- dovecot - dovecot.org

[Dovecot] still has error
by Farkas Levente 28 May '03

28 May '03

hi, after download the current cvs recompile, start I still has these errors: ------------------------------------- imap-login: May 28 13:49:02 Info: Login: jcsinalos [192.168.0.106] imap(jcsinalos): May 28 13:49:02 Panic: unreached dovecot: May 28 13:49:02 Error: child 31683 (imap) killed with signal 6 imap-login: May 28 13:49:03 Info: Login: jcsinalos [192.168.0.106] imap-login: May 28 13:49:03 Info: Login: jcsinalos [192.168.0.106] imap(jcsinalos): May 28 13:49:03 Panic: unreached dovecot: May 28 13:49:03 Error: child 31687 (imap) killed with signal 6 ------------------------------------- -- Levente "Si vis pacem para bellum!"

2 3

[Dovecot] 0.99.10-test7
by Timo Sirainen 28 May '03

28 May '03

- APPEND memory leak fixed - maildir syncing crash / some other errors fixed - pop3 mail fetching fixed - pop3 didn't advertise STLS in capabilities before http://dovecot.procontrol.fi/test/

1 0

[Dovecot] a few more error
by Farkas Levente 28 May '03

28 May '03

in the imap.log: imap(arvays): May 28 10:39:32 Error: Timeout while waiting for release of fcntl() lock for index file /home/arvays/Maildir/.INBOX/.imap.index imap(lfarkas): May 28 10:42:05 Panic: unreached dovecot: May 28 10:42:07 Error: child 19933 (imap) killed with signal 6 -- Levente "Si vis pacem para bellum!"

1 0

[Dovecot] serious problem
by Farkas Levente 28 May '03

28 May '03

hi, I've got dozen such errors: ---------------------------- imap-login: May 28 10:28:56 Info: Login: gpajlocz [192.168.1.182] imap(gpajlocz): May 28 10:28:56 Panic: file maildir-sync.c: line 472 (maildir_sync_uidlist): assertion failed: (ACTION(hash_rec) == MAILDIR_FILE_ACTION_NEW) dovecot: May 28 10:28:56 Error: child 17207 (imap) killed with signal 6 ---------------------------- and this users can't get into the system... -- Levente "Si vis pacem para bellum!"

1 0

[Dovecot] Dovecot 0.99.9.1 and openssl-0.9.7a-5
by Robin Whittle 28 May '03

28 May '03

Hi Timo and others, Thanks for Dovecot! I just got it running under Red Hat 9.0 after spending many hours unsuccessfully trying to get Courier IMAP to run, although I have done it in the past. Here is a compilation glitch which may already be well known: I compiled dovecot-0.99.9.1 on my newly installed and recently updated (via Red Hat Network) system and the compilation stopped due to a missing file in the SSL headers: In file included from /usr/include/openssl/ssl.h:179, from ssl-proxy-openssl.c:14: /usr/include/openssl/kssl.h:72:18: krb5.h: No such file or directory In file included from /usr/include/openssl/ssl.h:179, from ssl-proxy-openssl.c:14: The currently installed OpenSSL RPM was openssl-0.9.7a-5. This lacks the file /usr/include/openssl/krb5.h . I understand that openssl-0.9.7b is OK, but I wasn't sure about whether installing this as a tarball from http://www.openssl.org would upset any other programs on this machine, or confuse the RHN update system. I didn't really need SSL and I couldn't see a specific option for disabling it, other than a general instruction about disabling modules with a ./configure option. So I tried ./configure --help and compiled it with: ./configure --without-ssl This worked fine. I really like the report at the end of the ./configure stage saying how the program will be compiled! - Robin // Robin Whittle http://www.firstpr.com.au // Melbourne, Australia http://fondlyandfirmly.com

