Re: mail server replication
Rob, Rob Siemborski wrote: I hope this kind of functionality comes into cyrus once; I assume it is not there yet, I once asked this on the list before ;-) And maybe it shouldn't be too difficult, since there is already a synchronisation mechanism for NNTP in place, if I'm correct. This is very hard -- if not impossible -- with only two machines, at the very least you'd need three in order to maintain a quorum of what the "real view" of the server is -- For example, think about what would happen if there was a network partition and two different messages were delivered to the split servers at the same time. It's not an easy problem to solve-- keeping a hot spare is definately the way to go. I agree; but what I meant was a synchronisation mechanism that can synchronise a spool with/to a hot spare server - without having the need for a RAID unit with two scsi interfaces. (This way you remove another SPOF ;-)) Maybe I see this a bit like the OpenLDAP slave/master replication idea: something must "switch" in order to make the slave the master and vise versa. Cyrus could synchronise the slave with every message delivered, or when keeping a log and synchronising the messages apart. Although I think Gerard likes to see both servers active at the same time, I think the (master-slave kind) synchronisation would be a nice thing to start with. On the other hand, if I see what offlineimap can do, I assume it must be possible with just 2 servers to synchronise folders in a proper way (when keeping some history and logs on both sides, of course...), am I wrong? (This looks again a bit like bi-directional synchronisation as with unison, instead of master->slave think as with rsync or so.) Regards, Paul --- 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
Re: mail server replication
Paul Dekkers said: > Rob, > > Rob Siemborski wrote: > >>> I hope this kind of functionality comes into cyrus once; I >>> assume it >>> is not there yet, I once asked this on the list before ;-) And >>> maybe >>> it shouldn't be too difficult, since there is already a >>> synchronisation mechanism for NNTP in place, if I'm correct. >> >> This is very hard -- if not impossible -- with only two >> machines, at >> the very least you'd need three in order to maintain a quorum of >> what >> the "real view" of the server is -- >> >> For example, think about what would happen if there was a >> network >> partition and two different messages were delivered to the split >> servers at the same time. It's not an easy problem to solve-- >> keeping >> a hot spare is definately the way to go. > > I agree; but what I meant was a synchronisation mechanism that can > synchronise a spool with/to a hot spare server - without having > the need > for a RAID unit with two scsi interfaces. (This way you remove > another > SPOF ;-)) > > Maybe I see this a bit like the OpenLDAP slave/master replication > idea: > something must "switch" in order to make the slave the master and > vise > versa. Cyrus could synchronise the slave with every message > delivered, > or when keeping a log and synchronising the messages apart. > > Although I think Gerard likes to see both servers active at the > same > time, I think the (master-slave kind) synchronisation would be a > nice > thing to start with. On the other hand, if I see what offlineimap > can > do, I assume it must be possible with just 2 servers to > synchronise > folders in a proper way (when keeping some history and logs on > both > sides, of course...), am I wrong? (This looks again a bit like > bi-directional synchronisation as with unison, instead of > master->slave > think as with rsync or so.) Take a look at drbd http://www.drbd.org . You can do exactly what you are describing using drbd and heartbeat. -- Neil Schneider pacneil_at_linuxgeek_dot_net http://www.paccomp.com Key fingerprint = 67F0 E493 FCC0 0A8C 769B 8209 32D7 1DB1 8460 C47D Fires can't be made with dead embers, nor can enthusiasm be stirred by spiritless men. Enthusiasm in our daily work lightens effort and turns even labor into pleasant tasks. --James Baldwin --- 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
Re: cyrus-2.2.3 + cyrus-sasl + ldap problem
On Fri, 21 May 2004, Andrew B. Panphiloff wrote: why in first case saslauthd get "[EMAIL PROTECTED]" and "realm=" but in second case it get "user=jeka" and "realm=8ka.mipt.ru" ? How fix this behaviour ? Because SASL treats everything after the @ sign as a 'realm'. You'll need to fix your filter accordingly (e.g. [EMAIL PROTECTED]) -Rob -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456 Research Systems Programmer * /usr/contributed Gatekeeper --- 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
Re: mail server replication
On Fri, 21 May 2004, Paul Dekkers wrote: Although I think Gerard likes to see both servers active at the same time, I think the (master-slave kind) synchronisation would be a nice thing to start with. On the other hand, if I see what offlineimap can do, I assume it must be possible with just 2 servers to synchronise folders in a proper way (when keeping some history and logs on both sides, of course...), am I wrong? (This looks again a bit like bi-directional synchronisation as with unison, instead of master->slave think as with rsync or so.) I'm not familiar with what offlineimap does, but it isn't possible to have two servers syncing "in unison" without some user interaction to resolve conflicts when they occur after a network partition. Obviously a master/slave situation is much easier to achieve. -Rob -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456 Research Systems Programmer * /usr/contributed Gatekeeper --- 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
Moving mailboxes som one server to another?
