Henrique de Moraes Holschuh wrote:
Any ideas why we don't just kill non-native mmap? Without it Cyrus cannot
perform to satisfaction anyway...
Because mmap won't work on HP-UX (yes, it still performs well enough
with BDB, even though skiplist is useless without mmap). Maybe if Cyrus was
modifi
Rosenbaum, Larry M. wrote:
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:
Well, it has to wade through every mailbox to check out that they
are OK. Five minutes for startup can be quite
Rob Mueller wrote:
I don't know anyone who uses BDB for critical DB's in cyrus. They're too
big for seen/sub state db's, too slow for enumerating across the
mailboxes db, which really only leaves suplicate delivery and tls cache
dbs.
HP-UX people. The mmap() implementation on HP-UX won't do f
This applies to 2.2.12, but I believe the same issue is present on later
versions too. On HP-UX 11.*, accept() can return ENOBUFS for no particular
reason, typically when the client went away before the program got around
to calling accept() for it. Current Cyrus IMAP code would then terminate
the
Can anyone explain the locking scheme that is built in the accept()
loop in master/service.c (and service-thread.c, imap 2.2.12)?
I think I understand the code, but not what it's trying to
accomplish. Some lock files are created under socket/, but
only two per service ("id"'s 0 and 1, only 0 for
Pascal Mouret wrote:
Has anybody got an idea ? Where did I go wrong ? Could it be a result of
running version 2.2.12 cvt_cyrusdb while 2.0.16 is still up ?
Running two different versions (or even having them present on the
system) of BDB libraries is always somewhat a problem. Save yourself
a l
Short question - are the skiplist database files binary compatible
across all Cyrus versions (particulary 2.1 -> 2.2), so I don't
have to worry about any upgrading issues?
--mika
---
Cyrus Home Page: http://asg.web.cmu.edu/cyrus
Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
List Archives/Info:
Robert Covell wrote:
We are using deliver only to notify them of being over quota. Everything
else is coming in via sendmail. It does notify the sender of the user being
over quota but the user still gets the email. Just wonder if anything else
could be causing conflicting settings.
Yes, this is
RJ45 wrote:
Hello I haev lately this error mesage always the same time of day once a
day:
imaplog:Feb 7 04:00:00 venus ctl_deliver[429]: [ID 866726 local6.warning]
DBERROR db4: 6 lockers
This is (almost) completely normal. If the number of lockers keeps
rising constantly, that is not. It may ve
Jeff Wilde wrote:
I just wanted to verify if the Cyrus Package Supports the Latest Version
of Berkley DB. I haven't seen where it says otherwise, but just would
like a confirmation. I saw in the archives where it said it wasn't
supported at one time, but what is the status now?
It does compile
Rob Siemborski wrote:
[skiplist]
No, it doesn't have a specific in-core cache, it relies on the filesystem
cache alone (it mmap()s the db file and uses that directly).
Can you tell offhand if it does any of the mmap() stuff that's a no-no
on HP-UX platform (eg. same process mapping two overlappi
Richard Gilbert wrote:
Having watched the activity, the thing which seems to consume most of the
resources are the LIST "" * and LIST "" % commands. These return with a
response like "OK Completed (1.480 secs 28 calls)" but can take a very
long real time. Is there anything which can be done to s
Nicola Ranaldo wrote:
in imapd.c I read:
/*
* run once when process is forked;
* MUST NOT exit directly; must return with non-zero error code
*/
int service_init(int argc, char **argv, char **envp)
but mboxlist_init() is a void! it calls directly fatal() and so exit().
Master will not chec
Voutsinas Nikos wrote:
It seems that there is a problem when recovering (ctl_cyrusdb -r). I
always get SIGBUS (in __db_apprec.)
To test it just send SIGTERM to master and then restart it.
I tried that, and it just worked. Even kill -9 didn't seem to
trigger any problems. Perhaps you have some
Jay Levitt wrote:
Thinking further about this, I no longer like my idea of a
separate --perl-prefix configure option, because if I do
"configure --prefix=/home/jay", I should reasonably expect that nothing is
being installed into sitewide libraries.
You're right there. Getting cyradm and the p
Voutsinas Nikos wrote:
Just a simple question. Is anyone out there running succesfully
cyrus-imapd with DB 4.1.24 ?
Do you have some specific problems? It's not in production over here
yet, but during testing I haven't been able to pinpoint any problem that
would be due to BDB 4.1.
Sure, I'd
Ian McDonald wrote:
Please, what are the groups, if not using Unix auth?
AFAIK you'd have to use Kerberos. If you have --with-auth=unix, the
group lookups go to the normal getpwnam()/getgrnam() functions, which
finally consult anything your name service switch is configured to do.
Theoretical
Does anyone know what the current status on using mmap() on HP-UX is?
Cyrus documentation only mentions HP-UX 9 and 10 as not working (probably
because that chapter is just old) but I think 11.11 is supposed to be
greatly improved in this area.
