- dovecot - dovecot.org

[Dovecot] Adding Sieve Extensions
by James Butler 06 Apr '09

06 Apr '09

How can I add an extension to Dovecot's Sieve implementation? I would like to use 'editheader' and 'redirect'. Thank you! James

4 13

[Dovecot] v1.2: can't access other users shared INBOX
by Sascha Wilde 06 Apr '09

06 Apr '09

Hi *, when a user A shares his INBOX with another user B, the user B can't access its content: User A: g getacl INBOX * ACL INBOX A(a)example.com lrswipkxtecda B(a)example.com lrswipkxtecd g OK Completed User B: l list "" "*" * LIST (\Noselect \HasChildren) "/" "user" * LIST (\Noselect \HasChildren) "/" "user/A(a)example.com" * LIST (\HasChildren) "/" "INBOX" * LIST (\HasNoChildren) "/" "INBOX/Gesendet" * LIST (\HasChildren) "/" "user/A(a)example.com/foobar" * LIST (\HasNoChildren) "/" "user/A(a)example.com/INBOX" l OK List completed. s1 select "user/A(a)example.com" s1 NO [CANNOT] Invalid mailbox name s2 select "user/A(a)example.com/INBOX" s2 NO [NONEXISTENT] Mailbox doesn't exist: INBOX Actually there are two bugs to observe here: 1) "user/A(a)example.com" really should be accessible to user B. Why is it listed with "\Noselect"? 2) "user/A(a)example.com/INBOX" does not exist, so the error message is correct, but why does it appear in the listing in the first place? cheers sascha -- Sascha Wilde OpenPGP key: 4BB86568 http://www.intevation.de/~wilde/ http://www.intevation.de/ Intevation GmbH, Neuer Graben 17, 49074 Osnabrück; AG Osnabrück, HR B 18998 Geschäftsführer: Frank Koormann, Bernhard Reiter, Dr. Jan-Oliver Wagner

4 12

[Dovecot] big picture
by Mohsen Pahlevanzadeh 06 Apr '09

06 Apr '09

Dear all, You & every mail-man saw Big Picture of qmail. I assumed qmail is black box.I want to extend big picture, i use qmail admin ,Devocot & vpopmail. I draw a few relation.Can you tell me whether i draw correct?Also i need to if i didn't draw relation tell me that i will draw that. Yours, Mohsen