Is there another way of doing this than to copy the mailboxes and running reconstruct on every mailbox and every subfolder? Or to put it simple, does anyone have a list on how to best do this? Regards, BTJ -- --- Bjørn T Johansen (BSc,MNIF) Executive Manager [EMAIL PROTECTED] Havleik Consulting Phone : +47 67 54 15 17 Bjørnebærstien 57 Fax : +47 67 54 13 91 N-1348 Rykkinn Cellular : +47 926 93 298 http://www.havleik.no --- "The stickers on the side of the box said "Supported Platforms: Windows 98, Windows NT 4.0, Windows 2000 or better", so clearly Linux was a supported platform." --- --- 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
idle + sieve: only applies to INBOX?
cyrus-imapd-2.2.4 I use sieve extensively to filter my email to several mailboxes (dozens) and I noticed that Thunderbird 0.6 uses the IDLE extension. Cool! However, I also noticed that it is only applied to the INBOX mailbox, and not the others. So, how does idle work? Does it apply only to the currently selected mailbox or to all mailboxes as a whole? If it applies only to the currently selected mailbox, then should the client open one connection for each mailbox, select it and issue the idle command in order to get notifications for all mailboxes? --- 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
Re: IMAP/LMTP/Quota locking problem
Hi, --On Dienstag, 18. Mai 2004 8:53 Uhr -0400 Scott Adkins <[EMAIL PROTECTED]> wrote: --On Tuesday, May 18, 2004 9:13 AM +0200 Sebastian Hagedorn <[EMAIL PROTECTED]> wrote: thanks for the info. The patch was definitely added *after* 2.2.1 ... hmm, now that I think about it: seems like it was even added after 2.2.3! Is bug 1270 really fixed in CVS only? Good question! :) I hope it is fixed :) as we have learned it is in 2.2.4 ... If I remember right, it has something to do with SSL. Are you using SSL? Yes. The IMAP session in question used TLS ... Currently, we manage the problem by watching the sendmail syslog file for System I/O errors on the Cyrus mailer (grep for 'mailer=cyrus.*System.I.O') We don't get those. I wonder if we actually have separate problems. Or maybe not. However, in our case the sendmail processes get stuck as well, maybe due to our setting of DeliveryMode to interactive!? I find that interesting... after an IMAP process locks the quota file, the LMTP processes start to stack up. Right. After awhile, anything trying to be delivered to that user generates a System I/O error. Not for us. I'm not sure what exactly happens to the sendmail processes, but they seem to get stuck as well. I haven't yet had the opportunity to analyse that. Cheers, Sebastian Hagedorn -- Sebastian Hagedorn M.A. - RZKR-R1 (Gebäude 52), Zimmer 18 Zentrum für angewandte Informatik - Universitätsweiter Service RRZK Universität zu Köln / Cologne University - Tel. +49-221-478-5587 pgp7wOx9HfJbt.pgp Description: PGP signature
Re: cyrus-2.2.3 + cyrus-sasl + ldap problem
