- dovecot - dovecot.org

[Dovecot] Hardlinks in copy
by Andreas Aardal Hanssen 12 Aug '03

12 Aug '03

Hi, Dovecotters. In the Maildir code, dovecot-0.99.10/src/lib-storage/index/maildir/maildir-copy.c, I see that hardlinks are used. It seems that (line 185) if MAILDIR_COPY_WITH_HARDLINKS is exported, then copying is attempted by using link. Having problems reading the source / flow of control, I resort to this list. When a file is linked, the mtime is naturally kept on the destination. This gives the destination message the same internaldate as the original. The destination mailbox would then be likely to get an arrival with an earlier internaldate than its newest message. The IMAP protocol says to bump UIDVALIDITY when this happens. Does Dovecot always bump UIDVALIDITY when messages are copied, or is not the mtime used for internaldate? Best regards, Andreas -- Andreas Aardal Hanssen

2 1

[Dovecot] 0.99.10-test5
by Timo Sirainen 12 Aug '03

12 Aug '03

Again at http://dovecot.procontrol.fi/test/ if someone wants to test it. - new index file code should work now - mbox support is completely broken - caching isn't very smart yet. it always caches only what client requested at the time - header caching might have some problems actually now that I think of it.. I was anyway going to change a bit how it worked.

2 1

[Dovecot] CVS/test version config/build failures on FreeBSD
by Mike Brown 11 Aug '03

11 Aug '03

While diagnosing a problem I was having with IMAP and SSL (which I ended up fixing by simply upgrading from 0.99.9.1 to 0.99.10, and possibly affected by upgrading openssl libs from 0.9.7a to 0.9.7.b), I attempted to try the CVS version of Dovecot. When using the CVS version, apparently I have to first generate 'configure' using autogen.sh (?). This was the result of that, on FreeBSD 4.8-STABLE (2003-08-06): # ./autogen.sh automake: configure.in: installing `./install-sh' automake: configure.in: installing `./mkinstalldirs' automake: configure.in: installing `./missing' configure.in: 3: required file `./config.h.in' not found automake: configure.in: installing `./depcomp' configure.in:3: warning: do not use m4_patsubst: use patsubst or m4_bpatsubst configure.in:13: warning: AC_ARG_PROGRAM invoked multiple times configure.in:1126: warning: do not use m4_regexp: use regexp or m4_bregexp autoheader: `config.h.in' is created And then... # ./configure checking for a BSD-compatible install... /usr/bin/install -c checking whether build environment is sane... yes checking for gawk... no checking for mawk... no checking for nawk... nawk checking whether make sets ${MAKE}... yes checking whether to enable maintainer-specific portions of Makefiles... no checking for style of include used by make... GNU checking for gcc... gcc checking for C compiler default output... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking dependency style of gcc... gcc checking for strerror in -lcposix... no checking for gcc... (cached) gcc checking whether we are using the GNU C compiler... (cached) yes checking whether gcc accepts -g... (cached) yes checking dependency style of gcc... (cached) gcc checking how to run the C preprocessor... gcc -E checking for ANSI C header files... yes checking for gcc option to accept ANSI C... none needed checking for inline... inline checking build system type... i386-unknown-freebsdelf4.8 checking host system type... i386-unknown-freebsdelf4.8 checking for ranlib... ranlib checking for ld used by GCC... /usr/libexec/elf/ld checking if the linker (/usr/libexec/elf/ld) is GNU ld... yes checking for BSD-compatible nm... /usr/bin/nm -B checking whether ln -s works... yes loading cache /dev/null within ltconfig ./ltconfig: .: /dev/null: not a regular file ltconfig: you must specify a host type if you use `--no-verify' Try `ltconfig --help' for more information. configure: error: libtool configure failed I also attempted to use 0.99.11-test4 from http://dovecot.procontrol.fi/test/, but this failed to build: [...] Making all in maildir gcc -DHAVE_CONFIG_H -I. -I. -I../../../.. -I../../../../src/lib -I../../../../src/lib-mail -I../../../../src/lib-imap -I../../../../src/lib-index -I../../../../src/lib-index/maildir -I../../../../src/lib-storage -I../../../../src/lib-storage/index -I/usr/local/include -g -O2 -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -c maildir-copy.c make: don't know how to make ../../../../src/lib-storage/index/index-expunge.h. Stop *** Error code 1 Stop in /usr/local/src/dovecot/dovecot-0.99.11-test4/src/lib-storage/index. Just thought you'd like to know :) -Mike

