RE: Cyrus dies
Dear Sebastian, Thanks for the info especially the configuration suggestions from RedHat. We use a modified Slackware 9.1 installation with lots of our own stuff and a kernel we build ourselves. I just checked and we don't have a file called /proc/sys/vm/pagecache. There are two files called pagetable_cache and page-cluster which have "25 50" and "3" in them respectively. I suspect that taken together they are the same as pagecache. I'll try changing them tonight when there are not too many people around. Anyway, thanks again for the info. Best wishes Colin On Fri, 28 May 2004, Sebastian Hagedorn wrote: > Hi, > > --On Donnerstag, 27. Mai 2004 19:54 Uhr +0100 Colin Bruce > <[EMAIL PROTECTED]> wrote: > > > I have found out the cause of this problem but I don't know what the > > solution is. The good news from your point of view is that it isn't > > anything to do with Cyrus - its our backup software. > > > > The backup cycle starts at 9PM and the problem we are experiencing starts > > about 17 minutes later and finishes 20 minutes after that. The backup > > is still running at that point and continues for another 30 minutes or > > so. It didn't seem to be strongly related. However, I tried doing a manual > > full backup and that didn't cause a problem so I tried a manual > > incremental backup and that killed the server. I had top running at the > > time with 1 second delays between updates but it froze. However, it was > > showing kswapd at the top with 99% CPU time and bpkar second with 56%. > > When I killed the backup the imap system continued happily. > > we have seen similar problems using TSM (Tivoli Storage Manager) backup > software. See also below. > > --On Donnerstag, 27. Mai 2004 13:19 Uhr -0600 Michael Loftis > <[EMAIL PROTECTED]> wrote: > > > Linux has a LOT of known problems with huge amounts of memory...try > > reducing your memsize to 3Gb or 4Gb either by physically pulling chips or > > memsize= on the boot line Lilo you can say something like 'linux > > memsize=4096M' in GRUB you have to use 'e' to edit one of the options, > > then scroll down to the line containing kernel and add memsize= to > > the end of it by 'e' editing that line, then you can use b to boot. See > > if that helps your performance. > > This I can confirm. We have Dell 6450 systems with 8GB of RAM and the > caching has caused us no end of problems. Recentyl we reduced the memory to > 2 GB and things are better. However, now, after two years of problems we > reported to Red Hat, they finally came up with a suggestion about how to > tune the kernel, albeit somewhat vague: they recommended we lower the > values in /proc/sys/vm/pagecache. On RH AS 2.1 they default to: > > 2 50 90 > > We have now reduced them to: > > 2 10 60 > > Hmm, I just noticed that the RH AS 3 values default to: > > 1 15 100 > > Things *seem* to be working better this way, but we'll have to see ... --- 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 dies
On Thu, 27 May 2004, Rob Siemborski wrote: > If even the console isn't responding to the enter key I'm hard pressed to > believe that cyrus is at fault. Dear Rob et al, I have found out the cause of this problem but I don't know what the solution is. The good news from your point of view is that it isn't anything to do with Cyrus - its our backup software. The backup cycle starts at 9PM and the problem we are experiencing starts about 17 minutes later and finishes 20 minutes after that. The backup is still running at that point and continues for another 30 minutes or so. It didn't seem to be strongly related. However, I tried doing a manual full backup and that didn't cause a problem so I tried a manual incremental backup and that killed the server. I had top running at the time with 1 second delays between updates but it froze. However, it was showing kswapd at the top with 99% CPU time and bpkar second with 56%. When I killed the backup the imap system continued happily. I suspect that when the backup software is doing an incremental it builds some sort of list of all the files on the system in memory and by some means this causes it to lock solid. I don't know whether it is swapping at the time but even that shouldn't cause it to lock the way it does. It may be a problem with Linux I suppose. We use Veritas Netbackup (DataCentre) so we can refer the problem to them. This is not the first problem we have had with Veritas and I wish we had bought something else. Ah well such is life. Best wishes. Colin --- 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 hanging (possible saslauthd problem)