On Fri, 21 May 2004, Andrew B. Panphiloff wrote: > I have strange behaviour of cyrus-imapd and cyrus-sasl. > > imapd config : > > -- > tls_ca_file: /etc/ssl/cyrus.pem > tls_cert_file: /etc/ssl/cyrus.pem > tls_key_file: /etc/ssl/cyrus.pem > virtdomains: yes > defaultdomain: localhost > configdirectory: /var/imap > partition-default: /var/spool/imap > sievedir: /var/imap/sieve > altnamespace: no > unixhierarchysep: no > lmtp_downcase_rcpt: yes > admins: cyrus > allowanonymouslogin: no > popminpoll: 0 > autocreatequota: 10 > createonpost: yes > autocreateinboxfolders: Sent | Drafts | Templates | Trash > autosubscribeinboxfolders: Sent | Drafts > umask: 077 > sieveusehomedir: false > hashimapspool: true > allowplaintext: yes > sasl_mech_list: plain login > sasl_minimum_layer: 0 > sasl_pwcheck_method: saslauthd > sasl_auto_transition: no > tls_ca_path: /etc/ssl/certs > tls_session_timeout: 1440 > tls_cipher_list: TLSv1:SSLv3:SSLv2:!NULL:!EXPORT:!DES:!LOW:@STRENGTH > lmtpsocket: /var/run/cyrus/socket/lmtp > idlesocket: /var/run/cyrus/socket/idle > notifysocket: /var/run/cyrus/socket/notify > --- > > saslauthd.conf config: > > --- > ldap_servers: ldap://127.0.0.1 > ldap_bind_dn: cn=admin,o=8ka.mipt.ru > ldap_bind_pw: xxx > ldap_version: 3 > ldap_search_base: ou=Mail,o=8ka.mipt.ru > ldap_filter: mail=%u Change to ldap_filter: [EMAIL PROTECTED] and things will work. > why in first case saslauthd get "[EMAIL PROTECTED]" and "realm=" if you want to emulate imapd behavior, you need to 'testsaslauthd -u jeka -r 8ka.mipt.ru -p xxx' > but in second case it get "user=jeka" and "realm=8ka.mipt.ru" ? > How fix this behaviour ? libsasl splits fully qualified username before it is passed to saslauthd. An alternate fix is to download the cvs version of saslauthd and use -r option which reassembles fully qualified username before it is passed to authentication mechs. -- Igor --- 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
Re: idle + sieve: only applies to INBOX?
On Fri, 21 May 2004, Andreas wrote: So, how does idle work? Does it apply only to the currently selected mailbox or to all mailboxes as a whole? If it applies only to the currently selected mailbox, then should the client open one connection for each mailbox, select it and issue the idle command in order to get notifications for all mailboxes? Just the currently selected mailbox (if any). You can think of it as having the server acting as if a continuous stream of NOOP commands was arriving from the client. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456 Research Systems Programmer * /usr/contributed Gatekeeper --- 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
Re: Moving mailboxes som one server to another?
On Fri, 21 May 2004, [ISO-8859-1] Bjørn T Johansen wrote:
> Is there another way of doing this than to copy the mailboxes and
> running reconstruct on every mailbox and every subfolder? Or to put it
> simple, does anyone have a list on how to best do this?
Bjørn,
I think the generally accepted way of doing this is using IMAP to move
the mailboxes with mailutil or some other such program. Lots of people
have done this other ways, but it seems that they often end up either
losing information, or investing a lot of energy in custom/local scripts
to make the move work.
Search the archives (see the URL below) for "migrate" and you'll have
days worth of reading on this subject. In the end, download the pine
source of University of Washington and compile the "mailutil" program that
comes with it. An approximate command line is:
mailutil copy {oldserver/novalidate-cert/authuser=admin/user=username} \
{newserver/novalidate-cert/authuser=admin/user=username}
Good luck.