3 8

[Dovecot] Some benchmarking
by Timo Sirainen 11 Aug '03

11 Aug '03

So I finally got the new mail caching code fully working in normal conditions. Index rebuilding and some error correction still needs a bit work. mbox is also still broken. Anyway, I was mostly interested how fast Dovecot now works with my 367000 LKML mails, but thought I'd have to get something to compare the results against, so here's Dovecot vs. Courier. Results against older Dovecot would have been pretty boring, the older Dovecot could never perform threading without opening all the files and reading References-headers from them. Sorting would could have been done from cache if IMAP ENVELOPEs were cached, I'd estimate maybe 2-3x larger user CPU usage with it and a _lot_ larger cache file since it would have stored from, to, etc. fields in it. Results against Cyrus would be interesting. I'll probably do those sometimes later too.. Last I checked, Cyrus was very slow at inserting the mails into it's database, it took .. was it 1-3 hours to insert all those mails, while a simple copying took maybe 20 mins. Anyway, Cyrus's caching is pretty dummy as well so I'd expect Dovecot to be many times faster in threading and likely somewhat faster in sorting.

3 2

[Dovecot] Direct RADIUS support, not via PAM
by Jethro R Binks 08 Aug '03

08 Aug '03

Hi, I'm currently testing dovecot to authenticate user passwords against a radius server, via pam_radius_auth, and that seems to work fine. However, I am running other software (Exim MTA) that also authenticates via radius, using the radiusclient library from: http://www.mcs.de/~lf/radius/ Is there any chance that this library could be used by Dovecot as a further authentication option, and save me having two separate radius configurations for two different (but related) pieces of software, and the need to install pam_radius_auth, which I don't otherwise use? Jethro. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jethro R Binks Computing Officer, IT Services University Of Strathclyde, Glasgow, UK

2 1

[Dovecot] Slow bulk deleting
by Leslie Viljoen 07 Aug '03

07 Aug '03

Hi! I am finding Dovecot very slow to delete large numbers of messages. IS this a known problem? Leslie

2 2

[Dovecot] compile with dietlibc
by Jens Gutzeit 07 Aug '03

07 Aug '03

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, I'm used to test the software which I use with dietlibc, just to see if it could run without the bloat of glibc. While compiling dovecot 0.99.10 I got these errors: In file included from lib.h:20, from alarm-hup.c:26: compat.h:30: conflicting types for `uint_fast32_t' /opt/diet/include/stdint.h:25: previous declaration of `uint_fast32_t' I've solved this in a dirty way, but is this something which could be fixed in the future? Thanks in advance, Jens Gutzeit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/L/etLPHiyaqJLscRAuNGAJ9jlOQzx86T+NS9h7FfjoDi86R4ygCgs6gb 1XyNB93ybwiG4gFatz797qo= =WZod -----END PGP SIGNATURE-----

3 3

[Dovecot] thinking seriously about converting
by Eric S. Johansson 06 Aug '03

06 Aug '03

to dovecot from UW imap. I think I will go all the way to maildir while I'm at it. I was having performance problems with mbox format files and converted to mbx but it's not really yielding a huge performance gain. Deletes and moves are still rather slow at times. my mailboxes are typically 2000 to 5000 messages. I've scanned the mailing lists a bit and picked up some hints on the conversion. Any bits of advice before I convert? Since it's only my own account I'm converting (my wife likes mutt), I'm torn between two different ways of converting the mailboxes. First is to convert all of the mbx format mailboxes back to mbox. Then convert from mbox to maildir. Second is running both imap server simultaneously (obviously one on a different port number) and running some form of a tool to read from one imap server and write to the other. any better ideas? many thanks ---eric