1 0

[Dovecot] Dovecot in OpenBSD ports system
by Andrew Basterfield 27 May '03

27 May '03

Hi I have put together dovecot port for the OpenBSD ports system. The port has yet to be submitted, but is finished awaiting testing from a wider audience. I have been using dovecot successfully on OpenBSD/sparc64 -current for months now with no problems. Would you recommend including the CR+LF patch? http://cemetery.homeunix.org/OpenBSD/dovecot.tar.gz -- http://cemetery.homeunix.org

2 6

RE: [Dovecot] Postgres "issue"
by Maikel Verheijen 27 May '03

27 May '03

Ok: a little lame to answer my own question, but I created a CRUDE patch. It does NOT check why a connection failed, it just does a PQreset, and it SEEMS to work. Kind regards, Maikel Verheijen. Ps: Timo, can something with reconnect be implemented permanently? --8<-- --- src/auth/db-pgsql.c.ORIG Tue May 27 10:06:39 2003 +++ src/auth/db-pgsql.c Tue May 27 11:16:38 2003 @@ -53,9 +53,16 @@ res = PQexec(conn->pg, query); if (PQresultStatus(res) != PGRES_TUPLES_OK) { - i_error("PGSQL: Query \"%s\" failed: %s", - query, PQresultErrorMessage(res)); - failed = TRUE; + PQreset(conn->pg); + res = PQexec(conn->pg, query); + if (PQresultStatus(res) != PGRES_TUPLES_OK) { + i_error("PGSQL: Query \"%s\" failed: %s", + query, PQresultErrorMessage(res)); + failed = TRUE; + } else { + i_error("PGSQL: Reconnected because of failure!"); + failed = FALSE; + } } else { failed = FALSE; } --8<-- -----Original Message----- From: Maikel Verheijen [mailto:maikel@ladot.com] Sent: Tuesday, May 27, 2003 9:43 AM To: 'Dovecot (E-mail)' Subject: [Dovecot] Postgres "issue" Dear list, I have a small issue with dovecot talking to a postgresql server: If my postgres server get restarted, dovecot fails to reconnect to the postgres server. I run dovecot 0.99.9.1 (which is actually a cvs version of dovecot, of may 9 2003) I have the same postgres version for the server as I use for the libraries. Is it hard to get auto-reconnect to work? Kind regards, Maikel Verheijen Ps: the log message I get is this: May 22 17:33:01 testlfw3 dovecot-auth: PGSQL: Query "SELECT password FROM aliases WHERE alias = 'maikel(a)test.com' AND password != ''" failed: FATAL 1: This connection has been terminated by the administrator. server closed the connection unexpectedly ^IThis probably means the server terminated abnormally ^Ibefore or while processing the request. Which probably describes it all :)

1 0

[Dovecot] still problem with load
by Farkas Levente 27 May '03

27 May '03

hi, we've got (and getting more) serious problem with load. I'm just calculate that ther is about 300 mailbox with 20Gb mails. and the whole system is _very_ slow and just getting slower and slower. it's a fast lan every client has 100Mb connection to the server with gigabit crad and gigabit switches. when I push the delete button in a mail I usualy have to wait 10,20 or 30 seconds (!!!) to deleted and show me the next one. dovecot eat so much io to the hard disk, that we can't do too much. currently at the same time about 100 clients are connected mostly with OE6. the server is a P4 server with 1GB ram and about 1TB hard disk. there is only one samba and dovecot on the server. and the load use to be something like this: ---------------------------------------- 10:47:38 up 3 days, 15:48, 2 users, load average: 4.64, 3.58, 3.43 230 processes: 225 sleeping, 5 running, 0 zombie, 0 stopped CPU states: 6.3% user 8.9% system 0.0% nice 0.0% iowait 84.6% idle Mem: 1022840k av, 1013588k used, 9252k free, 0k shrd, 80320k buff 826372k actv, 47324k in_d, 16904k in_c Swap: 4193608k av, 80612k used, 4112996k free 783088k cached ---------------------------------------- as you can see it use almost all ram (as the 2.4 kernmel used to do) but no cache. so the ram is enough. the more interesting part is the load while the idle usualy between 80-90% the whole system doesn't use any cpu (or just minimal). BUT at the same time the load is around 3,4,5 (!!!) which is too much, and more than acceptable. what's worse the reason is the io load on the hard disks. every process are waiting for io. in this case I usualy service dovecot stop killall imap <wait about one minutes> service dovecot start and the load goes back to 0.3-0.6. after an hour the same happend and I have to do the same.:-((( what's more in this case samba case that all (!!!) clients stops for a few seconds. since everybody use OE6 they always download all messages with huge mailboxes what's more dovecot index the whole system all the time. I assume this cause the load. IMHO 300 mailbox and 100 concurrent user is not such a big thing. is there any way to restrict dovecot not to use so much io? thanks in advance. -- Levente "Si vis pacem para bellum!"