>>May 26 20:31:29 imap2 saslauthd[6296]: do_auth : auth failure: >> [user=tellingj] [service=imap] [realm=] [mech=shadow] [reason=Unknown] >> >> This happens even when the password that was typed in was correct. > >Can you run an strace on saslauthd and perhaps find out why it might be >feeling ill occasionally? > >I do know that on Solaris with unix sockets it can behave like this, but >I've never seen it on Linux. > >Is it possible there is some sort of intermittant hardware problem on the >machine? Dear Rob, It could be a problem with hardware but it is brand new and works all day when we have lots of people using it. It seems to fail at night when few people are using it. The thing is, if 10 people login most of them succeed but one or two would get a connection where the inbox never loads. If they try a few minutes later their inbox does load. There doesn't appear to be a pattern to it. I am not sure how we can run strace on saslauthd. Obviously I can do that now but that will just show saslauthd running correctly. We would have to be very lucky to get it to fail just as we run strace; or have I missed something. Best wishes Colin --- 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 dies
Dear Michael and Rob Sorry I should have said what the hardware is. We are running Cyrus on an IBM X345 with dual 3GHz processors, 8GB of memory, dual Gigabit network cards and 6 ultra 320 73GB SCSI disks which are not in a RAID configuration of any kind. I'm afraid there are no IDE drives in the machine. I've checked the cron tables and there are no cron jobs running at the time this happens. Unfortunately it isn't possible to run anything when it happens because the machine is totally locked. It doesn't even respond to me pressing the enter key on the console. I can try running top beforehand and hopefully when it happens it will show something but I think it will stop displaying. The last time master was started was May 17 at 18:36:53. What events could it be running? I know there are some mentioned in the config files but are there any that are hard coded that might be running? I had a look at the logfiles afterwards last night and everything stops. There are no entries in the log files between 9:17PM and 9:35PM. Afterwards everything returns to normal. Best wishes.. Colin --- 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 dies
Dear All, Another problem which may be related to the one I reported earlier but I don't think so. The facts cyrus imap 2.2.3 Linux 2.4.25 The server only runs cyrus imap. There are no other services on this machine and only a few administrators can login. The Problem As far as I can tell every night at approximately 21:15 the server dies. Here is some uptimes as that times approaches. [EMAIL PROTECTED]: uptime 21:01:19 up 40 days, 22:48, 21 users, load average: 0.88, 0.55, 0.33 [EMAIL PROTECTED]: uptime 21:17:33 up 40 days, 23:05, 21 users, load average: 0.18, 0.27, 0.31 [EMAIL PROTECTED]: uptime 21:17:37 up 40 days, 23:05, 21 users, load average: 0.18, 0.27, 0.31 [EMAIL PROTECTED]: uptime 21:17:55 up 40 days, 23:05, 21 users, load average: 0.74, 0.39, 0.34 Following that last one I cant type anything at all. I also can't login to the server and anyone using it will find that they can't read any e-mail. After 10 or 15 minutes like this it will start responding again and the load average will be falling from somewhere between 80 and 100. It will fairly quickly fall to values of less than 1 again. There are no cron or at jobs running at this time. This seems to happen most evenings. Does anyone know what might be the cause? Best wishes..... Colin Bruce --- 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 hanging (possible saslauthd problem)
Dear All, We have been running cyrus imap for a couple of months now and it works well most of the time. We have something like 20,000 registered users with around 800 simultaneous users most of the day. Lately we have noticed a strange problem which we suspect might be saslauthd but might not. I thought it might be worth asking if anyone else has seen this. The facts are: cyrus imap 2.2.3 saslauthd 2.1.8 We are currently using /etc/shadow so saslauthd only supports [EMAIL PROTECTED]: /usr/local/sbin/saslauthd -v saslauthd 2.1.18 authentication mechanisms: getpwent rimap shadow We are running on a Linux server [EMAIL PROTECTED]: uname -a Linux imap2 2.4.25 #1 SMP Thu Apr 15 21:58:02 BST 2004 i686 unknown unknown GNU/Linux The symptoms depend on the e-mail