--
Andrew Caird
---
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
Re: mail server replication
Yeah, master slave sort of thing would work nice. If the server losses an email or two that would be no problem. Plus I can always set up an MX record for both servers so it would be delivered no matter what. This way if one server goes down I have the other as a backup and if server one catches on fire and I cannot get anything back I have server two, and noone knows the difference. Its always good to have a backup :) Gerard On Fri, 2004-05-21 at 09:23, Rob Siemborski wrote: > On Fri, 21 May 2004, Paul Dekkers wrote: > > > Although I think Gerard likes to see both servers active at the same > > time, I think the (master-slave kind) synchronisation would be a nice > > thing to start with. On the other hand, if I see what offlineimap can > > do, I assume it must be possible with just 2 servers to synchronise > > folders in a proper way (when keeping some history and logs on both > > sides, of course...), am I wrong? (This looks again a bit like > > bi-directional synchronisation as with unison, instead of master->slave > > think as with rsync or so.) > > I'm not familiar with what offlineimap does, but it isn't possible to have > two servers syncing "in unison" without some user interaction to resolve > conflicts when they occur after a network partition. > > Obviously a master/slave situation is much easier to achieve. > > -Rob > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456 > Research Systems Programmer * /usr/contributed Gatekeeper > signature.asc Description: This is a digitally signed message part
virtdomains and GSSAPI auth
Hi, on a system running Cyrus IMAP 2.2.3 I switched to using virtual domains. GSSAPI authentication is used. From my imapd.conf: admins: jukka/[EMAIL PROTECTED] virtdomains: on defaultdomain: bart.stasoft.ch loginrealms: stasoft.ch After getting a TGT for jukka/admin, when starting cyradm, syslog gets May 21 16:25:10 bart imap[25634]: login: bart.stasoft.ch [192.168.12.104] jukka/[EMAIL PROTECTED] GSSAPI User logged in but it seems that I'm not authenticated as the admin user, at least "lm" doesn't show any mailboxes. If I change the "defaultdomain" option to "stasoft.ch" everything works fine; syslog says May 21 16:32:08 bart imap[25656]: login: bart.stasoft.ch [192.168.12.104] jukka/admin GSSAPI User logged in What do I have to change to successfully authenticate as an admin user if defaultdomain != Kerberos realm name? TIA, Jukka -- bashian roulette: $ ((RANDOM%6)) || rm -rf ~ --- 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
Re: complett relocation from old to new server and cyrus version
On Wed, May 12, 2004 at 10:49:18AM +0200, Pascal Gienger wrote: > 1. Dump your mailboxes-Database on the old system: > ctl_mboxlist -d > mailboxes.dump > Copy the resulting "mailboxes.dump" to the new server. ... Does one have to do any of that when upgrading say 2.2-beta to 2.4, or can one reasonably expect just changing the server binaries to be sufficient? (Assuming of course that say a berkley doesn't become a skiplist) Cheers, Patrick --- 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
Re: virtdomains and GSSAPI auth
Jukka Salmi --> info-cyrus (2004-05-21 16:41:27 +0200): > What do I have to change to successfully authenticate as an admin user > if defaultdomain != Kerberos realm name? Setting "virtdomains: userid" did the trick. Jukka -- bashian roulette: $ ((RANDOM%6)) || rm -rf ~ --- 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
Alternative to cyradm
Hello all, Is there an alternative to cyradm? I haven't been able to get this to run yet. I continue to get: IMAP.so: undefined symbol: db_version. -- Kent L. Nasveschuk <[EMAIL PROTECTED]> --- 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
Re: mail server replication
Dear Paul and others, I haven't tried it yet but it may be that DRBD (http://www.drbd.org) might be able to do what you want. We used it with a UW Imap server and I don't see why it shouldn't work with Cyrus. It is probably possible to split the users between two cyrus servers and have each group replicated to the other server so that each server could become a server with all users reasonably quickly. I suspect it would take a few minutes to fail over. As I say I haven't tried this so perhaps it won't work. Best wishes. Colin On Fri, 21 May 2004, Paul Dekkers wrote: > Rob, > > Rob Siemborski wrote: > > >> I hope this kind of functionality comes into cyrus once; I assume it > >> is not there yet, I once asked this on the list before ;-) And maybe > >> it shouldn't be too difficult, since there is already a > >> synchronisation mechanism for NNTP in place, if I'm correct. > > > > This is very hard -- if not impossible -- with only two machines, at > > the very least you'd need three in order to maintain a quorum of what > > the "real view" of the server is -- > > > > For example, think about what would happen if there was a network > > partition and two different messages were delivered to the split > > servers at the same time. It's not an easy problem to solve-- keeping > > a hot spare is definately the way to go. > > I agree; but what I meant was a synchronisation mechanism that can > synchronise a spool with/to a hot spare server - without having the need > for a RAID unit with two scsi interfaces. (This way you remove another > SPOF ;-)) > > Maybe I see this a bit like