3 8

[Dovecot] Postgres "issue"
by Maikel Verheijen 27 May '03

27 May '03

Dear list, I have a small issue with dovecot talking to a postgresql server: If my postgres server get restarted, dovecot fails to reconnect to the postgres server. I run dovecot 0.99.9.1 (which is actually a cvs version of dovecot, of may 9 2003) I have the same postgres version for the server as I use for the libraries. Is it hard to get auto-reconnect to work? Kind regards, Maikel Verheijen Ps: the log message I get is this: May 22 17:33:01 testlfw3 dovecot-auth: PGSQL: Query "SELECT password FROM aliases WHERE alias = 'maikel(a)test.com' AND password != ''" failed: FATAL 1: This connection has been terminated by the administrator. server closed the connection unexpectedly ^IThis probably means the server terminated abnormally ^Ibefore or while processing the request. Which probably describes it all :)

1 0

[Dovecot] pop3 timeout when sending first message
by Mike Brown 27 May '03

27 May '03

I'm using FreeBSD 4.8 with dovecot from the ports collection. First I am first trying to fetch mail via POP3. Clients are both OE6 and Mozilla. Both exhibit the same behavior: the client says it is receiving message 1 of (however many), but nothing happens; no message data is transferred. Mozilla seems to keep waiting forever, but OE6 times out after 60 seconds and lets me stop waiting. I tried both with and without SSL. Authentication works fine in either case, but the mail xfer just doesn't work. There's nothing of interest in the logs. By the way, there seems to be no mention of what syslog facility is used by dovecot. What is it? I'm avoiding syslog for now and writing to specific files. In the info log I get this when the client logs in: pop3-login: May 25 23:10:44 Info: Login: mike [my.IP.addr.here] In the other log I get this when OE6 disconnects, following the timeout and prompting me to wait or stop: pop3-login: May 25 23:11:53 Warning: SSL_read() syscall failed: EOF [216.241.45.45] I'm not sure what else to do to try to debug this... Help? -Mike

2 4

[Dovecot] Why dovecot is not multi-threaded?
by Ruben Alvarez 26 May '03

26 May '03

Hi, I'm looking for the most efficient OpenSource IMAP server and I am told that Dovecot Server is a good solution. Anyway, I have inspected the source code and I think Dovecot is not a multi-threaded server, but a forked server. I am surprised about this issue, because threads perfomance should be higher than creating a new process for each connection....I suppose. So, Could anybody tell me why Dovecot was not designed with multi-threaded support?. Perhaps, threads are not a good solution for Dovecot but I am very interested in which are the reasons for this. By the other hand, ¿do anybody know any multi-threaded OpenSource IMAP server?. I would like test performance of Dovecot server versus a multi-threaded one. Very thanks.