1 0

[Dovecot] Implementation of editheaders in dovecot
by Konstantin Lepa 06 Apr '09

06 Apr '09

Hello. I wrote the hook function for deliver. I want to add support of editheaders in the plug-in for dovecot. For this purpose I wrote the function rarules_get_stream. Remover of headrs works properly, but adding does not work. I took Timo Sirainen's advice from http://markmail.org/message/skb6arnll5gaopdr . Do I use a correct way of creation of a message? I give backtrace and a part of my code. Thanks, Konstantin uname -a: FreeBSD bsdman.virtualworld.com 6.4-RELEASE FreeBSD 6.4-RELEASE #0: Wed Nov 26 11:43:51 UTC 2008 root@dessler.cse.buffalo.edu:/usr/obj/ usr/src/sys/GENERIC i386 dovecot --version: 1.1.11 build: $ ./configure --localstatedir=/var --with-statedir=/var/db/dovecot -- without-shadow --enable-debug --with-ioloop=kqueue --disable-ipv6 -- with-gssapi --without-vpopmail --without-ldap --without-pgsql --with- mysql --without-sqlite --prefix=/usr/local --mandir=/usr/local/man -- infodir=/usr/local/info/ --build=i386-portbld-freebsd6.4 patches: in freebsd ports (gdb) bt #0 0x282f3a01 in memcpy () from /lib/libc.so.6 #1 0x080df555 in i_stream_concat_read_next (cstream=0x814aa00) at istream-concat.c:79 #2 0x080df784 in i_stream_concat_read (stream=0x814aa00) at istream- concat.c:136 #3 0x080de659 in i_stream_read (stream=0x814aa28) at istream.c:78 #4 0x080cd99a in i_stream_header_filter_read (stream=0x814e400) at istream-header-filter.c:322 #5 0x080de659 in i_stream_read (stream=0x814e428) at istream.c:78 #6 0x080dff26 in i_stream_crlf_read_common (cstream=0x814ad00) at istream-crlf.c:41 #7 0x080e01b3 in i_stream_crlf_read_lf (stream=0x814ad00) at istream- crlf.c:103 #8 0x080de659 in i_stream_read (stream=0x814ad28) at istream.c:78 #9 0x080e2611 in i_stream_tee_read (stream=0x814ae00) at istream- tee.c:122 #10 0x080de659 in i_stream_read (stream=0x814ae28) at istream.c:78 #11 0x080deea3 in i_stream_read_data (stream=0x814ae28, data_r=0xbfbfea88, size_r=0xbfbfea8c, threshold=1) at istream.c:322 #12 0x080d1744 in message_parser_read_more (ctx=0x8156468, block_r=0xbfbfea80, full_r=0xbfbfea27) at message-parser.c:117 #13 0x080d20ac in parse_next_body_to_eof (ctx=0x8156468, block_r=0xbfbfea80) at message-parser.c:411 #14 0x080d2ccf in message_parser_parse_next_block (ctx=0x8156468, block_r=0xbfbfea80) at message-parser.c:765 #15 0x080931d9 in index_mail_cache_parse_continue (_mail=0x8150060) at index-mail.c:1316 #16 0x0807838d in mbox_save_continue (_ctx=0x811bb00) at mbox-save.c:582 #17 0x080a05a8 in mailbox_save_continue (ctx=0x811bb00) at mail- storage.c:740 #18 0x0809d9df in mail_storage_copy (t=0x811b980, mail=0x8141860, flags=0, keywords=0x0, dest_mail=0x0) at mail-copy.c:33 #19 0x080a0632 in mailbox_copy (t=0x811b980, mail=0x8141860, flags=0, keywords=0x0, dest_mail=0x0) at mail-storage.c:763 #20 0x0805960d in deliver_save (namespaces=0x8121c68, storage_r=0xbfbfee14, mailbox=0x811a370 "Draft", mail=0x8141860, flags=0, keywords=0x0) at deliver.c:234 #21 0x2831fe8f in store_in (arg=0x811a370 "Draft", md=0xbfbfed00) at / home/pitman/work/rarules/src/actions.c:220 #22 0x2831f759 in rarules_execute_action (type=RARULES_ACTION_STORE_IN, arg=0x811a370 "Draft", md=0xbfbfed00) at /home/pitman/work/rarules/src/actions.c:57 #23 0x2831e70d in execute_actions (action=0x811a360, md=0xbfbfed00) at /home/pitman/work/rarules/src/rarules.c:98 #24 0x2831e692 in apply_rules (rule=0x811a320, md=0xbfbfed00) at /home/ pitman/work/rarules/src/rarules.c:80 #25 0x2831e56a in rarules_main (namespaces=0x8121c68, storage_r=0xbfbfee14, mail=0x8141860, db=0x8121000, destaddr=0xbfbfef62 "pitman(a)virtualworld.com", username=0x8113525 "pitman(a)virtualworld.com", mailbox=0x80f2a19 "INBOX") at /home/pitman/work/rarules/src/rarules.c:52 #26 0x2831da37 in rarules_deliver_mail (namespaces=0x8121c68, storage_r=0xbfbfee14, mail=0x8141860, destaddr=0xbfbfef62 "pitman(a)virtualworld.com", mailbox=0x80f2a19 "INBOX") at /home/pitman/work/rarules/src/rarules_plugin.c:126 #27 0x0805b69a in main (argc=3, argv=0xbfbfeeac) at deliver.c:1084 (gdb) f 1 #1 0x080df555 in i_stream_concat_read_next (cstream=0x814aa00) at istream-concat.c:79 79 memcpy(cstream->istream.w_buffer, data, data_size); (gdb) p data $4 = (const unsigned char *) 0x15 <Address 0x15 out of bounds> (gdb) p data_size $5 = 1 /////////////////////////////////////////My code///////////////////////////////////////////////////////// static int rarules_get_stream(struct mail * mail, struct message_size * hdr_size, struct message_size * body_size, struct istream ** stream_r) { struct rarules_mail * ra_mail = RARULES_MAIL_CONTEXT(mail); struct istream * stream = NULL; struct istream ** chunks = NULL; const unsigned int max_chunks = 4; unsigned int i = 0; unsigned int cnt = 0; unsigned int exclude_cnt = 0; unsigned int include_cnt = 0; int ret = 0; #if 1 {/* Test code */ const char * str = "Subject"; const char * str2 = "X-DSPAM: test value\n\n"; array_append(&ra_mail->exclude_headers, &str, 1); array_append(&ra_mail->include_headers, &str2, 1); } #endif DPRINT("Standard get_stream for mail storage format"); ret = ra_mail->module_ctx.super.get_stream(mail, hdr_size, body_size, &stream); include_cnt = array_count(&ra_mail->include_headers); exclude_cnt = array_count(&ra_mail->exclude_headers); if ((!exclude_cnt && !include_cnt) || ret < 0) { *stream_r = stream; return ret; } DPRINT("Create message from chunks with new headers and skip excluded headers"); chunks = t_new(struct istream *, max_chunks); chunks[cnt++] = header_filter(stream, &ra_mail->exclude_headers, exclude_cnt); if (include_cnt) { chunks[cnt++] = create_header_stream(&ra_mail->include_headers, include_cnt); } chunks[cnt++] = create_stream_for_msgbody(stream); chunks[cnt++] = NULL; *stream_r = i_stream_create_concat(chunks); DPRINT("Cleanup"); array_clear(&ra_mail->include_headers); array_clear(&ra_mail->exclude_headers); for (i = 0; chunks[i] != NULL; ++i) { i_stream_unref(&chunks[i]); } i_stream_unref(&stream); DPRINTF("Func `%s' executed", __FUNCTION__); return ret; } inline static struct istream * create_stream_for_msgbody(struct istream * stream) { struct istream * tmp = stream; struct message_size hdr_size; struct message_size body_size; DPRINTF("Func `%s' called", __FUNCTION__); message_get_body_size(stream, &body_size, NULL); message_get_header_size(stream, &hdr_size, NULL); i_stream_seek(stream, hdr_size.physical_size); stream = i_stream_create_limit(stream, body_size.physical_size); i_stream_unref(&tmp); return stream; } inline static struct istream * header_filter(struct istream * stream, ARRAY_TYPE(const_string) * headers_arr, unsigned int count) { const char * const * headers = NULL; DPRINTF("Func `%s' called", __FUNCTION__); if (count) { headers = (const char **)array_idx(headers_arr, 0); } return i_stream_create_header_filter(stream, HEADER_FILTER_EXCLUDE | HEADER_FILTER_NO_CR | HEADER_FILTER_HIDE_BODY, headers, count, null_header_filter_callback, NULL); } inline static struct istream * create_header_stream(ARRAY_TYPE(const_string) * headers_arr, unsigned int count) { unsigned int i = 0; struct istream * result = NULL; struct istream ** chunks = NULL; const char * const * headers = NULL; DPRINTF("Func `%s' called", __FUNCTION__); i_assert(count != 0); headers = (const char **)array_idx(headers_arr, 0); DPRINT("Create streams for new included headers from headers"); chunks = t_new(struct istream *, count + 1); for (i = 0; i < count; ++i) { struct istream * tmp = NULL; tmp = i_stream_create_from_data(headers[i], strlen(headers[i])); chunks[i] = i_stream_create_lf(tmp); i_stream_unref(&tmp); } chunks[count] = NULL; result = i_stream_create_concat(chunks); for (i = 0; chunks[i] != NULL; ++i) { i_stream_unref(&chunks[i]); } return result; }

