Re: [Dovecot] Dovecot-1.1.2 assertion failure in preparsed_parse_next_header_init

2008-08-30 Thread Timo Sirainen 
On Fri, 2008-08-29 at 15:03 -0500, Mike Abbott wrote:
> Panic: IMAP(user): file message-parser.c: line 684  
> (preparsed_parse_next_header_init): assertion failed: (ctx->part- 
>  >physical_pos >= ctx->input->v_offset)
..
> (gdb) p *ctx
> $4 = {parser_pool = 0x812b638, part_pool = 0x0, input = 0x8113618,
..
>broken = 1}

Ah, they were already detected to be broken. This should help then:
http://hg.dovecot.org/dovecot-1.1/rev/1d8bd4bc3038



signature.asc
Description: This is a digitally signed message part

[Dovecot] Dovecot-1.1.2 assertion failure in preparsed_parse_next_header_init

2008-08-29 Thread Mike Abbott 
Panic: IMAP(user): file message-parser.c: line 684  
(preparsed_parse_next_header_init): assertion failed: (ctx->part- 
>physical_pos >= ctx->input->v_offset)


Linux 2.6.24-19-386
Maildir on ext3

Looks similar to but different from http://dovecot.org/list/dovecot/2008-June/031523.html 
 .


#0  0xb7faf410 in __kernel_vsyscall ()
#1  0xb7e6f085 in raise () from /lib/tls/i686/cmov/libc.so.6
#2  0xb7e70a01 in abort () from /lib/tls/i686/cmov/libc.so.6
#3  0x080d2b65 in default_fatal_finish (type=,  
status=0)

at failures.c:149
#4  0x080d2bba in i_internal_fatal_handler (type=LOG_TYPE_PANIC,  
status=0,

fmt=0x80e5fec "file %s: line %d (%s): assertion failed: (%s)",
args=0xbfb4b1d4 "\017L\017\b?\002") at failures.c:423
#5  0x080d246c in i_panic (
format=0x80e5fec "file %s: line %d (%s): assertion failed: (%s)")
at failures.c:190
#6  0x080cc246 in preparsed_parse_next_header_init (ctx=0x812b650,
block_r=0xbfb4b24c) at message-parser.c:684
#7  0x080cbc0d in message_parser_parse_next_block (ctx=0x812b650,
block_r=0xbfb4b24c) at message-parser.c:753
#8  0x080cbd5e in message_parser_parse_body (ctx=0x812b650,
hdr_callback=0x80968a0 ,
context=0x812e658) at message-parser.c:816
#9  0x08097f21 in index_mail_parse_body (mail=0x8122668,
field=MAIL_CACHE_IMAP_BODYSTRUCTURE) at index-mail.c:753
#10 0x08098026 in index_mail_parse_bodystructure (mail=0x8122668,
field=MAIL_CACHE_IMAP_BODYSTRUCTURE) at index-mail.c:860
#11 0x0809852c in index_mail_get_special (_mail=0x8122668,
field=MAIL_FETCH_IMAP_BODYSTRUCTURE, value_r=0xbfb4b370)
at index-mail.c:983
#12 0x08078a18 in maildir_mail_get_special (_mail=0x6, field=0,
value_r=0xbfb4b370) at maildir-mail.c:437
#13 0x080643f4 in fetch_bodystructure (ctx=0x810b708, mail=0x8122668,
context=0x0) at imap-fetch.c:450
#14 0x08064960 in imap_fetch (ctx=0x810b708) at imap-fetch.c:309
#15 0x0805e44e in cmd_fetch (cmd=0x810b670) at cmd-fetch.c:152
#16 0x080625cc in client_command_input (cmd=0x810b670) at client.c:580
#17 0x08062675 in client_command_input (cmd=0x810b670) at client.c:629
#18 0x08062e43 in client_handle_input (client=0x810b418) at client.c:670
#19 0x0806305e in client_input (client=0x810b418) at client.c:725
#20 0x080da4a7 in io_loop_handler_run (ioloop=0x8109a60) at ioloop- 
epoll.c:201

