[LRUG] Sites drop off web

Matthew Rudy Jacobs matthewrudyjacobs at gmail.com
Mon Aug 8 04:05:53 PDT 2011


This may be overkill for you,
But it may kill two birds with one stone.

- set up a replication to a single slave
- when you want to dump, stop replication
- dump from the slave
- restart after the dump

Thereby you have a hot copy to failover to
As well as a mechanism for taking consistent database snapshots
Without affecting the performance of your service.

On Monday, 8 August 2011, Jay Caines-Gooby <jay at gooby.org> wrote:
> On 8 August 2011 11:53, Andrew Stewart <boss at airbladesoftware.com> wrote:
>> Tim,
>>
>>> I had a similar problem with CentOS. mysqldump was stalling everything
for a couple of minutes when it ran.
>>>
>>> Saw some pretty epic query times popping up in NewRelic RPM around that
time. The site just stalled for a few minutes.
>>>
>>> Pingdom reported a minute or two of downtime.
>>
>> Thanks for the information.  I use mysqldump too -- with increasing
reluctance because I've been seeing the
>> same behaviour as you.  As a stopgap measure I wrapped it in a timeout[1]
but I think the real solution is to
>> move to Postgres and whatever backup mechanism it has.
>
> You can also use the --single-transaction option to mysqldump which
> prevents it from locking up your app.
>
>
> --
> Jay Caines-Gooby
> http://jay.gooby.org
> jay at gooby.org
> +44 (0)7956 182625
> twitter, skype & aim: jaygooby
> gtalk: jaygooby at gmail.com
> _______________________________________________
> Chat mailing list
> Chat at lists.lrug.org
> http://lists.lrug.org/listinfo.cgi/chat-lrug.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lrug.org/pipermail/chat-lrug.org/attachments/20110808/57ad63b8/attachment.html>


More information about the Chat mailing list