2 1

[Dovecot] logging in POP3
by Jens Laas 26 May '03

26 May '03

I have begun testing dovecot as (primarly) a POP replacement. So far the only lacking bit is the logging. It would be great if pop3 could log a line at client exit with the following info: (wishlist follows :) o nr deleted mails + total bytes deletes o nr mails left in spool + bytes left o time spent o indication if TLS was used or not If you would lika generic logging API I have one you can use: (Only tested under Linux. It is GPL) http://robur.slu.se/jensl/jellog.tar.gz Excerpt from headerfile: /* Functions for setting the destination of logger_facilities. - filename is the file to append log entries to. If file does not exist it is created with umask 0666 (modified as usual by open() call). - filestream is the FILE stream to use. - fd is the filedescriptor to use. - priority is the syslog priority to use as used in the syslog() call. - timestamp is used as a boolean. True means begin line with a timestamp. */ int logger_set_file(logger_facility facility, char *filename, int timestamp); int logger_set_iostream(logger_facility facility, FILE *filestream, int timestamp); int logger_set_fd(logger_facility facility, int fd, int timestamp); int logger_set_syslog(logger_facility facility, int priority, int timestamp); /* fmt = [sdt]+ s = string d = integer t = time-string Example: logger( LOGGER_ERROR, "sds", "Errorcode: ", -1234, ". Aborting!"); See end of logger.c for more examples. */ int logger(logger_facility facility, char *fmt, ...); Dovecot looks great! Just thought I'd send you this wishlist. If you could point me to the right places in the code I am willing to help adding this myself. Cheers, Jens Låås ----------------------------------------------------------------------- 'This mail automatically becomes portable when carried.' ----------------------------------------------------------------------- Jens Låås Email: jens.laas(a)data.slu.se Department of Computer Services, SLU Phone: +46 18 67 35 15 Vindbrovägen 1 P.O. Box 7079 S-750 07 Uppsala SWEDEN -----------------------------------------------------------------------

