Does anybody know if Cyrus has to be down before I can safely run
ctl_cyrusdb -r under truss? If so, I'll have to do it off-peak.
I did manage to get a couple minutes of truss output the other day but I
had to kill it before it finished. It looked very repetitive, with a
lot of this:
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
fstat(4, 0xFFBFE600) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
fstat(4, 0xFFBFE600) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
fstat(4, 0xFFBFE600) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
fstat(4, 0xFFBFE600) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
fstat(4, 0xFFBFE600) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE578) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE5F0) = 0
fcntl(4, F_SETLKW, 0xFFBFE8C0) = 0
fstat(4, 0xFFBFE9E8) = 0
stat("/var/imap/mailboxes.db", 0xFFBFE960) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "A1028B\r s k i p l i s t".., 2919372) = 2919372
fcntl(4, F_SETLKW, 0xFFBFE9D8) = 0
fcntl(4, F_SETLKW, 0xFFBFE4D8) = 0
> -----Original Message-----
> From: Andrew Morgan [mailto:[EMAIL PROTECTED]
> Sent: Monday, March 13, 2006 10:58 AM
> To: Rosenbaum, Larry M.
> Cc: info-cyrus@lists.andrew.cmu.edu
> Subject: Re: Any ideas? Cyrus takes 5 minutes to start
>
> How about running ctl_cyrusdb -r under truss to see what it is doing
> during that time? I am unsure if you have to stop Cyrus first to do
this.
>
> Andy
>
> On Mon, 13 Mar 2006, Rosenbaum, Larry M. wrote:
>
> > Any ideas on this issue?
> >
> > -----Original Message-----
> > From: Rosenbaum, Larry M.
> > Sent: Saturday, February 25, 2006 2:21 PM
> > To: [EMAIL PROTECTED]
> > Subject: Cyrus takes 5 minutes to start
> >
> > We have just upgraded from an ancient version of Cyrus to a new one
on
> > some new hardware. I have discovered that Cyrus takes 4-5 minutes
to
> > start:
> >
> > Feb 25 13:42:15 email master[327]: [ID 965400 local6.notice] process
> > started
> > Feb 25 13:42:15 email ctl_cyrusdb[346]: [ID 702911 local6.notice]
> > recovering cyrus databases
> > Feb 25 13:43:07 email ctl_cyrusdb[346]: [ID 275131 local6.notice]
> > skiplist: recovered /var/imap/mailboxes.db (30547 records, 2591408
> > bytes) in 52 seconds
> > Feb 25 13:43:07 email ctl_cyrusdb[346]: [ID 275131 local6.notice]
> > skiplist: recovered /var/imap/annotations.db (0 records, 144 bytes)
in 0
> > seconds
> > Feb 25 13:47:04 email ctl_cyrusdb[346]: [ID 127214 local6.notice]
done
> > recovering cyrus databases
> > Feb 25 13:47:04 email master[327]: [ID 139525 local6.notice] ready
for
> > work
> >
> > What was it doing between 13:43:07 and 13:47:04? Is there any way
to
> > speed this up?
> >
> > Software: Cyrus v2.3.1 on Solaris 9
> > Hardware: Sun v210 with 2 GB memory, 2 1.3GHz processors, 9 GB swap
> > Database types: using the recommended default types
> > Size: 2000 users, 30,547 folders (i.e. there were 30,547 lines in
the
> > old /var/imap/mailboxes)
> >
> > Larry
> >
> > ----
> > Cyrus Home Page: http://asg.web.cmu.edu/cyrus
> > Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
> > List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
> >
----
Cyrus Home Page: http://asg.web.cmu.edu/cyrus
Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