client. However, I think pine shows it best. I can login and read e-mail quite happily most of the time. However, sometimes it accepts my username and password and then says "opening inbox" for ever. Usually, while this is in progress a message will flash up about an untagged response. However, it is not visible long enough to see what the message is about. In any event it will never open the inbox. We have noticed that this is accompanied by a message in the auth file that saslauthd writes failures to such as: May 26 20:31:29 imap2 saslauthd[6296]: do_auth : auth failure: [user=tellingj] [service=imap] [realm=] [mech=shadow] [reason=Unknown] This happens even when the password that was typed in was correct. We have also noticed that if this is happening then it does it for most (if not all) people. However, after a minute or two it stops again and everything is fine for a little while. If the same thing happens with other clients the behaviour may vary. For example, Netscape just reports an incorrect password immediately and continues to do so each time the person tries to enter the password for a few minutes. After a minute or two of this it works fine again. What I am not sure about is whether it is happening every so often or whether I am seeing it every so often and others are seeing the same thing at different times. Has anyone seen this type of thing before? Does anyone know how to fix it? Could it be a locking thing? Is there some resource that is being dead locked? We will be moving to Kerberos in the not too distant future but if there is a quick and easy fix now I would be most grateful. Best wishes.... Colin Bruce --- 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 Kevin, Yes, we ran UW impa for some time with drbd doing replication and it was very successful. Of course UW imap doesn't have any databases so I don't know whether that would cause a problem with drbd. I guess the worst case would be a reconstruct perhaps but even that would be much better than losing the contents of a server. We has used drbd in various places and it is very good. Best wishes... Colin On Fri, 21 May 2004, Kevin P. Fleming wrote: > 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 > --- 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: reading quotas
Dear Pascal,
Thanks for your help. I'll do that instead.
Best wishes
Colin
On Thu, 13 May 2004, Pascal Gienger wrote:
> Colin Bruce <[EMAIL PROTECTED]> wrote:
>
> > Dear All,
> >
> > This may be a stupid question but is there any reason why someone should
> > not be able to obtain their own e-mail quota. If I do something like
> >
> > telnet imaphost 143
> > . login ccx004 password
> > . getquota user.ccx004
> >
> > I get "NO Permission denied"
>
> GETQUOTA is reserverd to a user with admin privilege ("admins:" in
> imapd.conf). Use GETQUOTAROOT instead to get your quota usage:
>
> telnet server 143
> . login myusername mypasswort
> . getquotaroot INBOX
> * QUOTAROOT INBOX INBOX
> * QUOTA INBOX (STORAGE 519605 3145728)
> . OK Completed
>
> Pascal
>
---
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
reading quotas
Dear All, This may be a stupid question but is there any reason why someone should not be able to obtain their own e-mail quota. If I do something like telnet imaphost 143 . login ccx004 password . getquota user.ccx004 I get "NO Permission denied" If I do the same as an administrative user (e.g. cyrus) I get the quota. I had a look at the code and there is a specific exclusion in cmd.getquota which prevents anyone other than an administrative user from getting the quota. (I am typing this from memory so I may have got the function name wrong). Anyway, the strange thing is that someone earlier showed me them running Netscape 7.x and reading their quota. Now how did it do that? Have I just got the commands wrong? Any ideas would be gratefully received. Best wishes.... Colin Bruce --- 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: MOre problems with DBD
Dear Ken, Thanks for the suggestion. I'll do that tomorrow. I managed to get it to convert all the DBs to skiplist (except subs which are flat files) and it seems to be working a treat. No DB4 or PANIC errors in the log files and going well. I wish I'd never set eyes on BDB but I learn from my mistakes. Best wishes... Colin On Thu, 15 Apr 2004, Ken Murchison wrote: > > > > Does this mean that something is still using BDB and if so how can I > > find out what and how do I get rid of it as well? > > Simply not using BDB for any of the databases won't solve this. You > need to compile Cyrus without support for BDB: --with-bdb=no --- 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: How to get rid of Berkeley DB
