- dovecot - dovecot.org

[Dovecot] dovecot + mysql (complements)
by MAGALLON Lionel 12 Jan '05

12 Jan '05

Hi, I've done some supplementary tests with tcpdump. Apparently, with a login containing the % character, there's no query send to MySQL. Do you know if the % character is filtered by dovecot ? Thanks

2 1

[Dovecot] dovecot + mysql
by MAGALLON Lionel 12 Jan '05

12 Jan '05

Hi, I've got a problem of authenticating users when they've got the % character in their login name. It always returns -ERR Authentication failed. I've done some tests (explained below). 1. the mysql answer to SELECT password FROM users WHERE uid = 'foo%domain.com' is 123 2. the mysql answer to SELECT password FROM users WHERE uid = 'foo(a)domain.com' is 123 3. with a password_query like this : password_query = SELECT password FROM users WHERE uid = '%u' it works fine with foo(a)domain.com but not with foo%domain.com 4. everybody can login with password 123 with a password_query like this : password_query = SELECT password FROM users WHERE uid = 'foo%domain.com' => I don't know what's done with the % character when passed to the %u variable. Can somebody enlight me ? Thanks a lot.

1 0

[Dovecot] link() failed / maildir_copy_with_hardlinks=no / maildir-save.c.patch
by David C. Reeve 12 Jan '05

12 Jan '05

Hi, I have recently been having a small problem with Dovecot (dovecot-0.99.13) not honouring the "maildir_copy_with_hardlinks" config option, specifically when it is set to "no". This results in errors to the syslog looking like: Jan 12 09:59:21 imap imap(dave): link(/afs/.../tmp/1105541958.P1137Q0M41428.imap, /afs/.../new/1105541958.P1137Q0M41428.imap:2,S) failed in maildir_copy: Invalid cross-device link The supplied patch will, when maildir_copy_with_hardlinks=no, rename files inside maildir's instead of hard linking them, hence honouring the configuration option. A word of caution! This all relies on the filenames in maildir being unique (which they should be, and this shouldn't be a problem, but..). It is possible (although unlikely) that renaming the file could destroy a file in the target directory; no checking is done to assure this is not the case! (If you are paranoid you can always write another patch). Yes I am using AFS (openafs specifically) to hold my maildir's, which does not support hard links. However AFS assures that rename operations are atomic so we can use that instead (this may be the case for other filling systems that do not allow hard links). Hope that helps some one else ;) Dave -- David C. Reeve <David.Reeve(a)pnsol.com> Predictable Network Solutions

1 0

[Dovecot] dovecot-auth zombie procs
by jpollara＠lawnchair.net 12 Jan '05

12 Jan '05

Hi, every now and then it seems as if dovecot-auth zombie's out. When this happens nobody can log in. If I kill all the dovecot-auth processes and then restart dovecot everything seems to work again. I'm authenticating against an openldap server on a debian sarge box so I wonder if this has anything to do with it. All my clients are using Outlook. Any help would be greatly appreciated. Thanks. Josh

2 5

[Dovecot] Re: maildir and /var/spool/mail
by C.Lee Taylor 12 Jan '05

12 Jan '05

Greetings ... > Not sure what to do next. Have imap working in that I can see messages > in inbox. When I create a folder it won't create one. Looking at maillog > shows that I don't have rights to folder /var/spool/mail/.sent or > whatever folder I am trying to create. I have procmailrc and > dovecot.conf point to /home/Maildir but for some reason something is > tied to /var/spool/mail. Of course I am running FC3 and sendmail. > Someone have any ideas as to what to try next? I'm not sure of the rest of the history with this, but my FC3 installation has [root@nasrec etc]# cat /etc/procmailrc PATH=/bin:/usr/bin:/usr/bin MAILDIR=$HOME/mail DEFAULT=$HOME/Maildir/ #VERBOSE=on #LOGFILE=/var/log/procmail [root@nasrec etc]# ll /etc/procmailrc -rw-r--r-- 1 root root 109 May 2 2004 procmailrc I do remember having to delete the old mail box of one of my installs, where I had started using dovecot before putting in my procmailrc file ... I stop dovecot, imap* and pop3* ... Removed all the wrong mailboxs both in user folder and in /var/spool/mail ... Ran mail -s Test user Type a message here < Press Control-D here > Message gets delieverd in maildir format, start dovecot and all seemed well. I hope this helps. Mailed Lee

1 0

[Dovecot] Howto host multiple domains
by Mesut Bayrak 11 Jan '05

11 Jan '05

i dont know howto host multiple domains with dovecot are there any way to make it ?