#21 0x080d9628 in io_loop_run (ioloop=0x8109a60) at ioloop.c:308
#22 0x0806ab5d in main (argc=Cannot access memory at address 0x2acc
) at main.c:293
(gdb) frame 6
#6  0x080cc246 in preparsed_parse_next_header_init (ctx=0x812b650,
block_r=0xbfb4b24c) at message-parser.c:684
684 i_assert(ctx->part->physical_pos >= ctx->input->v_offset);
(gdb) p ctx
$1 = (struct message_parser_ctx *) 0x812b650
(gdb) p ctx->input
$2 = (struct istream *) 0x8113618
(gdb) p ctx->part
$3 = (struct message_part *) 0x812e7c0
(gdb) p *ctx
$4 = {parser_pool = 0x812b638, part_pool = 0x0, input = 0x8113618,
  parts = 0x812e730, part = 0x812e7c0, hdr_flags = 3,
  flags = MESSAGE_PARSER_FLAG_SKIP_BODY_BLOCK, last_boundary = 0x0,
  boundaries = 0x0, skip = 0, last_chr = 0 '\0', want_count = 0,
  hdr_parser_ctx = 0x0,
  parse_next_block = 0x80cc1e0 ,
  part_seen_content_type = 0, broken = 1}
(gdb) p *ctx->part
$5 = {parent = 0x812e778, next = 0x0, children = 0x0, physical_pos =  
688,

  header_size = {physical_size = 90, virtual_size = 93, lines = 0},
  body_size = {physical_size = 505, virtual_size = 514, lines = 9},
  flags = 72, context = 0x0}
(gdb) p *ctx->input
$6 = {v_offset = 759, stream_errno = 0, mmaped = 0, blocking = 1,  
closed = 0,

  seekable = 1, eof = 0, real_stream = 0x81135f0}

Re: [Dovecot] dovecot 1.1.2 assertion

2008-08-05 Thread Timo Sirainen 


On Aug 5, 2008, at 3:48 AM, Nuno Lopes wrote:



Do you still have the files? You forgot to send the  
dovecot.index.log file. Also perhaps dovecot.index.cache could help  
when figuring this out (but it contains message headers so you may  
not want to send it to this mailing list at least).




Hi, I still have them.
I sent both dovecot.index and dovecot.index.log the last time, but  
now I've also added the cache file in a tgz.


Oh. I guess that's why Mail.app showed some weird input in the mail.  
Anyway fixed now: http://hg.dovecot.org/dovecot-1.1/rev/d674c05d725d


PGP.sig
Description: This is a digitally signed message part

Re: [Dovecot] dovecot 1.1.2 assertion

2008-08-04 Thread Timo Sirainen 

On Jul 24, 2008, at 9:22 PM, Nuno Lopes wrote:


Timo Sirainen wrote:

On Jul 25, 2008, at 2:36 AM, Nuno Lopes wrote:

I've upgraded to last version of dovecot and am now getting the  
following assertions:


Jul 25 00:00:34 bld3 dovecot: Panic: IMAP([EMAIL PROTECTED]):  
file index-sort-string.c: line 542 (index_sort_add_ids_range):  
assertion failed: (left_idx == 0 && right_idx == rightmost_idx)


Can you reproduce this? Could you send that mailbox's dovecot.index  
and dovecot.index.log files to me? They don't contain any sensitive  
information.


Both bugs are fully reproducible by trying to order a given field.

In the mailbox corresponding to the indexes I send you it happens  
every time I try to order by subject.


Would you like me to send you files corresponding to the other bug  
or even more examples of this one if I can find them?


Do you still have the files? You forgot to send the dovecot.index.log  
file. Also perhaps dovecot.index.cache could help when figuring this  
out (but it contains message headers so you may not want to send it to  
this mailing list at least).




PGP.sig
Description: This is a digitally signed message part

Re: [Dovecot] dovecot 1.1.2 assertion

2008-07-24 Thread Nuno Lopes 

Here go the files

Timo Sirainen wrote:

On Jul 25, 2008, at 2:36 AM, Nuno Lopes wrote:

I've upgraded to last version of dovecot and am now getting the 
following assertions:


Jul 25 00:00:34 bld3 dovecot: Panic: IMAP([EMAIL PROTECTED]): file 
index-sort-string.c: line 542 (index_sort_add_ids_range): assertion 
failed: (left_idx == 0 && right_idx == rightmost_idx)


Can you reproduce this? Could you send that mailbox's dovecot.index 
and dovecot.index.log files to me? They don't contain any sensitive 
information.




Both bugs are fully reproducible by trying to order a given field.

In the mailbox corresponding to the indexes I send you it happens every 
time I try to order by subject.


Would you like me to send you files corresponding to the other bug or 
even more examples of this one if I can find them?



