I've been seeing lots of index cache file errors (using mbox on Solaris
8 sparc 64-bit, but with 32-bit Dovecot) since I switched my account to
Dovecot 1.2.0 (we're still on 1.0.15 mostly, but I'm hoping to upgrade
to 1.1.17+ this summer, or 1.2.x if it's stable enough).
e.g.
Error: Corrupted index cache file
<path>/.imap/INBOX/dovecot.index.cache: field index too large (47 >= 25)
I've tried renaming/removing the .imap/INBOX directory, but the
"Corrupted index cache file" errors return within a day or so.
This morning, Thunderbird got itself in a mess retrying something that
caused multiple core dumps:
> Jul 14 08:52:53 dovecot: Error: child 17694 (imap) killed with signal 10
> (core dumped)
> Jul 14 08:52:54 dovecot: Error: child 17697 (imap) killed with signal 10
> (core dumped)
> Jul 14 08:52:55 dovecot: Error: child 17700 (imap) killed with signal 10
> (core dumped)
Here's a backtrace
> #0 mail_cache_get_record (cache=0x124e28, offset=289, rec_r=0xffbeeb6c)
> at mail-cache-lookup.c:30
> 30 if (rec->size < sizeof(*rec)) {
> (gdb) bt
> #0 mail_cache_get_record (cache=0x124e28, offset=289, rec_r=0xffbeeb6c)
> at mail-cache-lookup.c:30
> #1 0x84e2c in mail_cache_lookup_iter_next_record (ctx=0xffbeeb60)
> at mail-cache-lookup.c:186
> #2 0x84f14 in mail_cache_lookup_iter_next (ctx=0xffbeeb60,
> field_r=0xffbeeb50)
> at mail-cache-lookup.c:219
> #3 0x85650 in mail_cache_lookup_headers_real (view=0x128510, dest=0x129728,
> seq=11416, field_idxs=0xffbeec74, fields_count=2)
> at mail-cache-lookup.c:500
> #4 0x85848 in mail_cache_lookup_headers (view=0x128510, dest=0x129728,
> seq=11416, field_idxs=0xffbeec74, fields_count=2)
> at mail-cache-lookup.c:555
> #5 0x6ba94 in index_mail_get_raw_headers (mail=0x128f20,
> field=0xff142ee8 "Message-ID", value_r=0xffbeecfc)
> at index-mail-headers.c:614
> #6 0x6c084 in index_mail_get_first_header (_mail=0x128f20,
> field=0xff142ee8 "Message-ID", decode_to_utf8=false, value_r=0xffbeede4)
> at index-mail-headers.c:771
> #7 0x7893c in mail_get_first_header (mail=0x128f20,
> field=0xff142ee8 "Message-ID", value_r=0xffbeede4) at mail.c:118
> #8 0xff1419a0 in mail_log_add_hdr (mail=0x128f20, str=0x10edc8,
> name=0xff142d68 "msgid", header=0xff142ee8 "Message-ID")
> at mail-log-plugin.c:277
> #9 0xff141c5c in mail_log_action (dest_trans=0x128f20, mail=0x128f20,
> event=MAIL_LOG_EVENT_DELETE, data=0x0) at mail-log-plugin.c:318
> #10 0xff141e5c in mail_log_mail_update_flags (_mail=0x128f20,
> modify_type=MODIFY_ADD, flags=12) at mail-log-plugin.c:376
> #11 0x78af8 in mail_update_flags (mail=0x128f20, modify_type=MODIFY_ADD,
> flags=12) at mail.c:184
> #12 0x2ab3c in cmd_store (cmd=0x11bec8) at cmd-store.c:192
> #13 0x2b5f8 in cmd_uid (cmd=0x11bec8) at cmd-uid.c:27
> #14 0x2ccc4 in client_command_input (cmd=0x11bec8) at client.c:611
> #15 0x2cb68 in client_command_input (cmd=0x11bec8) at client.c:660
> #16 0x2cd4c in client_handle_next_command (client=0x11bc40,
> remove_io_r=0xffbef23f) at client.c:701
> #17 0x2ce40 in client_handle_input (client=0x11bc40) at client.c:713
> #18 0x2cf78 in client_input (client=0x11bc40) at client.c:752
> #19 0xc6c04 in io_loop_handler_run (ioloop=0x1186d8) at ioloop-poll.c:203
> #20 0xc63f4 in io_loop_run (ioloop=0x1186d8) at ioloop.c:338
> #21 0x369f4 in main (argc=1107968, argv=0xffbef484, envp=0xffbef494)
> at main.c:323
I didn't have these problems with 1.1.x. I guess I ought to get out the
imaptest tool and see if I can reproduce this reliably ...
Best Wishes,
Chris
--
--+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+-
Christopher Wakelin, c.d.wake...@reading.ac.uk
IT Services Centre, The University of Reading, Tel: +44 (0)118 378 8439
Whiteknights, Reading, RG6 2AF, UK Fax: +44 (0)118 975 3094
