Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 05 Aug 2006 20:31:14 +0200
From:      Miroslav Lachman <000.fbsd@quip.cz>
To:        rick-freebsd@kiwi-computer.com
Cc:        freebsd-geom@freebsd.org
Subject:   Re: gmirror Cannot add disk ad5 to gm0 (error=22)
Message-ID:  <44D4E3F2.5060709@quip.cz>
In-Reply-To: <20060803171025.GA23405@megan.kiwi-computer.com>
References:  <44D06650.1030803@quip.cz> <20060802183001.GA14279@megan.kiwi-computer.com> <44D10D1D.9040700@quip.cz> <20060802210709.GA15310@megan.kiwi-computer.com> <44D126EF.9070503@quip.cz> <44D12A80.9040802@quip.cz> <20060802233255.GB16385@megan.kiwi-computer.com> <44D1BE0B.9090709@quip.cz> <20060803171025.GA23405@megan.kiwi-computer.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Rick C. Petty wrote:
> On Thu, Aug 03, 2006 at 11:12:43AM +0200, Miroslav Lachman wrote:

[...]

> This certainly sounds like a disk-related problem.  Likely your previous
> failures were due to the same problems.  Time to send the disks back to the
> manufacturer for replacement..  :-/

Disk was replaced with new one and synchronization was much slower then 
usual (more than 5 hours instead of 1:30)

If I run gstat, disks are idle (0% busy), then I star dd command to 
write some test file to disk and see that dd writing is about 10 times 
slower then on another "single disk" system.

root@track =  2x 250GB Seagate SATA disks in gmirror + P4 3GHz + 1GB RAM
root@grimm2 = 1x 300GB Samsung ATA disk + AMD Burton 2500 + 256MB RAM
both machines has 6.1-RELEASE and similar disk partitions

root@track /vol0/# dd if=/dev/zero of=/vol0/test bs=1m count=10
10+0 records in
10+0 records out
10485760 bytes transferred in 1.639139 secs (6397114 bytes/sec)

root@track /vol0/# dd if=/dev/zero of=/vol0/test bs=1m count=128
128+0 records in
128+0 records out
134217728 bytes transferred in 24.923024 secs (5385291 bytes/sec)

root@track /vol0/# dd if=/dev/zero of=/vol0/test bs=1m count=1024
1024+0 records in
1024+0 records out
1073741824 bytes transferred in 190.751349 secs (5629013 bytes/sec)


root@grimm2 ~/# dd if=/dev/zero of=/vol0/test bs=1m count=10
10+0 records in
10+0 records out
10485760 bytes transferred in 0.173180 secs (60548295 bytes/sec)

root@grimm2 ~/# dd if=/dev/zero of=/vol0/test bs=1m count=128
128+0 records in
128+0 records out
134217728 bytes transferred in 2.479431 secs (54132468 bytes/sec)

root@grimm2 ~/# dd if=/dev/zero of=/vol0/test bs=1m count=1024
1024+0 records in
1024+0 records out
1073741824 bytes transferred in 21.136426 secs (50800539 bytes/sec)

grimm2 is almost 10 times faster!
Is it expected, that gmirror is 10 time slower? I don't think so.

Does anybody know how can I "debug" this slowness?

>>After manual reboot, there is no ad5 device. I hope new drive helps, but 
>>I am still nervous, because I have similar troubles with 2 machines 
>>(both replaced with new one - so I played with 4 machines)...
> 
> 
> Chance of one "new" disk being bad-- pretty low.
> Chance of two new disks being bad-- even lower.
> Chance of three or more disks going bad around the same time-- much higher.
> 
> I've noticed this type of behavior before.  Someone correct me if I'm
> wrong but it appears that you probably got a bad batch of disks.  Try
> throwing a different set of disks on the boxes (preferrably a different
> manufacturer).  I would also try swapping with brand new high-quality
> cables (just because they're cheaper than new disks).
> 
> -- Rick C. Petty

Thanks for your time and help

If anything can goes bad, then everything will go bad :(

(I am sorry if my english is bad, I think you know what I mean)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44D4E3F2.5060709>