Thanks,
--
Nuno Lopes


dovecot.index
Description: Binary data
…+ëG
[EMAIL PROTECTED] [EMAIL PROTECTED]@€€€‰J†HJ†HJ†HJ†H€€€„Ê€€€„ 0Ë€€€„ [EMAIL PROTECTED]@€€€‰^†HJ†H^†HJ†H€€€„ [EMAIL PROTECTED]@[EMAIL PROTECTED]@[EMAIL PROTECTED]	¨ª
[EMAIL PROTECTED]@[EMAIL PROTECTED]@Ž+ëG€€€øY ZxaÈ
b cldÀr ul%v°*wô/xD5y”:{à?| E}\J~°OU€XZ°_‚üdƒLjŠ oŒøtLz ‘ô„’PŠ—¬˜•™LšŸ Ÿ§ø¤¨<ª©¯ªä´­È·®л¯,Á°ñtȲÌͳ Ó´lØ·ØÛº Ἰã½é¾`îÀ¸óÁH÷ÂìùàýÄàÅ4Æ€	Çh
È´ÉÊL€€€„ [EMAIL PROTECTED]@[EMAIL PROTECTED]"€€€„ [EMAIL PROTECTED]@[EMAIL PROTECTED] [EMAIL PROTECTED]<[EMAIL PROTECTED] [EMAIL PROTECTED]@€€€‰Ú†HJ†HÚ†H؆H€€€„ [EMAIL PROTECTED]@€€€‰Ú†HJ†HÚ†Hü†H€€€„ [EMAIL PROTECTED]@€€€‰†HJ†H†Hü†H€€€„ [EMAIL PROTECTED]@€€€‰†HJ†H†HÆH€€€„ [EMAIL PROTECTED] [EMAIL PROTECTED]@€€€‰Ð†HJ†HІHÆH€€€„ [EMAIL PROTECTED]@€€€‰Ð†HJ†HІH,†H€€€„ [EMAIL PROTECTED] [EMAIL PROTECTED]	[EMAIL PROTECTED],†H€€€„ [EMAIL PROTECTED]	[EMAIL PROTECTED] [EMAIL PROTECTED]	€€€ˆ‘Íwy{€‚‚€€€ˆ‘Íwy{€‚‚€€€„ [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]@€€€‰›†HJ†H›†H­†H€€€„ [EMAIL PROTECTED]@€€€‰¯†HJ†H¯†H­†H€€€„ [EMAIL PROTECTED]@€€€‰¯†HJ†H¯†H͆H€€€„ [EMAIL PROTECTED][EMAIL PROTECTED] [EMAIL PROTECTED][EMAIL PROTECTED] [EMAIL PROTECTED][EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED](*€€€„ [EMAIL PROTECTED]
[EMAIL PROTECTED] 0Í€€€„ [EMAIL PROTECTED] [EMAIL PROTECTED]|6†Hi6†H|6†Hi6†H€€€„ [EMAIL PROTECTED]@[EMAIL PROTECTED] [EMAIL PROTECTED]@€€€‰‘6†Hi6†H‘6†H‘6†H€€€„ [EMAIL PROTECTED]@€€€‰•6†Hi6†H•6†H‘6†H€€€„ [EMAIL PROTECTED]@€€€‰7†Hi6†H7†H7†H€€€„ [EMAIL PROTECTED] [EMAIL PROTECTED]@$€€€Œ$7†Hi6†H7†H7†Hù†Hé€€€„ [EMAIL PROTECTED]@Ž+ëG€€€„Í [EMAIL PROTECTED],€€€Œ T$pf†HÍÁ¹”€€€„ [EMAIL PROTECTED][EMAIL PROTECTED]
€€€„ [EMAIL PROTECTED]@Ž+ëG€€€„Î [EMAIL PROTECTED](0€€€„ [EMAIL PROTECTED]@[EMAIL PROTECTED]
€€€„ [EMAIL PROTECTED],[EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]@$€€€Œ$ü`‡Hü`‡Hý`‡Hý`‡Hý`‡Hû€€€„ 0Ѐ€€„ [EMAIL PROTECTED]@$€€€Œ$ü`‡Hü`‡Hý`‡Hý`‡Ha‡H+€€€„ [EMAIL PROTECTED]@Ž+ëG€€€„Ð [EMAIL PROTECTED] [EMAIL PROTECTED]@$€€€Œ$4a‡Ha‡H4a‡Hý`‡Ha‡H+€€€„ [EMAIL PROTECTED] [EMAIL PROTECTED] 0Ñ€€€„ [EMAIL PROTECTED]@$€€€Œ$Œa‡Hˆa‡HŒa‡Hˆa‡Ha‡H+€€€„ [EMAIL PROTECTED]@$€€€Œ$Œa‡Hˆa‡HŒa‡Hˆa‡Hss‡H`€€€„ [EMAIL PROTECTED]@Ž+ëG€€€„Ñ [EMAIL PROTECTED]<€€€„ [EMAIL PROTECTED]@$€€€Œ$Œa‡Hˆa‡HŒa‡Hˆa‡Hêw‡H“€€€„ [EMAIL PROTECTED]@Ž+ëG€€€„Ò [EMAIL PROTECTED] [EMAIL PROTECTED]@$€€€Œ$y‡Hy‡HŒa‡Hˆa‡Hêw‡H“€€€„ 0Ó€€€„ [EMAIL PROTECTED]@$€€€Œ$y‡Hy‡Hy‡Hy‡Hêw‡H“€€€„ [EMAIL