2 1

[Dovecot] possible dovecot problem (fwd)
by Mark E. Mallett 23 May '03

23 May '03

I'm just forwarding this verbatim from somebody who works here. Comments ? mm ----- Forwarded message from Dave Chamberlain <davec(a)mv.mv.com> ----- > Subject: possible dovecot problem > From: "Dave Chamberlain" <davec(a)mv.mv.com> > Date: Tue, 20 May 2003 21:29:48 -0400 (EDT) > To: mem(a)mv.mv.com (Mark E. Mallett) > > Ok, I'm kind of confused at what dovecot's imap implementation is doing. > > I seem to be spending more time than I need to correctly decipher the headers of > email messages. > > I have included here the text from 7.4.2 as it explains the response from a > fetch of an ENVELOPE. The last paragraph specifies that the FROM, SENDER > and REPLYTO fields cannot be nil because some of them must be present per > RFC822 and the IMAP spec will default values. I have many examples where > this is not the case, where the fetch of the envelope is returning NIL for > some of these values. > > In the example below, the message shows that the from, replyto and sender > are all set to jaentc(a)aukware.com (ok, a spelling error, but that was her > typing it in wrong). The next field after those three is NIL, that is the > TO line (more of my text after the IMAP conversation) > > a uid fetch 902 ENVELOPE > * 113 EXISTS > * 1 RECENT > * 104 FETCH (UID 902 ENVELOPE ("Tue, 20 May 2003 16:05:04 -0500" "From A Friend: > 'TheWMURChannel.com - News - Law Raises Age Children Must Be In Seat Restraints'" > (("Janet" NIL "jaentc" "aukware.com")) > (("Janet" NIL "jaentc" "aukware.com")) > (("Janet" NIL "jaentc" "aukware.com")) NIL NIL NIL NIL > "<200305202105.h4KL5xo20860(a)qbert.ibsys.com>")) > > Following is the complete header from the message. There is a valid To address > supplied, and it is in one of the allowed formats. I would have thought that > possibly it was the format that was causing the problem. I first theorized that > it didn't understand the emailaddress (personalname) format. However, the > From field is in that format, and was correctly picked up (notice that "Janet" > exists above as the Personal name, so it did understand that format). > > I have found multple cases where the FROM was NIL, the SUBJECT was NIL and > the TO is NIL, yet when looking at the header they are visible. > > This is causing problems with more than just my things, OE is having a problem > with it as well. > > > > > >From jaentc(a)aukware.com Tue May 20 21:26:43 2003 UTC > Received: (qmail 24772 invoked from network); 20 May 2003 17:52:17 -0400 > Received: from iridium.mv.net (HELO mv.mv.com) (qmailr(a)199.125.85.17) > by mercury.mv.net with SMTP; 20 May 2003 17:52:17 -0400 > X-Peer-Info: remote-ip 199.125.85.17 local-ip 199.125.85.40 > local-name mercury.mv.net > Received: (qmail 21564 invoked by uid 134); 20 May 2003 17:26:53 -0400 > Delivered-To: davec(a)mv.com > Received: (qmail 3533 invoked from network); 20 May 2003 17:06:26 -0400 > Received: from qbert.ibsys.com (66.187.197.24) > by iridium.mv.net with SMTP; 20 May 2003 17:06:26 -0400 > X-Peer-Info: remote-ip 66.187.197.24 local-ip 199.125.85.17 > local-name iridium.mv.net > Received: from 10.0.0.24 (cfusion3.ibsys.com [10.3.0.53]) > by qbert.ibsys.com (Switch-2.1.0/Switch-2.1.0) with ESMTP id h4KL5xo20860 > for <davec(a)mv.com (Dave)> Tue, 20 May 2003 16:05:59 -0500 (CDT) > Message-Id: <200305202105.h4KL5xo20860(a)qbert.ibsys.com> > Content-type: text/plain > Date: Tue, 20 May 2003 16:05:04 -0500 > From: jaentc(a)aukware.com (Janet) > Subject: From A Friend: 'TheWMURChannel.com - News - Law Raises Age Children Must Be > In Seat Restraints' > To: davec(a)mv.com (Dave) > > > Section 7.4.2 > > ENVELOPE > A parenthesized list that describes the envelope structure of a > message. This is computed by the server by parsing the > [RFC-2822] header into the component parts, defaulting various > fields as necessary. > > The fields of the envelope structure are in the following > order: date, subject, from, sender, reply-to, to, cc, bcc, > in-reply-to, and message-id. The date, subject, in-reply-to, > and message-id fields are strings. The from, sender, reply-to, > to, cc, and bcc fields are parenthesized lists of address > structures. > > An address structure is a parenthesized list that describes an > electronic mail address. The fields of an address structure > are in the following order: personal name, [SMTP] > at-domain-list (source route), mailbox name, and host name. > > [RFC-2822] group syntax is indicated by a special form of > address structure in which the host name field is NIL. If the > mailbox name field is also NIL, this is an end of group marker > (semi-colon in RFC 822 syntax). If the mailbox name field is > non-NIL, this is a start of group marker, and the mailbox name > field holds the group name phrase. > > If the Date, Subject, In-Reply-To, and Message-ID header lines > are absent in the [RFC-2822] header, the corresponding member > of the envelope is NIL; if these header lines are present but > empty the corresponding member of the envelope is the empty > string. > > Note: some servers may return a NIL envelope member in the > "present but empty" case. Clients SHOULD treat NIL and > empty string as identical. > > Note: [RFC-2822] requires that all messages have a valid > Date header. Therefore, the date member in the envelope can > not be NIL or the empty string. > > Note: [RFC-2822] requires that the In-Reply-To and > Message-ID headers, if present, have non-empty content. > Therefore, the in-reply-to and message-id members in the > envelope can not be the empty string. > > If the From, To, cc, and bcc header lines are absent in the > [RFC-2822] header, or are present but empty, the corresponding > member of the envelope is NIL. > > If the Sender or Reply-To lines are absent in the [RFC-2822] > header, or are present but empty, the server sets the > corresponding member of the envelope to be the same value as > the from member (the client is not expected to know to do > this). > > Note: [RFC-2822] requires that all messages have a valid > From header. Therefore, the from, sender, and reply-to > members in the envelope can not be NIL. ----- End forwarded message -----

