I have a problem with dovecot eating sometimes up to 90% CPU.
While making such high cpu load, dovecot calls gettimeofday() more than
2 times per second:
mail2# cat k.dov.l.txt | grep CALL | grep gettimeofday | wc -l
1195552
mail2# head -n 1 k.dov.l.txt
20910 imap 0.00 CALL gettimeofda
Pete French пишет:
I have a problem with dovecot eating sometimes up to 90% CPU.
Are you using kqueue support ? I also had this problem and the solution
for me was to recompile it without kqueue support. Now it runs fine.
-pete.
Yes, I'm using kqueue support.
--
С уважением, Савчук Т
Daniel Gerzo пишет:
Hello Pete,
Friday, September 22, 2006, 12:15:29 PM, you wrote:
Yes, I'm using kqueue support.
Try taking it out and the problem should go away.
this is indeed a nice workaround (at least if it helps :-)), but
somebody(tm) should have a look why the k
Pete French пишет:
this is indeed a nice workaround (at least if it helps :-)), but
somebody(tm) should have a look why the kqueue support is broken
there in the first place.
I think it's a problem in Dovecot, not FreeBSD - there are a number of
messages regarding this on the Dovecot lists
ilya пишет:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hello BSD Gurus!
I was wondering if someone could suggest a great piece of FREEBSD 6.2
ported software, which can be effectively used as project management
tool, tasks, calendar, etc. We would like it to run on our server.
Therefore, we
I tried to use pam_group to allow accessing imap(dovecot) only for users
in certain group (users/groups stored in AD and checked out via
LDAP/Kerberos), but pam_group is checking applicant's group membership.
I'm sure, that in many cases is more useful to check group membership of
target (authe
I tried to use pam_group to grant access to imap(dovecot) only for users
in certain group (users/groups stored in AD and checked out via
LDAP/Kerberos), but pam_group is checking applicant's group membership.
I'm sure, that in many cases is more useful to check group membership of
target (authe
My SATA HDD with UFS2 crashed. While checking HDD fsck said, that alternate
super block at block 32 is not present. In 'man fsck' I saw, that in UFS2
(my file system) alternate super block is usually located in block 160 (For
UFS1 - in 32). So the question is: why fsck trying to find alternate
supe
On 11/3/05, Xin LI <[EMAIL PROTECTED]> wrote:
On 11/3/05, Taras Savchuk <[EMAIL PROTECTED]> wrote:
> My SATA HDD with UFS2 crashed. While checking HDD fsck said, that alternate
> super block at block 32 is not present. In 'man fsck' I saw, that in UFS2
> (
