[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: how to counteract slowdown



Hi

--On Monday, November 12, 2001 02:21:36 PM -0800 Mike Fedyk <mfedyk matchmail com> wrote:

On Mon, Nov 12, 2001 at 04:18:59PM -0000, Patrick C. F. Ernzer wrote:
Hello,

as far as I understood ext3 will more or less hog a machine when writing
away the journal. A customer is having a slowdown every 5 minues for
about  30 seconds, the machine becomes more or less unusable.

This is an NFS server serving 300 Gigs spread over 2 NFS shares.

I'm wondering what would be the best course of action:
 a) make the journal bigger?
 b) make the journal smaller?
 c) switch from ordered to writeback?

Can somebody give me a hint?


What was the previous FS? Ext2?

Previous FS was ext2.

What is the current mode? The default (ordered)?

Yes. It has been switched (see below) 20 hours ago and there has been no performance hit since.

What are the write patterns for this FS? Large contiguous writes? random writes? Read, write, read? (I'm thinking of each individual use, and overall usage)

Write patterns are mixed, current ext3 FS are:


/dev/sda12 on /tmp type ext3 (rw,data=writeback)
/dev/sda13 on /usr/local type ext3 (rw,data=journal)
/dev/sda14 on /var type ext3 (rw,data=journal)
/dev/sda15 on /var/opt/openmail type ext3 (rw,data=journal)
/dev/sdb5 on /home type ext3 (rw,data=writeback)
/dev/sdc1 on /home/ftp/pub/Archive type ext3 (rw,data=writeback)


they are exported as follows


/tmp    *.think(rw,root_squash) 192.168.10.*(rw,root_squash)
/home/netstation   192.168.10.155(rw,root_squash,anonuid=555,anongid=555)
/home   172.16.3.15(rw,no_root_squash) 192.168.10.16(rw,no_root_squash)
192.168.10.15(rw,no_root_squash)
virtual-039.think(rw,root_squash,anonuid=14,anongid=50)
*.think(rw,root_squash)  192.168.10.*(rw,root_squash)
/home/ftp/pub/Archive   172.16.3.15(rw,no_root_squash)
192.168.10.16(rw,no_root_squash) 192.168.10.15(rw,no_root_squash)
192.168.10.176(rw,no_root_squash)
virtual-039.think(rw,root_squash,anonuid=14,anongid=50)


uname -a


Linux jake.think 2.4.7-10enterprise #1 SMP Thu Sep 6 16:48:20 EDT 2001
i686 unknown
(can not go to 2.4.9 due to PCI issues, waiting for a newer RH kernel that fixes bug 55470)



Distribution is: Red Hat Linux release 7.1 (Seawolf)


--On Monday, November 12, 2001 02:27:13 PM -0800 Andrew Morton <akpm zip com au> wrote:

Are you able to monitor the server during the slowdown?  Running
`top', `ps' and `vmstat 1' would be useful.

This will be difficult as: * the journal mode has been switched in the mean time * it's a production machine and switching back and forth is not really an option.

RU

PCFE
--
from the desk of Patrick C. F. Ernzer
Red Hat Europe

Attachment: pgp00006.pgp
Description: PGP signature


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]