the OpenLDAP slave/master replication idea: > something must "switch" in order to make the slave the master and vise > versa. Cyrus could synchronise the slave with every message delivered, > or when keeping a log and synchronising the messages apart. > > Although I think Gerard likes to see both servers active at the same > time, I think the (master-slave kind) synchronisation would be a nice > thing to start with. On the other hand, if I see what offlineimap can > do, I assume it must be possible with just 2 servers to synchronise > folders in a proper way (when keeping some history and logs on both > sides, of course...), am I wrong? (This looks again a bit like > bi-directional synchronisation as with unison, instead of master->slave > think as with rsync or so.) > > Regards, > Paul > > --- > 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
Re: mail server replication
Colin Bruce wrote: I haven't tried it yet but it may be that DRBD (http://www.drbd.org) might be able to do what you want. We used it with a UW Imap server and I don't see why it shouldn't work with Cyrus. It is probably possible to split the users between two cyrus servers and have each group replicated to the other server so that each server could become a server with all users reasonably quickly. I suspect it would take a few minutes to fail over. As I say I haven't tried this so perhaps it won't work. drbd in combination with "heartbeat" and a journalling filesystem can do exactly this. You can have Cyrus IMAP running on both servers (different users), with the Cyrus storage areas mirrored to the other server via drbd. When heartbeat notices that one of the servers has died, it can mount the other server's storage area (since it has a copy) and start up Cyrus (and take over the other server's IP address as well, of course). Users would notice a service disruption, but it's not likely any mail would be lost and they would only have to reconnect. If their mail client is set to only connect/check their mailboxes every few minutes, they may not notice the switchover at all :-) --- 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
Problem w/ 2.2.4 and unixhierarchysep: yes
Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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
Re: Problem w/ 2.2.4 and unixhierarchysep: yes
Here is an odd update to this... I tried to create another user with a dot and it works fine. It seems john.smith causes the error whereas al.jones does not. Has anyone seen anything like this? Thanks. AJ AJ wrote: Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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
rare infinite loop in cyrus-imapd-2.2.3 seendb processing
Hi, I think I hit a rare infinite loop somewhere in the seendb processing code of cyrus-imapd-2.2.3, during which an imapd process takes 99% CPU cycles and keep accessing the hard disk and updating the user seendb. I cannot reproduce the problem. So I can only describe what I was doing. I was experimenting to migrate an imap folder from UW-IMAP to Cyrus IMAP v2.2.3 on Red Hat Linux 9, with both IMAP servers running on different IP addresses of the same machine. First, I accessed an Cyrus IMAP account "wenzhuo" using Mozilla Mail and kept it open. Then I deleted the IMAP account and recreated it immediately. After that, I tried to run the migration script. No sooner had I started the script, did I notice the significant slowdown of the system. The system exhibits the following symptoms: 1. One Cyrus imap process eats 99% CPU cycles. 2. vmstat shows several block writes per second (bi) and around 200 block reads per second (bo), and around 130 interrupts per second. 3. The timestamp of the flat seendb file wenzhuo.seen kept changing. Here is the system log of the misbehaving process during its lifetime: May 21 17:56:45 matrix master[15715]: about to exec /usr/lib/cyrus-imapd/imapd May 21 17:56:45 matrix imap[15715]: executed May 21 17:58:30 matrix imap[15715]: accepted connection May 21 17:58:30 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:30 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:30 matrix imap[15715]: open: user wenzhuo opened INBOX May 21 17:58:32 matrix imap[15715]: accepted connection May 21 17:58:32 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:32 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:32 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:33 matrix imap[15715]: accepted connection May 21 17:58:34 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:34 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:34 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:34 matrix imap[15715]: accepted connection May 21 17:58:34 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:34 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:34 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:35 matrix imap[15715]: accepted connection May 21 17:58:35 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:35 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:35 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:37 matrix imap[15715]: accepted connection May 21 17:58:37 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:37 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:37 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:38 matrix imap[15715]: accepted connection May 21 17:58:38 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:38 