Dear Mark, Thanks for the help. Your idea worked a treat. I also had to mess around with the files in the db directory but I worked all that out it seemed to work okay. Just in case anyone else wants to do that (as I've copied this to the list). Here's what I did to convert the seen database from Berkeley DB to skiplist. 1. stop the server 2. run db_recover in var/imap/db 3. rm the log file from var/imap/db and any left over __db files 4. cd var/imap/user 5. run a script which for each seen file in every directory did a db_dump and a db_load 6. repeat steps 2 and 3 at this point 7. run a script which Mark suggested that converts every file from BDB to skiplist. Once all the conversions are complete rename all the skiplist files to replace the seen files. 8. repeat steps 2 and 3 9. now edit impad.conf and set seenstate_db: to skiplist 10. Make sure that the var/imap/db directory contains nothing other than the file called skipstamp. 11. start master and all should be fine. There should be no errors in imapd.log about PANICs or DB errors. This worked for me anyway so perhaps it might be useful for others. I risked it because the seen files could be lost and it wouldn't be a total disaster. Perhaps on other databases it might not be much good. Best wishes. Colin On Thu, 15 Apr 2004, Mark Keasling wrote: > Did you try converting all of the ".seen" berkeley db files to skiplist before > moving them? After stopping cyrus, something like this might work for you: > > cd var/user > for f in */*.seen > do > cvt_cyrusdb `pwd`/$f berkeley `pwd`$f.new skiplist > done > --- 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
MOre problems with DBD
Dear All, Yet another problem with BDB (at least I think it is) On a test system I think I have completely removed BDB. However, I still get these errors showing up in the log files. Apr 15 17:21:27 r118-1 ctl_cyrusdb[20562]: DBERROR db4: PANIC: Invalid argument Apr 15 17:21:27 r118-1 ctl_cyrusdb[20562]: DBERROR: critical database situation Apr 15 17:21:45 r118-1 reconstruct[20565]: DBERROR db4: PANIC: fatal region error detected; run recovery Apr 15 17:21:45 r118-1 reconstruct[20565]: DBERROR: critical database situation Apr 15 17:22:04 r118-1 reconstruct[20569]: DBERROR db4: PANIC: fatal region error detected; run recovery Apr 15 17:22:04 r118-1 reconstruct[20569]: DBERROR: critical database situation Does this mean that something is still using BDB and if so how can I find out what and how do I get rid of it as well? Best wishes Colin --- 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: How to get rid of Berkeley DB
Dear Mark, Thanks for the info. That idea worked on my test system (which is a copy of the live system). I'll try it on the real system tonight when everyone has gone home. There are other problems now as well. I wish I'd never heard of BDB :-( Best wishes... Colin On Thu, 15 Apr 2004, Mark Keasling wrote: > Did you try converting all of the ".seen" berkeley db files to skiplist before > moving them? After stopping cyrus, something like this might work for you: > --- 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
How to get rid of Berkeley DB
Dear All, I am running cyrus 2.2.3 on a linux system but at the end of my tether with it. It is now a live environment with around 20,000 users. At the moment it is down. I have discovered that Berkeley DB is utter rubbish and totally unusable. However, there appears to be no way to get rid of it. I am running the cvt_cyrusdb by hand and it is behaving in a very strange way indeed. I am trying to convert the seen DB. I pick a letter (lets say u) and do cd var/user/u for f in *.seen do cvt_cyrusdb `pwd`/$f berkeley `pwd`$f.new skiplist done after that has finished I have a nice set of files which appear to be skiplist. I then do for f in *.seen do mv $f.new $f done All this is as the user cyrus. Anyway the new seen files are fine. Now I move on to another letter (lets say v) and repeat the exercise I get another set of nice skiplist files. However, when I look back at the files under u they are all now 4096 bytes long and appear to be corrupted BerkeleyDB files. How on earth is this happening? If I try to do another letter, (lets say w) they also work fine but now v is corrupted and so on. Each time I do a directory the previous one gets damaged. Is there anything I can do. If we stick with Berkeley we get several crashes a day with it. If I were writing this stuff I would dump Berkeley straight away. It seems to be the most appalling rubbish I have come across for some time. --- 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
