On Wed, Mar 19, 2003 at 04:34:32PM -0600, Jeremy Sanders wrote:
> We remounted the imap and postfix partitions as ext2 and it is smoking.
> Now we just have to figure out how to tweak ext3 to work better. I think
> a bigger journal size might help, it is flushing to disk too often and
> the pop3d p
On Wed, 19 Mar 2003, Jeremy Sanders wrote:
> We remounted the imap and postfix partitions as ext2 and it is smoking.
Switch to XFS. That will fix your performance troubles, probably...
And mount the spools "noatime".
--
"One disk to rule them all, One disk to find them. One disk to bring
th
We remounted the imap and postfix partitions as ext2 and it is smoking.
Now we just have to figure out how to tweak ext3 to work better. I think
a bigger journal size might help, it is flushing to disk too often and
the pop3d processes are piling up waiting on it to flush...
Jeremy
>>> Henrique d
On Wed, 19 Mar 2003, Jeremy Sanders wrote:
> changed to skiplist. It's not all pop3, but it is mostly pop3 in the D
> state. There are some lmtpd, cleanup(postfix), and every once in a while
> kjournald is in DW. When the load really spikes like 30 pop3d processes
Looks like LDAP is hosing your sy
> was consistently below 5. The processor is mostly idle. I thought
the
Find out which processes are in D or R state. Are them all pop3d ?
What
processes are contending for the CPU during the CS spikes?
What database backends are you using in Cyrus? What berkeley DB
version?
The db is db3-3.3
On Mon, 17 Mar 2003, Jeremy Sanders wrote:
> Why would pop3d have all the hot backup db files open? If 50 pop3d
There are no hot-backup files in your lsof output. Just the usual db log
files. Run the db statistics stuff and see if you can find trouble spots,
such as too many lockers, or something
Why would pop3d have all the hot backup db files open? If 50 pop3d
processes are trying to access the same 4 files, I can see why there
would be some contention there
pop3d 1423 cyrus memREG 58,9 270336324606
/var/spool/imap/db/__db.002
pop3d 1423 cyrus memREG
Hello,
I'm running Cyrus IMAP 2.1.12 on a Redhat 7.3 box running kernel
2.4.20. The imap partition is on a Compaq RA4100 with a Compaq
Fiber-channel HBA in the server. It is an lvm ext3 partition running w/
noatime,data=ordered.
Here is a vmstat output, notice the spike in context switches:
[EMA