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:38 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:39 matrix imap[15715]: accepted connection May 21 17:58:39 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:39 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:40 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:41 matrix imap[15715]: accepted connection May 21 17:58:41 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:41 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:41 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:42 matrix imap[15715]: accepted connection May 21 17:58:42 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:42 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:42 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:43 matrix imap[15715]: accepted connection May 21 17:58:43 matrix imap[15715]: login: localhost.localdomain [127.0.0.1] wenzhuo plaintext May 21 17:58:43 matrix imap[15715]: open: user wenzhuo opened sent-mail May 21 17:58:43 matrix imap[15715]: seen_db: user wenzhuo opened /var/lib/imap/user/w/wenzhuo.seen May 21 17:58:44 matrix imap[15715]: accepted connection May 21 17:58:44 matrix imap[15715]: login: loca
Re: Problem w/ 2.2.4 and unixhierarchysep: yes
The pieces begin to come together here.. hopefully someone else benefits from this post. I have managed to track the problem down to not just accounts with a dot in the mailbox name. This problem is occuring on mailboxes with quotas only. Mailboxes that do not have quotas do not experience this issue. My imapd.conf file is below, does anyone know why this is happening? Once I issue these command in cyradm, this issue happens for the mary.jones mailbox. localhost> sq user/mary.smith 8192 quota:8192 localhost> lq user/mary.smith STORAGE 1/8192 (0.01220703125%) Here is imapd.conf: configdirectory: /var/cyrus/imap partition-default: /var/cyrus/spool/imap admins: cyrus sievedir: /var/cyrus/sieve sendmail: /usr/sbin/sendmail hashimapspool: true sasl_pwcheck_method: saslauthd sasl_mech_list: PLAIN LOGIN altnamespace: yes unixhierarchysep: yes Thanks. AJ AJ wrote: Here is an odd update to this... I tried to create another user with a dot and it works fine. It seems john.smith causes the error whereas al.jones does not. Has anyone seen anything like this? Thanks. AJ AJ wrote: Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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
Re: Problem w/ 2.2.4 and unixhierarchysep: yes
AJ wrote: Here is an odd update to this... I tried to create another user with a dot and it works fine. It seems john.smith causes the error whereas al.jones does not. Has anyone seen anything like this? user/john.smith may be corrupt in some way. Try reconstructing the mailbox and see what happens. AJ wrote: Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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 -- Kenneth Murchison Oceana Matrix Ltd. Software Engineer 21 Princeton Place 716-662-8973 x26 Orchard Park, NY 14127 --PGP Public Key--http://www.oceana.com/~ken/ksm.pgp --- 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-imapd-2.2.4 "cannot continue without libsasl2."
I am attempting to compile Cyrus-imap-2.2.4 on FreeBSD 5.2.1/RELEASE, and I am having problems. cyrus-sasl-2.1.18 is correctly installed from source. However, when running configure, regardless of options, I continue to get the error "Cannot continue without libsasl2". I have tried some of the ideas from the archives, including copying the sasl libraries and include files to a mess of different places (/usr/include, /usr/local, etcetera). I've used --with-sasl= with a number of different directories, all to no avail. It really seems that it cannot find sasl.h. The last few lines of output from configure are: checking for gsskrb5_register_acceptor_identity... no checking sasl/sasl.h usability... no checking sasl/sasl.h presence... no checking for sasl/sasl.h... no configure: error: Cannot continue without libsasl2. Get it from ftp://ftp.andrew.cmu.edu/pub/cyrus-mail/. But, of course, I do have it. My ./configure line is: ./configure --with-sasl=/usr/local/lib --with-perl --with-auth=unix --with-dbdir=/usr/local/bdb --with-bdb-incdir=/usr/local/bdb/include --with-bdb-libdir=/usr/local/bdb/lib --with-openssl=/usr/local/ssl --without-ucdsnmp After plugging at this for the last several hours, I'm giving in and asking for help ... ? Mark Nernberg --- 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
Re: Problem w/ 2.2.4 and unixhierarchysep: yes
This problem does not appear in 2.2.3, I just wiped my entire 2.2.4 install and installed 2.2.3 and no issues. Ideas? AJ AJ wrote: The pieces begin to come together here.. hopefully someone else benefits from this post. I have managed to track the problem down to not just accounts with a dot in the mailbox name. This problem is occuring on mailboxes with quotas only. Mailboxes that do not have quotas do not experience this issue. My imapd.conf file is below, does anyone know why this is happening? Once I issue these command in cyradm, this issue happens for the mary.jones mailbox. localhost> sq user/mary.smith 8192 quota:8192 localhost> lq user/mary.smith STORAGE 1/8192 (0.01220703125%) Here is imapd.conf: configdirectory: /var/cyrus/imap partition-default: /var/cyrus/spool/imap admins: cyrus sievedir: /var/cyrus/sieve sendmail: /usr/sbin/sendmail hashimapspool: true sasl_pwcheck_method: saslauthd sasl_mech_list: PLAIN LOGIN altnamespace: yes unixhierarchysep: yes Thanks. AJ AJ wrote: Here is an odd update to this... I tried to create another user with a dot and it works fine. It seems john.smith causes the error whereas al.jones does not. Has anyone seen anything like this? Thanks. AJ AJ wrote: Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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
