xfs
[Top] [All Lists]

Re: Delaylog

To: xfs@xxxxxxxxxxx
Subject: Re: Delaylog
From: Fabricio Archanjo <farchanjo@xxxxxxxxx>
Date: Tue, 14 Sep 2010 18:32:35 -0300
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:content-type :content-transfer-encoding; bh=Xx31hdiK3wzPkQFLL09TQkJ8ZMSCU9sXTywBHINEDTg=; b=HL3LYZ2OmNR5U1cQ2kErRSwKtnhljuPtqYkJcSZN4JJA+fDGIeuNFahc7V2VQrh2op QoM5sxe7ZsXNMOm7xmmUvNUp5PHP8rPp54byqCfC4N8tl7YrphFi2sYMOi6ubHa2XeXz AxuMYAe/hy3V/xs0ZOx+gI0GlIE2Hr08Blaq8=
Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type:content-transfer-encoding; b=e58HaJHn7T1cJYu9tx/lxyhxYyfLzIsMFN5MPW949Qt4EolP7/GmvFSvxR5RPOmQcm A6Cx7VGiqOFIL1vkgv2wJkV2DiBZCFKXmF8oPvgRxQrm7z2rVmHOUwROCtXz3FAPGc5I 9Up5bm1w5jN6r67+JTslk8APF2HeTdlLCyM+M=
In-reply-to: <4C8FD50C.1030905@xxxxxxxxxxxxxxxxx>
References: <AANLkTin6FSBK3tMXhSxDuLdjyb+R0LhNnaa2CKt37FcV@xxxxxxxxxxxxxx> <201009142106.24448.arekm@xxxxxxxx> <4C8FD50C.1030905@xxxxxxxxxxxxxxxxx>
When i changed to delaylog option my database performance increased,
so how i have 2 databases one master another slave, i can keep it. If
i have problem on my filesystem, i'll need to changed the slave to
master and recover my master.


Thanks all.

On Tue, Sep 14, 2010 at 5:03 PM, Stan Hoeppner <stan@xxxxxxxxxxxxxxxxx> wrote:
> Arkadiusz Miskiewicz put forth on 9/14/2010 2:06 PM:
>> On Tuesday 14 of September 2010, Fabricio Archanjo wrote:
>>> Hey all,
>>>
>>>    I just trying delaylog in my server that has a mysql database. When
>>> i monted my /var/lib/mysql with delaylog option, it showed me:
>>> "Enabling EXPERIMENTAL delayed logging feature - use at your own
>>> risk". Ok, i know it's experimental, but what kind of problem could i
>>> have using delaylog?
>>
>> ... and what problems in case of system hang or power loss when compared to
>> nodelaylog mode?
>
> This was covered in prior posts IIRC.  Delaylog holds more write
> transactions in memory in an effort to decrease the amount of disk I/O
> and optimize write patterns.  The more blocks waiting in the in memory
> log, the more data will be lost due to power outage, controller/disk
> failure, storage HBA/network failure (iSCSI/FC), kernel panics, etc.
>
> Same failure modes as before, but with potentially greater loss of
> data--unless there is an undiscovered bug that can wreck the entire
> filesystem.  ;)  Which I believe is the reason for the "experimental"
> boilerplate.
>
> --
> Stan
>
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs
>

<Prev in Thread] Current Thread [Next in Thread>