can really help you
> -
> > memory tables as example are making much more troubles as myisam
> >
> > Am 10.11.2011 14:26, schrieb Steven Staples:
> > > Good morning list,
> > >
> > > I am having issues with my replication setup. it seems that every few
> > > weeks/months something happens, and I have to restart it, skip a row,
> > > or delete, rsync and restart replication from scratch.
> > >
> > > The databases that are being replicated, are rather large... there are
> > > about
> > > 12 new tables every year, each consisting of about 2-4g in size each
> > > (today's total database size is 83g).
> > >
> > > Again, things work fine for a while, and then there are issues.  The
> > > backup/slave is on the same network/datacenter, so network speed isn't
> an
> > > issue.  I just don't understand why inconsistencies keep arising.   Is
> > there
> > > a better way to do live backups, or have a hot space in the event of a
> > > catastrophe?   Is there 3rd party software that would better achieve
> data
> > > integrity or something?
> > >
> > > Any help here would be appreciated.
>
>
>
>  --
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe:
> http://lists.mysql.com/mysql?unsub=1
>
>


-- 
Best Regards,

Prabhat Kumar
MySQL DBA

My Blog: http://adminlinux.blogspot.com
My LinkedIn: http://www.linkedin.com/in/profileprabhat

Thread
Relication IssuesSteven Staples10 Nov
  • Re: Relication IssuesReindl Harald10 Nov
    • RE: Relication IssuesSteven Staples10 Nov
      • Re: Relication IssuesPrabhat Kumar10 Nov