MOre info on the Berkeley problem
When I do the second directory I get the following error messages in the log file. Apr 14 21:44:59 imap2 cvt_cyrusdb[9875]: DBERROR db4: DB_ENV->log_flush: LSN of 2/2099190 past current end-of-log of 1/1426062 Apr 14 21:44:59 imap2 cvt_cyrusdb[9875]: DBERROR db4: Database environment corrupt; the wrong log files may have been removed or incompatible database files imported from another environment Apr 14 21:45:00 imap2 cvt_cyrusdb[9893]: DBERROR db4: DB_ENV->log_flush: LSN of 2/2099190 past current end-of-log of 1/1427846 Apr 14 21:45:00 imap2 cvt_cyrusdb[9893]: DBERROR db4: Database environment corrupt; the wrong log files may have been removed or incompatible database files imported from another environment Apr 14 21:45:00 imap2 cvt_cyrusdb[9911]: DBERROR db4: DB_ENV->log_flush: LSN of 2/2099190 past current end-of-log of 1/1429591 Apr 14 21:45:00 imap2 cvt_cyrusdb[9911]: DBERROR db4: Database environment corrupt; the wrong log files may have been removed or incompatible database files imported from another environment and the files in teh first directory are corrupted. best wishes. Colin --- 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 will not run properly
Dear Lyn, We had a similar problem this afternoon and some people helped out with sugestions. What we did was 1. stop the master server 2. delete files in the /var/imap/db directory 3. restart the master I suspect there might be a need to run utilities such as db_recover after step 1 if it were a live environment but if you are just setting up then that should be okay. I should add that I am new to all this so if someone else says something different take their advice before mine. Hope this helps. Colin On Thu, 8 Apr 2004, Lyn St George wrote: > Hallo all > > I'm in the process of newly installing Cyrus IMAP 2.2.3 > and sasl 2.1.18 on RH Linux, and have two problems > that I'm hoping someone can help with. > > 1. libsasl2.so.2 can't be found - it doesn't exist but > libsasldb.so.* does. I copied one to the other and > this seemed to fix it, but this can't be the correct way. > > 2. /usr/cyrus/bin/master appears to start well enough, though > with lots of DBERRORs due to files not yet created and listed > as missing. However telneting to it on port 143 gets a telnet > connection response (Escape character is '^]'.) but not the > '* OK ..' Cyrus response. And tailing the imapd.log shows > multitudinous instances of > > test master[9578]: service imapd pid 9942 in READY state: terminated abnormally > test master[9943]: about to exec /usr/cyrus/bin/imapd > test imapd[9943]: executed > test master[9578]: process 9943 exited, status 127 > > I see that others have posted similar logs, but can't find > any post with a solution. Is there a known solution to this? > > TIA > > Lyn St George > > > --- > 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: Desperate
On Thu, 8 Apr 2004, Rob Siemborski wrote: > On Thu, 8 Apr 2004, Colin Bruce wrote: > > > Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving log file: > > /usr/local/packages/cyrus/var/db/log.01 > > Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving database file: > > /usr/local/packages/cyrus/var/mailboxes.db > > Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving log file: > > /usr/local/packages/cyrus/var/db/log.01 > > . > > Are you using the skiplist backend for the mailbox list? > > > As a matter of interest how are we supposed to stop cyrus if the machine it is > > running on is being shutdown. I can't find any mention of how to do that in the > > documentation. > > Kill the master process. > Dear Rob, Thanks for the reply. Another person (Roman Hochuli) made a similar suggestion about using skiplist. The thing is we do kill the master process (using kill PID) but that seems to mostly cause some sort of database corruption. Roman said that the DB stuff is not so stable and that skiplist would be better. However, I am not so sure what we are using anymore. The configuration options were: ./configure --prefix=/usr/local/packages/cyrus-2.2.3 --enable-netscapehack \ --with-bdb-libdir=/usr/local/BerkeleyDB.4.2/lib \ --with-bdb-incdir=/usr/local/BerkeleyDB.4.2/include \ --with-sasl=/usr/local/lib/sasl2 --with-perl=/usr/bin/perl \ --with-cyrus-prefix=/usr/local/packages/cyrus-2.2.3/sbin \ --with-service-path=/usr/local/packages/cyrus-2.2.3/sbin \ --sysconfdir=/usr/local/packages/cyrus-2.2.3/etc \ --with-openssl --with-auth=unix and the imapd.conf file is [EMAIL PROTECTED]:/usr/local/packages/cyrus/etc# cat imapd.conf configdirectory: /usr/local/packages/cyrus/var defaultdomain: coventry.ac.uk seenstate_db: berkeley duplicatesuppression: 0 admins: root cyrus sieveusehomedir: 0 allowallsubscribe: 1 allowusermoves: 1 unix_group_enable: 0 autocreatequota: 5120 quotawarn: 80 hashimapspool: 1 sasl_pwcheck_method: saslauthd sasl_auto_transition: 0 lmtp_over_quota_perm_failure: 1 partition-default: /usr/spool/mail/cyrus partition-spool1: /mnt/disk1 partition-spool2: /mnt/disk2 partition-spool3: /mnt/disk3 partition-spool4: /mnt/disk4 partition-spool5: /mnt/disk5 partition-spool6: /mnt/disk6 I haven't seen a skiplist option in ./configure but maybe I missed it. It seems we should switch to using skiplist but how do we go about doing it? Do the databases need to be converted in some way or do we just shutdown and restart? Anyway, enough from me. Thanks for your help and advice. Best wishes Colin --- Home Page: http://asg.web.cmu.edu/cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
Desperate
Dear All, We have been running cyrus for roughly 8 hours in a production environment and it has failed. It will not start at all. It just sits their doing nothing. The following errors are in the log file although these are follow by hundreds of lines similar to this. It is the latest stable version of cyrus that we are trying to run. Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving log file: /usr/local/packages/cyrus/var/db/log.01 Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving database file: /usr/local/packages/cyrus/var/mailboxes.db Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: archiving log file: /usr/local/packages/cyrus/var/db/log.01 Apr 6 13:02:37 imap2 ctl_cyrusdb[2264]: done checkpointing cyrus databases Apr 6 13:02:37 imap2 master[2256]: process 2264 exited, status 0 Apr 6 13:03:23 imap2 imap[642]: DBERROR db4: PANIC: fatal region error detected; run recovery Apr 6 13:03:23 imap2 imap[642]: DBERROR: critical database situation Apr 6 13:03:56 imap2 imap[2255]: unable to tell master 1: Broken pipe Apr 6 13:03:56 imap2 imap[2255]: unable to tell master 2: Broken pipe We have run recovery as instructed but that does nothing. Does anyone know how to fix this? If not does anyone know how difficult it would be to to convert back to UW imap. It seems to be much more stable than cyrus. As a matter of interest how are we supposed to stop cyrus if the machine it is running on is being shutdown. I can't find any mention of how to do that in the documentation. Best wishes. Colin --- Home Page: http://asg.web.cmu.edu/cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
Re: Problem correcting my silly mistake
> What error did you get when you tried the 'dm' command? If you followed > the sequence of steps as you described, I'm betting it was > "deletemailbox: Permission denied" > The problem is by default your 'cyradm' . Dear Dwight, Thanks for the reply. Saddly, it isn't that error. The error was "operating system error" as I recall. I suspect it was me messing around that broke something pretty bad. Anyway, its one way to find out how things work :-) Thanks for the info anyway. Best wishes Colin Bruce --- Home Page: http://asg.web.cmu.edu/cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
Problem correcting my silly mistake
Dear All, We have just started implementing a Cyrus mail system to replace a UW one. Anyway, I've managed to make a silly mistake which we can resolve by installing from scratch. However, I wonder if someone can explain how to solve the problem properly as starting from scratch won't be possible once it is all live. I created a mailbox by typing (in cyradm) cm ccx004 instead of cm user.ccx004 How do I delete it. I tried dm ccx004 and that didn't work. I can't find it on the file system anywhere. When I do lm ccx004 in cyradm it says no such mailbox but when I do quota | grep ccx004 it shows the quota that I set. I've tried doing a reconstruct -r and also a quota -f but neither made any difference. I guess in my many and varied attempts to get rid of it I deleted it from the file system so the easiest thing in this case is to reinstall from scratch. However, my question is how should I deal with such errors in a live system? Best wishes. Colin Bruce --- Home Page: http://asg.web.cmu.edu/cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