Cyrus 2.0.16 has been running just fine with no_mmap,
Ian McDonald wrote:
Try using "--with-auth=unix" instead of --without-des
Does this bit of the manual mean that if I do not want to store
authorization information in /etc/passwd, because the mailbox users have
nothing to do with the unix users, I need to use Kerberos?
--with-auth=unix only imp
Mika Iisakkila wrote:
> ...so I suppose something in lib/cyrusdb_db3.c still isn't quite
> compatible with BDB 4.1. I get this same error in Linux and HP-UX 11.11,
> so I suppose it's general and not a compilation screwup. Any
> off-the-shelf ideas on how to fix this, be
Some time back I posted a small patch for compiling SASL2 & imapd with
Berkeley DB 4.1 (rather straightforward, just followed the Sleepycat
upgrade docs; worked fine for SASL 2.1.7). But there seems to be a
problem which I didn't notice before, since I had duplicate delivery
suppression turned off.
Phil Ellett wrote:
> I am running Cyrus IMAP 2.1.9 and have been having problems with deleted
> folders still appearing in the mailbox list.
> ctl_mboxlist -d > mailboxes.txt
> vi mailboxes.txt and remove entries for deleted folders
> ctl_mboxlist -u < mailboxes.txt
I've found that you need to de
Mika Iisakkila wrote:
> Any hopes for updates to current stable SASL2 and Cyrus 2.1
> to accommodate BDB 4.1?
Boy, was that easy; patches attached. I have no idea if
these actually realize any of the performance gains, but
at least both sasl2 and imapd compile and run.
I can't give a
Amos Gouaux wrote:
> Speaking of which, perhaps some have noticed that Berkeley DB 4.1
> was just released. From the announcement:
That sounds like a real performance win, but even though the databases
ought to be compatible, apparently the API has changed from 4.0
enough that SASL 2 won't even
(sorry Birger, I once again replied privately instead of to the list
as I inteded; any chance of getting the Reply-To to point to the list?)
[EMAIL PROTECTED] wrote:
> connects on the respective ports (or when configuring prefork). It
> also does some background tasks (ctl_mboxlist etc.) What
Peter Pilsl wrote:
> ok: this is a linux-system currently running under kernel2.4.16
...
> But on the last load-test some different problem took place: lmtpd
> consumed 1,2GB memory and therefore made the system unusable. (the
> system went into 'permanent swapping' mode :)
> The problem was not r
On Tue, 4 Dec 2001, Tarjei Huse wrote:
> thanks, does this info also apply if you shut down cyrus prior to backup?
Nope -- if you can shut down cyrus, backing up by normal means is OK.
But when cyrus is running, the mailbox/delivery database state on the
disk is a "moving target", and you can't g
Tarjei Huse wrote:
> PS: Some people noted that in addition to backing up the /var/imap
> and /var/spool/imap folders, you need to export some of the .db files to
> txt. Does anyone want to fill me out on this??
To export mailboxes.db to a text file:
ctl_mboxlist -d > tempfile
To res
Ken Murchison wrote:
> Assuming you are delivering via LMTP, it refuses delivery with a 452
> response (temporary failure).
...
> These are both MTA issues. A stock sendmail config will queue the
> message and continue to attempt delivery for a specified period (7
> days?) before it finally gives
Tarjei Huse wrote:
> Hi, has anyone come around to updating the sasl-ldap patch to 1.5.27?
Which one of them? :-) I just used the "IP-Plus" patch for 1.5.21
that is floating around (can't even remember where I got it).
The patch doesn't work automatically anymore, but it is so simple
that doing i
William K. Hardeman wrote:
> /var/imap and /var/sieve directories. We were migrating from one machine to
> another with identical Cyrus-Berkeley-SASL setups. I don't know how well
> this method would work when mixing-n-matching software versions.
I'd be extra careful in any case. Once the main
Lee Hoffman wrote:
> Thanks a bunch for the advice. I deleted db from /etc/nsswitch.conf and
> viola, master stopped segfaulting on launch.
...
> Oct 21 13:09:27 grass master[1520]: about to exec
> /usr/local/cyrus/bin/imapd
> Oct 21 13:09:27 grass master[1508]: process 1520 exited, signaled to
>
Lee Hoffman wrote:
> Ive followed the directions to a T, while compiling cyrus sasl (w/ldap
> support) and cyrus (2.0.16) (with ssl support) on a debian 2.2r3 distro
> box. The compilation and installation report no errors. I follow the
> installation directions, and ensure that all the directorie
Hello,
I've been involved with migrating a large system to imapd-2.0.16, and
some of the clients are now experiencing corrupted headers. Some newlines
seem to be eaten away, and so the following header line is concatenated to
the preceding one. I've yet to see a good problem report myself, but
th
34 matches
Mail list logo