Re: Problem w/ 2.2.4 and unixhierarchysep: yes
There were substantial changes in the handling of quotas in 2.2.4. However, we're unable to replicate your problem. Can you generate a GDB backtrace from a core dump to show where the segfault is occuring? On Fri, 21 May 2004, AJ wrote: This problem does not appear in 2.2.3, I just wiped my entire 2.2.4 install and installed 2.2.3 and no issues. Ideas? AJ AJ wrote: The pieces begin to come together here.. hopefully someone else benefits from this post. I have managed to track the problem down to not just accounts with a dot in the mailbox name. This problem is occuring on mailboxes with quotas only. Mailboxes that do not have quotas do not experience this issue. My imapd.conf file is below, does anyone know why this is happening? Once I issue these command in cyradm, this issue happens for the mary.jones mailbox. localhost> sq user/mary.smith 8192 quota:8192 localhost> lq user/mary.smith STORAGE 1/8192 (0.01220703125%) Here is imapd.conf: configdirectory: /var/cyrus/imap partition-default: /var/cyrus/spool/imap admins: cyrus sievedir: /var/cyrus/sieve sendmail: /usr/sbin/sendmail hashimapspool: true sasl_pwcheck_method: saslauthd sasl_mech_list: PLAIN LOGIN altnamespace: yes unixhierarchysep: yes Thanks. AJ AJ wrote: Here is an odd update to this... I tried to create another user with a dot and it works fine. It seems john.smith causes the error whereas al.jones does not. Has anyone seen anything like this? Thanks. AJ AJ wrote: Hi, I have an odd problem. I am running 2.2.4 fresh install and I have an issue when using a mailbox with a . in the name, such as user/john.smith I have the unixhierarchysep: yes line in my imapd.conf. Whenever I try to delete a message or move a message to another folder with this mailbox, I generate these errors: May 21 18:44:58 linux-beta master[2745]: process 2753 exited, signaled to death by 11 May 21 18:44:58 linux-beta master[2745]: service imap pid 2753 in BUSY state: terminated abnormally And I get errors on the client that said the connection was terminated before the command could complete. It looks like the files get "copied" but not moved, so it appears to be a delete issue. Users that do not have a . in their mailbox name do not have this problem. Has anyone seen this? I have another version of 2.2.4 running, and this does not happen there, but that version was upgraded. The only other difference is that the version that has this issue is running bdb 4.2.x and the other system is running 4.1.x. I just wanted to throw this out there to see if anyone knows anything about this. Thanks. AJ --- 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 -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456 Research Systems Programmer * /usr/contributed Gatekeeper --- 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
Not seeing sub-folders when migrating/upgrading
I am migrating from an older cyrus-imapd/postfix/webcyradm/squirrelmail solution on RH 7.3 to a new one with new versions of each on Fedora Core 1. All is setup and working fine with the exception of one problem: On my old box I have a user called 'link0033'. For testing, I created this same user on my new box with the same password. On the old system, I tar'd up the user's directory, copied it to my new box, and untar'd it. I set the appropriate permissions, then logged in 'link0033' on my new box via Squirrelmail. The problem is I only see the Inbox, Drafts, Sent, and Trash folders. The good news is that each has mail in it as they should. The bad news is that on my old box, I had created sub-folders of Work and Personal. Each has mail in it on the old server. When I log in via Squirrelmail, I don't see these folders. However, at the command line I can see them, see they have the permissions, and see the content within them. For kicks, I used the cyradm program as another test with: cyradm --user link0001 --server localhost --auth plain Once in, I did an "lm" and I only see the Inbox, Drafts, Sent, and Trash folders. So I now have two different programs with the same results. This is pointing me at a cyrus issue. Interestingly, on my new server, I was able to move the Personal folder via the command line, go to Squirrelmail -> and create a new Personal folder, then delete it at the command line and move the original back. When I log back into Squirrelmail, I can see the Personal folder *and* the contents of it that were brought over from my old server. So, is there something I need to be passing as a flag to cyrus? Perhaps a build option I should have used? My new system has cyrus-imapd-2.2.4. My older system has 2.1.12. Did something change between versions that's to blame? Is there a patch or tool I can use? Thanks in advance, -- Andrew Davis North County Computers http://www.nccomp.com Email: [EMAIL PROTECTED] Local: 760-525-4689 Toll Free: 877-735-4689 http://www.nccomp.com";>North County Computers --- 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