2 7

[Dovecot] Dovecot and SquirrelMail
by Ken Wilson 22 May '03

22 May '03

Hello, Does anyone know how to get dovecot working with squirrelmail v1.4? We have an existing uw-imap setup, with mail stored in ~/mail and inboxes in /var/spool/mail/user. I've tried setting $imap_server_type = other in squirrelmail, but I do not see any mail in my inbox. This is my setting in dovecot.conf default_mail_env = maildir:~/mail/:INBOX=/var/spool/mail/%n Any help would be grealty appreciated, as I want to leave uw-imap and get the speed increases that dovecot offers. Thanks Ken All e-mail from Wilberforce College is governed by our legal disclaimer which is available at http://webmail.wilberforce.ac.uk/disclaimer.txt

3 2

[Dovecot] ssl still not working
by Farkas Levente 21 May '03

21 May '03

hi, I'm just download the latest cvs and try to use imaps. in mozilla I've got the following message window: ----------------------------- mail.int.bppiac.hu received a message with incorrect Message Authentication Code. If the error occurs frequently, contact the website administrator. ----------------------------- and there is only one OK button:-) and this happens always. what can be the reason and what can I do? thanks. -- Levente "Si vis pacem para bellum!"

6 11

[Dovecot] SSL problems
by Timo Sirainen 21 May '03

21 May '03

Anyone with SSL problems, please try 0.99.10-test4 or included patch. If it fixes it, I'll make it faster by caching the generated keys. http://dovecot.procontrol.fi/test/

2 1

[Dovecot] dovecot with ssl on rhl9
by Elliot Peele 20 May '03

20 May '03

Has anyone gotten dovecot to work on Red Hat Linux 9 with ssl support? It seems to work just fine on Red Hat Linux 8.0, but doesn't seem to work Red Hat Linux 9. When I try to use openssl to connect to the server i get: [elliot@localhost elliot]$ openssl s_client -connect rktest.physics.ncsu.edu:993 CONNECTED(00000003) depth=0 /OU=IMAP server/CN=imap.example.com/emailAddress=postmaster@example.com verify error:num=18:self signed certificate verify return:1 depth=0 /OU=IMAP server/CN=imap.example.com/emailAddress=postmaster@example.com verify return:1 20949:error:140943FC:SSL routines:SSL3_READ_BYTES:sslv3 alert bad record mac:s3_pkt.c:1052:SSL alert number 20 20949:error:140790E5:SSL routines:SSL23_WRITE:ssl handshake failure:s23_lib.c:226: I'm trying to use evolution 1.3.3 to connect via imaps. Whenever I click on the mail folder it just says 'Scanning folders in "IMAP server imap.example.com" (100%)' and then nothing happens. Anyone have any ideas? Thanks Elliot

1 0

RE: [Dovecot] 0.99.10-test1
by Maikel Verheijen 19 May '03

19 May '03

> > Does this mean you made the .subscriptions nfs-safe already in this > > version? > Well, no, but that's not maildir-specific code :) Ah, ok, My mistake! :) Regards, Maikel.

1 0

[Dovecot] 0.99.10-test1
by Timo Sirainen 19 May '03

19 May '03

http://dovecot.procontrol.fi/test/ Maildir syncing now works pretty much as described in my earlier mail. It was quite a large change, so please test this :) So the new and great features: - It's faster. - If someone renames maildir file just before we try to use it, we now resync the maildir and try to find it again. Before we just failed. It's also possible that the filename changed again while we resynced, so we try max. 10 times before failing (yes, we know if we should try resyncing so we don't always scan it 10 times). - Read-only maildirs work now. If we don't have write access to dovecot-uidlist, we'll show only those mails which are already there. - Out of quota/disk space failures are handled well now. All except updating dovecot-uidlist file which is a bit tricky. - :CONTROL=dir in default_mail_env can specify where to save dovecot-uidlist and .customflags files. Useful for above problems. I think maildir support is getting near perfect now. Only thing left is to deal with failure to write dovecot-uidlist when out of quota. Oh, and NFS safe .customflags file.

