Re: Performance of disk i/o with 3ware

On 11/6/07, Nico -telmich- Schottelius <
nico-freebsd-performance@xxxxxxxxxxxxxxx> wrote:

Manjunath R Gowda [Tue, Nov 06, 2007 at 09:54:09AM -0800]:
On 11/6/07, Nico -telmich- Schottelius <
nico-freebsd-performance@xxxxxxxxxxxxxxx> wrote:

Hello!

I've the problem that sometimes there are many disk waiting processes
(sysctl -n vm.vmtotal), but systat -vmstat shows da0 and da1 busy with
0-10%.

I woudn't worry about how many are sleeping. But, How long they sleep
would
be interesting data.

And also why they sleep. Having about 700 processes in sleep state plus
400 processes in diskwait state makes me wonder somehow.



At a given time only few can run, remaining once will sleep which is an
expected behavior.

There are also some processes from cron, which are older and also in
state "I" (as shown by ps).


You have to dig deep and see what those cron jobs do.

I guess that the disk i/o is at about 100%, but wondering why I see
those strange values.

Why do you think it should be 100%? Are you running any disk I/O intense
application?

Because it's the well known bottleneck in that server:

- it has 4 cpus, which are about 30-50% idle
- it has 4 GiB ram, of which 2GiB is mostly inactive
- it has 4 10k rpm hds in 2 raid1 disk arrays, one for
the mailboxes + root and one for the qmail-queue

Normally systat -vmstat shows 80-100% busy state on the disk arrays,
but currently it's at about 10%, which cannot be right.

I think I'll have a look closer look at the patch we used from CVS to
patch the
3ware twa driver.


Looks like your blaming the twa driver based on your experience. OS is a
complex piece of software and involves many components, scheduler, vm,
scsi-stack, driver, problem could be any where in this. That's why specific
data would be useful. Something like on 6.2 running x takes y and on
6.2+patch it takes z would be useful.

Anyone an idea,
a) why systat -vmstat shows so small busy values?
b) how to debug it further?

You can try experimenting with diffrent I/O loads to make sure that
there
is a problem before start debugging it.

There is a big problem, that even results in taking about 30 seconds
until the '220' messages comes from qmail when connecting via telnet
from outside to it.



Your adding one more variable, network here. Verify that when you did
cvsup the network driver was not updated.

To summarise, what runs on the server:

- qmail + patches => has /var/qmail/queue on it's own disk-array
- courier imapd
- vpopmail (pop3 auth)
- mysql (needed by vpopmail)
- clamd
- spamassassin
- Webmail (horde on apache)

So it's mainly a customer mailserver, with nothing special installed,
which may have some load, but not in the way it's not explainable.

So I'm still hoping somebody with similar problems reads this mail ;-)

Sincerly

Nico

--
Think about Free and Open Source Software (FOSS).
http://nico.schottelius.org/documentations/foss/the-term-foss/

PGP: BFE4 C736 ABE5 406F 8F42 F7CF B8BE F92A 9885 188C

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFHMMFxuL75KpiFGIwRAoKxAKDjh7aN8n/b7B3YPqcdKuqMPv9CGACgsGuf
lI6wV4jUuP8GnUhd37oW/mQ=
=4Yn4
-----END PGP SIGNATURE-----


_______________________________________________
freebsd-performance@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "freebsd-performance-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • Re: Old Mac Mini - what system to upgrade to?
    ... Which Mini model, by the way? ... to sleep, and its time is the point at which the sleep sequence ... to reinitialize itself because the network environment has changed. ... I can imagine three causes for a hangup: it can't access the information on the disk, that information needs updating, or lookupd has crashed. ...
    (comp.sys.mac.system)
  • Re: Performance of disk i/o with 3ware
    ... I've the problem that sometimes there are many disk waiting processes ... And also why they sleep. ... You can try experimenting with diffrent I/O loads to make sure that there ... mysql (needed by vpopmail) ...
    (freebsd-performance)
  • Re: Old Mac Mini - what system to upgrade to?
    ... into a hibernation mode when put to sleep (sometimes referred to as ... loaded back from disk, otherwise the wakeup time is unaffected. ... My Mac Mini appears to default to not using this mode. ... Hibernate mode of 0 means normal "instant sleep" with no writing to hard ...
    (comp.sys.mac.system)
  • Re: LaCie disks
    ... disk is not turned off when the computer is turned off. ... as long as their other drives that do shut down at computer shut down. ... I have asked LaCie for their comment. ... I had no problem at all with my G4 who put them to sleep, ...
    (comp.sys.mac.apps)