3 9

[Dovecot] dovecot-1.1.13 auth-worker killed
by Jiri Novosad 06 Apr '09

06 Apr '09

Hi, I get these errors with dovecot 1.1.13 in my logs: dovecot: Mar 27 12:00:34 Error: auth-worker(default): *** glibc detected *** dovecot-auth: double free or corruption (!prev): 0x00002b30d5a7dd90 *** dovecot: Mar 27 12:00:34 Error: auth(default): worker-server(jan,147.251.48.141): Aborted: Worker process died unexpectedly dovecot: Mar 27 12:00:34 Error: child 21838 (auth-worker) killed with signal 6 (core not dumped) dovecot: Mar 27 12:10:23 Error: child 22711 (auth-worker) killed with signal 11 (core not dumped) dovecot: Mar 27 12:10:23 Error: auth(default): worker-server(thomas,209.85.146.140): Aborted: Worker process died unexpectedly Haven't seen this with previous versions. But somehow I fail to get the core dump. I do 'ulimit -c unlimited' before starting dovecot, I set * soft core 8192 * hard core 8192 in limits.conf, /proc/sys/kernel/core_pattern is set to /tmp/core. Regards, Jiri Novosad My dovecont -n: # 1.1.13: /packages/run.64/dovecot-1.1.13//etc/dovecot.conf # OS: Linux 2.6.18-92.1.22.el5 x86_64 Red Hat Enterprise Linux Server release 5.3 (Tikanga) ext3 base_dir: /var/run/dovecot/ log_path: /var/log/dovecot/log info_log_path: /var/log/dovecot/info_log protocols: pop3s imaps pop3 imap listen(default): * listen(imap): * listen(pop3): *:110 ssl_listen(default): ssl_listen(imap): ssl_listen(pop3): *:995 ssl_cert_file: /etc/ssl/cert-anxur.fi.muni.cz.pem ssl_key_file: /etc/ssl/serverkey-anxur.fi.muni.cz.pem ssl_parameters_regenerate: 24 login_dir: /var/run/dovecot/login login_executable(default): /packages/run.64/dovecot-1.1.13/libexec/dovecot/imap-login login_executable(imap): /packages/run.64/dovecot-1.1.13/libexec/dovecot/imap-login login_executable(pop3): /packages/run.64/dovecot-1.1.13/libexec/dovecot/pop3-login login_greeting: HI! login_greeting_capability(default): yes login_greeting_capability(imap): yes login_greeting_capability(pop3): no verbose_proctitle: yes first_valid_uid: 200 mail_location: mbox:/home/%u/mail/:INBOX=/var/mail/%u:INDEX=/home/%u/mail/.imap mail_debug: yes mail_full_filesystem_access: yes maildir_copy_with_hardlinks: no mbox_write_locks: fcntl mbox_dirty_syncs: no mbox_lazy_writes: no dbox_rotate_days: 0 mail_drop_priv_before_exec: yes mail_executable(default): /packages/run.64/dovecot-1.1.13/libexec/dovecot/imap mail_executable(imap): /packages/run.64/dovecot-1.1.13/libexec/dovecot/imap mail_executable(pop3): /packages/run.64/dovecot-1.1.13/libexec/dovecot/pop3 mail_plugin_dir(default): /packages/run.64/dovecot-1.1.13/lib/dovecot/imap mail_plugin_dir(imap): /packages/run.64/dovecot-1.1.13/lib/dovecot/imap mail_plugin_dir(pop3): /packages/run.64/dovecot-1.1.13//lib/dovecot/pop3 mail_log_max_lines_per_sec: 20 imap_max_line_length(default): 131072 imap_max_line_length(imap): 131072 imap_max_line_length(pop3): 65536 imap_client_workarounds(default): delay-newmail outlook-idle tb-extra-mailbox-sep imap_client_workarounds(imap): delay-newmail outlook-idle tb-extra-mailbox-sep imap_client_workarounds(pop3): pop3_uidl_format(default): %08Xu%08Xv pop3_uidl_format(imap): %08Xu%08Xv pop3_uidl_format(pop3): %08Xv%08Xu pop3_client_workarounds(default): pop3_client_workarounds(imap): pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh auth default: executable: /packages/run.64/dovecot-1.1.13/libexec/dovecot/dovecot-auth verbose: yes debug: yes passdb: driver: pam args: dovecot userdb: driver: passwd