2 1

[Dovecot] can not connect to dovecot using ssl
by Jay Gustafson 11 Jan '05

11 Jan '05

I can connect using imap, but when I try using ssl (imaps) I get a can not connect to server error. I have in my dovecot.conf protocols = imaps imap imap_listen = * imaps_listen = * I DO get a connection when I telnet to port 993 of my server (both localhost and remote work). Any help much appreciated.. Jay

2 1

[Dovecot] Re: delivery agent in 1.0?
by Apostolos Papayanakis 11 Jan '05

11 Jan '05

On Thu, 23 Dec 2004 Geo Carncross wrote: > How about a dovecot-inject tool that writes a message to whatever is > considered the $LOCAL user? It wouldn't have to do much but write the > message (safely) and update indexes -- something dovecot already has to > do. I hope there will soon be a plain delivering tool for dovecot 1.x. I could use it right away as the delivering step with procmail, which would run it under recipient's uid and context (all recipients are real unix users in my case). Without a local delivery agent, dovecot will always seem slower than its rivals when opening INBOXes with new mails, especially large ones, because of index creation delays. A good LMTP implementation has its benefits, but many people have to do without it. -- Apostolis Papayanakis apap(a)ccf.auth.gr, 2310-998416 On Thu, 23 Dec 2004 dovecot-request(a)dovecot.org wrote: > Date: Thu, 23 Dec 2004 11:44:15 -0500 > From: Geo Carncross <geocar-dovecot(a)internetconnection.net> > Subject: Re: [Dovecot] delivery agent in 1.0? > To: Timo Sirainen <tss(a)iki.fi> > Cc: " Branko F. Gra?nar " <bfg(a)noviforum.si>, dovecot(a)dovecot.org > Message-ID: <1103820255.3898.126.camel(a)midget.intranet> > Content-Type: text/plain; charset=iso-8859-2 > > On Thu, 2004-12-23 at 17:52 +0200, Timo Sirainen wrote: > > On 23.12.2004, at 16:06, Branko F. Graθnar wrote: > > > > > Is there any plan to include delivery agent in 1.x? > > > > I'll need to implement one in next few months, so yes. > > > > Hmm. This brings to my mind again that it would need Sieve, and mvmf > > might be a good base for it, but it's still not open source (Mark? :) > > It may be difficult to please everyone. I could have use for [some > elements of] a dovecot-lda but zero use for sieve... > > > > > I would be very happy, if there would be lmtp delivery agent included > > > with dovecot 1.x... This would make creating large mailsystem cluster > > > creation very simple + indexes would be always up2date. > > > > I'm not sure when I'll get around to implementing LMTP server. It's > > actually a bit annoying to implement because it needs to be able to > > deliver mail into multiple accounts, and if each one stores mails with > > different UID, it needs to either run as root (bad) or launch multiple > > processes and do IPC with them. > > How about a dovecot-inject tool that writes a message to whatever is > considered the $LOCAL user? It wouldn't have to do much but write the > message (safely) and update indexes -- something dovecot already has to > do. > > That way: > > * People who do delivery without LMTP baggage aren't treated as second- > class citizens. > > * Sieve implementations merely have to treat "keep" and "fileinto" as > subprocesses. > > * LMTP implementations can do the work they've already have to do > (regarding mapping email addresses to local users) and simply hook into > the sieve processor or the dovecot-inject, or anything else the > administrator/user likes. > > Formally, dovecot-inject would need the following information: > > 1. the envelope recipient ($RECIPIENT) > 2. the envelope sender/return-path/errors-to-address ($SENDER) > 3. the IMAP folder to write the message to (argv[1]) > > On systems with shared-uids, dovecot-inject would also need the concept > of the "localpart" of the address thats applicable. dovecot already > needs to know this in order to ACCESS the maildrop. ($LOCAL) > > This would keep things very easy to integrate without resorting to > shared libraries, "plugin system" nonsense, or forcing people to make > extra levels of indirection (lmtpclient) in order to benefit from pre- > login indexing... > > > > I suppose I'd implement it so that the actual LMTP talking would be > > done in a process very similar to imap/pop3-login processes, ie. > > completely unprivileged. Then it would tell master process to launch > > delivery agent processes to specific users, and those delivery agents > > then would get the mail from lmtp process via .. um. read-only shared > > memory + semaphores might be better than just copying the data over > > pipes/sockets? > > blah blah blah... > >