2 1

RE: [Dovecot] 0.99.10-test1
by Maikel Verheijen 19 May '03

19 May '03

Hi Timo, > I think maildir support is getting near perfect now. Only > thing left is to deal with failure to write dovecot-uidlist > when out of quota. Oh, and NFS safe .customflags file. Does this mean you made the .subscriptions nfs-safe already in this version? Kind regards, Maikel Verheijen.

2 1

[Dovecot] Maildir synchronization
by Timo Sirainen 17 May '03

17 May '03

I thought I'd send this here too, since I spent so much time writing it as comment to maildir-sync.c :) Now only thing left to do is to write the code to match it. /* Here's a description of how we handle Maildir synchronization and it's problems: We want to be as efficient as we can. The most efficient way to check if changes have occured is to stat() the new/ and cur/ directories and uidlist file - if their mtimes hasn't changed, there's no changes and we don't need to do anything. Problem 1: Multiple changes can happen within a single second - nothing guarantees that once we synced it, someone else didn't just then make a modification. Such modifications wouldn't get noticed until a new modification occured later. Problem 2: Syncing cur/ directory is much more costly than syncing new/. Moving mails from new/ to cur/ will always change mtime of cur/ causing us to sync it as well. Problem 3: We may not be able to move mail from new/ to cur/ because we're out of quota, or simply because we're accessing a read-only mailbox. MAILDIR_SYNC_SECS ----------------- Several checks below use MAILDIR_SYNC_SECS, which should be maximum clock drift between all computers accessing the maildir (eg. via NFS), rounded up to next second. Our default is 1 second, since everyone should be using NTP. Note that setting it to 0 works only if there's only one computer accessing the maildir. It's practically impossible to make two clocks _exactly_ synchronized. cur directory ------------- We have maildir_cur_dirty variable which is set to cur/ directory's mtime when it's >= time() - MAILDIR_SYNC_SECS and we _think_ we have synchronized the directory. When maildir_cur_dirty is non-zero, we don't synchronize the cur/ directory until a) cur/'s mtime changes b) opening a mail fails with ENOENT c) time() > maildir_cur_dirty + MAILDIR_SYNC_SECS This allows us to modify the maildir multiple times without having to sync it at every change. The sync will eventually be done to make sure we didn't miss any external changes. The maildir_cur_dirty is set when: - we change message flags - we expunge messages - we move mail from new/ to cur/ - we sync cur/ directory and it's mtime is >= time() - MAILDIR_SYNC_SECS It's unset when we do the final syncing, ie. when mtime is older than time() - MAILDIR_SYNC_SECS. new directory ------------- If new/'s mtime is >= time() - MAILDIR_SYNC_SECS, always synchronize it. maildir_cur_dirty-like feature might save us a few syncs, but that might break a client which saves a mail in one connection and tries to fetch it in another one. new/ directory is almost always empty, so syncing it should be very fast anyway. Normally we move all mails from new/ to cur/ whenever we sync it. If it's not possible for some reason, we set maildir_have_new flag on which instructs synchronization to check files in new/ directory as well. maildir_keep_new flag is also set which instructs syncing to not even try to move mails to cur/ anymore. If client tries to change a flag for message in new/, we try to rename() it into cur/. If it's successful, we clear the maildir_keep_new flag so at next sync we'll try to move all of them to cur/. When all of them have been moved, maildir_have_new flag is cleared as well. Expunges will also clear maildir_keep_new flag. If rename() still fails because of ENOSPC or EDQUOT, we still save the flag changes in index with dirty-flag on. When moving the mail to cur/ directory, or when we notice it's already moved there, we apply the flag changes to the filename, rename it and remove the dirty flag. If there's dirty flags, this should be tried every time after expunge or when closing the mailbox. broken clients -------------- Originally the middle identifier in Maildir filename was specified only as <process id>_<delivery counter>. That however created a problem with randomized PIDs which made it possible that the same PID was reused within one second. So if within one second a mail was delivered, MUA moved it to cur/ and another mail was delivered by a new process using same PID as the first one, we likely ended up overwriting the first mail when the second mail was moved over it. Nowadays everyone should be giving a bit more specific identifier, for example include microseconds in it which Dovecot does. There's a simple way to prevent this from happening in some cases: Don't move the mail from new/ to cur/ if it's mtime is >= time() - MAILDIR_SYNC_SECS. The second delivery's link() call then fails because the file is already in new/, and it will then use a different filename. There's a few problems with this however: - it requires extra stat() call which is unneeded extra I/O - another MUA might still move the mail to cur/ - if first file's flags are modified by either Dovecot or another MUA, it's moved to cur/ (you _could_ just do the dirty-flagging but that'd be ugly) Because this is useful only for very few people and it requires extra I/O, I decided not to implement this. It should be however quite easy to do since we need to be able to deal with files in new/ in any case. It's also possible to never accidentally overwrite a mail by using link() + unlink() rather than rename(). This however isn't very good idea as it introduces potential race conditions when multiple clients are accessing the mailbox: Trying to move the same mail from new/ to cur/ at the same time: a) Client 1 uses slightly different filename than client 2, for example one sets read-flag on but the other doesn't. You have the same mail duplicated now. b) Client 3 sees the mail between Client 1's and 2's link() calls and changes it's flag. You have the same mail duplicated now. And it gets worse when they're unlink()ing in cur/ directory: c) Client 1 changes mails's flag and client 2 changes it back between 1's link() and unlink(). The mail is now expunged. d) If you try to deal with the duplicates by unlink()ing another one of them, you might end up unlinking both of them. So, what should we do then if we notice a duplicate? First of all, it might not be a duplicate at all, readdir() might have just returned it twice because it was just renamed. What we should do is create a completely new base name for it and rename() it to that. If the call fails with ENOENT, it only means that it wasn't a duplicate after all. */