3 7

[Dovecot] conflict name of si_code on NetBSD
by Tatsuyoshi 06 Apr '09

06 Apr '09

Hi. I compiled 1.2.rc2 on NetBSD. [toyota@sernpidal]% uname -rsm NetBSD 3.1 i386 [toyota@sernpidal]% ./configure; make --- snip --- gcc -DHAVE_CONFIG_H -I. -I../.. -std=gnu99 -g -O2 -Wall -W -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wformat=2 -Wbad-function-cast -MT child-wait.o -MD -MP -MF .deps/child-wait.Tpo -c -o child-wait.o child-wait.c In file included from child-wait.c:4: lib-signals.h:9: error: parse error before '.' token *** Error code 1 Stop. Error occurred. I changed some code, and find out the source of the trouble. [toyota@sernpidal]% grep -w si_code /usr/include/sys/siginfo.h #define si_code _info._code /** si_code */ /** si_code */ The 'si_code' name is conflicted, and I need change the name of 'si_code'. The following is quick patch for NetBSD, just change si_code to sicode. [toyota@sernpidal]% diff -crN dovecot-1.2.rc2_bak dovecot-1.2.rc2 diff -crN dovecot-1.2.rc2_bak/src/lib/lib-signals.c dovecot-1.2.rc2/src/lib/lib-signals.c *** dovecot-1.2.rc2_bak/src/lib/lib-signals.c Thu Apr 2 02:08:21 2009 --- dovecot-1.2.rc2/src/lib/lib-signals.c Mon Apr 6 12:45:42 2009 *************** *** 27,36 **** static bool signals_initialized = FALSE; static struct io *io_sig = NULL; ! const char *lib_signal_code_to_str(int signo, int si_code) { /* common */ ! switch (si_code) { case SI_USER: return "kill"; #ifdef SI_KERNEL --- 27,36 ---- static bool signals_initialized = FALSE; static struct io *io_sig = NULL; ! const char *lib_signal_code_to_str(int signo, int sicode) { /* common */ ! switch (sicode) { case SI_USER: return "kill"; #ifdef SI_KERNEL *************** *** 43,49 **** switch (signo) { case SIGSEGV: ! switch (si_code) { case SEGV_MAPERR: return "address not mapped"; case SEGV_ACCERR: --- 43,49 ---- switch (signo) { case SIGSEGV: ! switch (sicode) { case SEGV_MAPERR: return "address not mapped"; case SEGV_ACCERR: *************** *** 51,57 **** } break; case SIGBUS: ! switch (si_code) { case BUS_ADRALN: return "invalid address alignment"; case BUS_ADRERR: --- 51,57 ---- } break; case SIGBUS: ! switch (sicode) { case BUS_ADRALN: return "invalid address alignment"; case BUS_ADRERR: *************** *** 60,66 **** return "object-specific hardware error"; } } ! return t_strdup_printf("unknown %d", si_code); } static void sig_handler(int signo, siginfo_t *si, void *context ATTR_UNUSED) --- 60,66 ---- return "object-specific hardware error"; } } ! return t_strdup_printf("unknown %d", sicode); } static void sig_handler(int signo, siginfo_t *si, void *context ATTR_UNUSED) diff -crN dovecot-1.2.rc2_bak/src/lib/lib-signals.h dovecot-1.2.rc2/src/lib/lib-signals.h *** dovecot-1.2.rc2_bak/src/lib/lib-signals.h Thu Apr 2 01:59:55 2009 --- dovecot-1.2.rc2/src/lib/lib-signals.h Mon Apr 6 12:46:00 2009 *************** *** 5,12 **** typedef void signal_handler_t(const siginfo_t *si, void *context); ! /* Convert si_code to string */ ! const char *lib_signal_code_to_str(int signo, int si_code); /* Set signal handler for specific signal. If delayed is TRUE, the handler will be called later, ie. not as a real signal handler. */ --- 5,12 ---- typedef void signal_handler_t(const siginfo_t *si, void *context); ! /* Convert sicode to string */ ! const char *lib_signal_code_to_str(int signo, int sicode); /* Set signal handler for specific signal. If delayed is TRUE, the handler will be called later, ie. not as a real signal handler. */ -- Tatsuyoshi Toyota rs27(a)tatsuyoshi.net