Re: [Dovecot] dovecot 1.1.2 assertion

2008-07-24 Thread Timo Sirainen 

On Jul 25, 2008, at 2:36 AM, Nuno Lopes wrote:

I've upgraded to last version of dovecot and am now getting the  
following assertions:


Jul 25 00:00:34 bld3 dovecot: Panic: IMAP([EMAIL PROTECTED]):  
file index-sort-string.c: line 542 (index_sort_add_ids_range):  
assertion failed: (left_idx == 0 && right_idx == rightmost_idx)


Can you reproduce this? Could you send that mailbox's dovecot.index  
and dovecot.index.log files to me? They don't contain any sensitive  
information.




PGP.sig
Description: This is a digitally signed message part

[Dovecot] dovecot 1.1.2 assertion

2008-07-24 Thread Nuno Lopes 

Hi,
 I've upgraded to last version of dovecot and am now getting the 
following assertions:


Jul 25 00:00:34 bld3 dovecot: Panic: IMAP([EMAIL PROTECTED]): file 
index-sort-string.c: line 542 (index_sort_add_ids_range): assertion 
failed: (left_idx == 0 && right_idx == rightmost_idx)


Jul 25 00:00:34 bld3 dovecot: IMAP([EMAIL PROTECTED]): Raw backtrace: 
imap [0x555d30f0] -> imap [0x555d3157] -> imap [0x555d27d8] 
-> imap(index_sort_list_finish_string+0xe8e) [0x555a01de] -> 
imap(index_storage_search_next_nonblock+0x432) [0x5559caa2] -> 
imap(mailbox_search_next+0x21) [0x555a4051] -> imap(imap_sort+0xdc) 
[0x55577d3c] -> imap(cmd_sort+0x33d) [0x55570e1d] -> 
imap(cmd_uid+0x54) [0x55571894] -> imap [0x555722ac] -> imap 
[0x55572362] -> imap [0x55572b7f] -> imap(client_input+0x62) 
[0x55572da2] -> imap(io_loop_handler_run+0xf8) [0x555da0c8] -> 
imap(io_loop_run+0x1d) [0x555d918d] -> imap(main+0x462) 
[0x5557a552] -> /lib64/libc.so.6(__libc_start_main+0xf4) 
[0x2ae31494c154] -> imap [0x5556c859]



in another instance:
Jul 25 00:08:48 bld4 dovecot: Panic: IMAP([EMAIL PROTECTED]): file 
index-sort-string.c: line 607 (index_sort_add_ids_range): assertion 
failed: (skip > 0)
Jul 25 00:08:48 bld4 dovecot: IMAP([EMAIL PROTECTED]): Raw backtrace: 
imap [0x555d30f0] -> imap [0x555d3157] -> imap [0x555d27d8] 
-> imap(index_sort_list_finish_string+0xedc) [0x555a022c] -> 
imap(index_storage_search_next_nonblock+0x432) [0x5559caa2] -> 
imap(mailbox_search_next+0x21) [0x555a4051] -> imap(imap_sort+0xdc) 
[0x55577d3c] -> imap(cmd_sort+0x33d) [0x55570e1d] -> 
imap(cmd_uid+0x54) [0x55571894] -> imap [0x555722ac] -> imap 
[0x55572362] -> imap [0x55572b7f] -> imap(client_input+0x62) 
[0x55572da2] -> imap(io_loop_handler_run+0xf8) [0x555da0c8] -> 
imap(io_loop_run+0x1d) [0x555d918d] -> imap(main+0x462) 
[0x5557a552] -> /lib64/libc.so.6(__libc_start_main+0xf4) 
[0x2b64897fc154] -> imap [0x5556c859]


If this is not enough for you to solve it ...

Some problems were solved upgrading from 1.1.1 but these ones remained.

Thanks,
--
Nuno