1 0

[Dovecot] Postfix locking equivalence?
by Warren Togami 17 May '03

17 May '03

/etc/postfix/main.cf.default: mailbox_delivery_lock = flock, dotlock Is flock equivalent to dovecot's fnctl locking? For mbox what type of locking settings should I be using? (No NFS, only local filesystem.) I'm using procmail to deliver mail to my mboxes. Does the above postfix setting effect procmail's locking method? I use rules like this: :0 * ^X-BeenThere: dovecot(a)procontrol.fi DOVECOT If I understand this correctly, :0: in procmailrc would explicitly use dotlocks instead of "kernel-locks" as procmail calls it. I'm not fully understanding all this locking stuff. Warren Togami warren(a)togami.com p.s. Timo, I received your test message after my prior posting attempt bounced. I had this error message: Reporting-MTA: dns; danu.procontrol.fi Arrival-Date: Sat, 17 May 2003 12:52:08 +0300 (EEST) Final-Recipient: rfc822; dovecot(a)procontrol.fi Action: failed Status: 5.0.0 Diagnostic-Code: X-Postfix; mail forwarding loop for dovecot(a)procontrol.fi

2 1

[Dovecot] test
by Timo Sirainen 17 May '03

17 May '03

..is this thing working.. I messed up a bit when adding spamassassin to postfix filter.

1 0

[Dovecot] Postfix locking equivalence?
by Warren Togami 17 May '03

17 May '03

/etc/postfix/main.cf.default: mailbox_delivery_lock = flock, dotlock Is flock equivalent to dovecot's fnctl locking? For mbox what type of locking settings should I be using? (No NFS, only local filesystem.) Warren Togami warren(a)togami.com

1 0

[Fwd: Re: [Dovecot] help: The current command did not succeed. The mail server responded: Invalid mask.]
by Timo Sirainen 16 May '03

16 May '03

Just FYI, I guess this message was meant here :)

1 0