1 0

[Dovecot] is it possible to recover files from a dovecot.inboxe.cache file?
by Alfred Nutile 06 Apr '09

06 Apr '09

I found this file in the mail boz and noticed it had 20 emails in it (though you can only read really the header the body is cryptic cause it is a bin file?) Just not sure if I can and how to use this to restore the lost emails in that inbox. Thanks

2 1

[Dovecot] Segfault in Sieve 1.2
by Bernhard Schmidt 05 Apr '09

05 Apr '09

Hello, two days ago I've upgraded to 1.2rc2 and the new sieve implementation (version 0.1.4). Today I got the first mail that got stuck in the postfix queue because deliver was dying with a segfault. The mail is spam, thus I don't have any issues sharing it: http://users.birkenwald.de/~berni/temp/deliver-segfault.mail It happens with this small sieve script (my normal one is larger, but this is the part that triggers the segfault): http://users.birkenwald.de/~berni/temp/deliver-segfault.sieve Note that I've been running this script unaltered for months, first with Dovecot 1.1 and CMUSieve and now with various versions of Dovecot 1.2 and the new sieve implementation for about three weeks, this is the first mail that got stuck. I got a coredump, but without debugging symbols it doesn't tell that much. If you can't reproduce it I'll compile it with debugging symbols and try again. (gdb) bt full #0 0x00007fa0f87ecbd7 in strstr () from /lib/libc.so.6 No symbol table info available. #1 0x00007fa0f8341933 in ?? () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #2 0x00007fa0f83377a3 in sieve_address_match () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #3 0x00007fa0f833ba7b in ?? () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #4 0x00007fa0f8332537 in sieve_interpreter_continue () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #5 0x00007fa0f8332728 in sieve_interpreter_run () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #6 0x00007fa0f833f8ca in ?? () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #7 0x00007fa0f833fb3d in sieve_execute () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #8 0x00007fa0f8326f9d in ?? () from /usr/lib/dovecot/lda/lib90_sieve_plugin.so No symbol table info available. #9 0x000000000041a5ad in main () No symbol table info available. Platform is Debian Lenny amd64. Regards, Bernhard

2 1

[Dovecot] v1.2.rc2 released
by Timo Sirainen 04 Apr '09

04 Apr '09

http://dovecot.org/releases/1.2/rc/dovecot-1.2.rc2.tar.gz http://dovecot.org/releases/1.2/rc/dovecot-1.2.rc2.tar.gz.sig Somehow I forgot to check that "make distcheck" finished successfully in rc1, so rc1 didn't actually even fully compile. No other changes since rc1 than the compile fix.

3 5

[Dovecot] THREAD X-REFERENCES2 index bug?
by Nathan Kinkade 04 Apr '09

04 Apr '09

Hi all, $ sudo dovecot --version 1.2.beta3 I just started using Dovecot a few weeks ago, and my client is also using the X-REFERENCES2 thread algorithm. The client is misbehaving when trying to list messages in my INBOX, but only when a threaded message exists. I have tracked it down to the thread index that Dovecot is returning, but the problem is only extant when using the X-REFERENCES2 thread algorithm. Here is a snippet of transcript from imap client that comes with Dovecot: $ export MAIL=maildir:~/Maildir:INBOX=~/Maildir/INBOX $ /usr/lib/dovecot/imap <snip> 1 SELECT "INBOX" <snip> 2 THREAD X-REFERENCES2 UTF-8 ALL * THREAD (0)(2)(1 3)(4) 2 OK Thread completed. There are, in fact, only 4 messages in my INBOX, and the reference to (0) seems spurious. When I do the same test, but passing REFERENCES instead of X-REFERENCES2 Dovecot returns an index which I believe to be correct: <snip> 2 THREAD REFERENCES UTF-8 ALL * THREAD (1 3)(2)(4) 2 OK Thread completed. <snip> Is this the normal and expected behavior of X-REFERENCES2 that the client just needs to take into account, or is the reference to thread index (0) an error? Thanks! Nathan

2 6