2 1

[Dovecot] corrupt mbox, mailboxes not found, and message read status
by Jared 04 Aug '03

04 Aug '03

I'm running dovecot 0.99.10-0.rc2 deb packages for Debian Woody from braincells.com. My old environment was a server running UW-IMAP for IMAP and POP services. We are still using mbox for all mailboxes. I use IMP for web based access. Forgive me if these issues have already been discussed or fixed in cvs. Any additional pointers to this info. would be appreciated. I have noticed a few issues. ---- I've seen a whole slew of entries like the following: pop3(username): Error indexing mbox file /home/username/mail/inbox: LF not found where expected According to this mailing list post from June 16, 2003: http://dovecot.procontrol.fi/list/dovecot/2003-June/001668.html it may be because of adjacent From mail headers in a message. I've had one user that this was not the case. They said they were deleting lots of messages in IMP and when he went back to his inbox, it was empty. When I looked at the mbox file, a message had been truncated. When I deleted the imap index and removed that truncated message, the mailbox became visible again. Has the adjacent From issue been fixed? Is there a way to find out more specifically what part of the mbox file dovecot is unhappy about? ie. a line number. ---- Inside IMP several users have set up a filter to send messages marked as spam by SpamAssassin into a spam folder. When they log-in or click the filter icon, the filter runs. However, now when the the filter runs, it reports: "Problem with filter rule 1: folder does not exist: mail/spam". However, the spam folder is still there, and when I open that folder in IMP, it opens just fine. I've already changed IMP's folder prefix from 'mail/' to ''. I've also tried deleting the imap index. The message still shows up and the filter fails to run. ---- With the old set-up, when I check mail with a pop3 client, the UW pop server modifies the Status and/or X-Keywords headers for the message. When the mailbox is viewed with an IMAP client such as IMP, messages that have already been downloaded via pop are marked as read. This doesn't seem to happen with dovecot. I found a Oct 19 2003 thread ( http://dovecot.procontrol.fi/list/dovecot/2002-October/000774.html ) on the mailing list where Timo said he was intending to use the X-Keywords trick. Did this ever get implemented? ---- Thanks, -jared

5 6

[Dovecot] why wont /usr/pkg/libexec/dovecot/imap see my Maildir
by Charlie Allom 04 Aug '03

04 Aug '03

When I invoke /usr/pkg/libexec/dovecot/imap from the command loine on the mail host, it logs me in, but won't see my emails: 13:22 yeled@lazy:~$ /usr/pkg/libexec/dovecot/imap * PREAUTH [CAPABILITY IMAP4rev1 SORT THREAD=REFERENCES MULTIAPPEND * UNSELECT LITERAL+ IDLE CHILDREN LISTEXT LIST-SUBSCRIBED] Logged in * as yeled 0 select inbox * FLAGS (\Answered \Flagged \Deleted \Seen \Draft) * OK [PERMANENTFLAGS (\Answered \Flagged \Deleted \Seen \Draft \*)] * Flags permitted. * 0 EXISTS * 0 RECENT * OK [UIDVALIDITY 1059959635] UIDs valid * OK [UIDNEXT 1] Predicted next UID 0 OK [READ-WRITE] Select completed. I know I have many emails in there, and I can see them when I login via telnet localhost 443. Any tips? would be nice to use fetchmail with ssh here. C. -- charlie(a)rubberduck.com - Melbourne, Australia http://rubberduck.com/~yeled/ PGP: 0x14AA7941 || finger yeled(a)lazy.spodder.com

2 2