2 2

[Dovecot] 1.0-test60
by Timo Sirainen 11 Jan '05

11 Jan '05

http://dovecot.org/test/ I've been running this for two hours now without problems. I guess it's pretty stable. - Keyword support finally included. Not too well tested, but I couldn't get it to break. Doesn't store the keywords into maildir/mbox yet. Hopefully soon. - Major reorganization of code in dovecot-auth, and not yet finished. Load balancing between multiple auth procsses (ie. auth_count > 1) was before done by connecting to each one's socket separately. Now there's a single socket which a balancer process listens to and distributes the requests to worker processes. What's left to do is make a single auth process support handling multiple auth sections in config file. This makes it possible to: - have ordering-sensitive fallbacking for multiple passdbs - have fallbacking work with non-plaintext authentication - allow for all kinds of new great features more easily. For example we could add PAM-like "required" vs. "sufficient" selection for auth sections, or some rules when each passdb is used (eg. based on username, domain or IP range). After that it should also be easy to run dovecot-auth standalone because it would read its own config file. I'm also planning on making it compatible with Postfix's master process, so you could directly add it to master.cf (actually I think I only have to move one fd from 3 to 6 and it'd work). - Added "passdb" userdb. Basically means that the passdb already returned all the information needed for userdb. With SQL this means you could do: SELECT password, uid AS userdb_uid, gid AS userdb_gid, home AS userdb_home FROM ... - If authentication failed because of some internal error, tell it to client instead of showing a generic "Authentication failed." error. - MySQL driver supports load balancing between multiple servers. Just add a new host=xx to connect string. It's not currently supported to have different usernames/passwords. Probably not worth it? - rawlog code rewritten. Should work better. - If userdb returns non-numeric UID or GID, look it up from passwd/group. - Rewritten message address parser. The old code could have lost some spaces and might have even returned invalid MIME string because of it. This affected only clients which use FETCH ENVELOPE command. - mbox: If lines began with NUL characters, things broke. Also fixed another potential problem with parsing mboxes. As a result hopefully these fix a problem with expunge sometimes leaving a few extra characters to beginning of mbox. - Added mail_debug setting which logs where and how Dovecot is trying to locate mails. Helpful when initially figuring out why Dovecot isn't working right. auth_debug=yes now also logs all SQL queries and LDAP lookups and some other things. - etc.

3 2

[Dovecot] IMAP crashes with signal 10 (SIGBUS)
by Joseph Tam 11 Jan '05

11 Jan '05

I wrote: > Any idea on what's causing this? > > Description: > imap process crashed with SIGBUS. > > Versions: > OS: Solaris 8, 9 > Dovecot: 1.0-test57 > Gcc: 3.2 same with 3.4.3 > > Backtrace: > #0 mbox_file_seek (ibox=0x8d9b0, view=0x8dc38, seq=2, deleted_r=0xffbef40c) at mbox-file.c:119 > ... Subsequent versions still crashed and I looked into the problem further. Specifically it crashes at this line offset = *((const uint64_t *)data); when "data" assumes a non-64 bit aligned value. At first I thought that 64-bit ints should always be aligned on 64-bit boundaries, but when I isolated this code into a standalone executable to demonstrate it (no arguments to cause a crash, one argument is OK): int main (int argc, char *argv[]) { int a[]={1,2,3,4}; long long int i64; void *p=(void *) (&a[argc]);; printf("Pointer to 32-bit int = %x\n",p); i64 = *((long long int *)p); printf("Successful casting to 64-bit int: value=%016llx\n",i64); } I found out that the presence of any optimisation (-O, -O2, -O3) will cause this to crash, yet it will work if no optimisation is present even though a pointer being cast is still not aligned on a 64-bit boundary (but aligned on a 32-bit boundary). I check the gcc bug web page page came up with this reference which may explain this behaviour. http://gcc.gnu.org/bugs.html#known - see section "Casting does not work as expected when optimization is turned on" http://mail-index.netbsd.org/tech-kern/2003/08/11/0001.html They claim that this casting violates ISO aliasing rules and the results like crashing are to be expected. Maybe this statement ought to be rewritten memcpy(&offset,data,sizeof(uint64_t)); There is one other statement at mbox-synbc.c:319 (1.0-test59) that may need this consideration. Joseph Tam <tam(a)math.ubc.ca>

2 2