Thanks, you were right, I had forgotten that there would be indexes for
each folder (of course). So I had done a 'find -maxdepth' type command
to find and delete them, and only hit the ones in Maildir. Now that I
have deleted all of them, there are no more errors.
On 01.03.2018 17:54, Aki Tuomi wrote:
Hi!
The problem seems to occur during mail index processing. Can you try
deleting all dovecot.index* files under this user's maildir. There
might
be some more under various folders.
Aki
On 03.01.2018 08:51, Owen B. Mehegan wrote:
The EC2 instance that I run Dovecot on (personal mail server) was
forcibly "retired" yesterday, and as a result I ended up rebuilding
from backups. In the process I switched from Ubuntu Precise to Ubuntu
Xenial (12.04 to 16.04), and this in turn caused me to upgrade Dovecot
from 2.0.19 to 2.0.22. Now, although accessing mail seems to work OK
using my various clients, I am seeing these errors in my logs:
Jan 3 05:56:58 -- dovecot: imap(owen): Panic: file
mail-index-sync-keywords.c: line 227 (keywords_update_records):
assertion failed: (data_offset >= sizeof(struct mail_index_record))
Jan 3 05:56:58 -- dovecot: imap(owen): Error: Raw backtrace:
/usr/lib/dovecot/libdovecot.so.0(+0x7aee2) [0x7fe9c08dbee2] ->
/usr/lib/dovecot/libdovecot.so.0(+0x7afcd) [0x7fe9c08dbfcd] ->
/usr/lib/dovecot/libdovecot.so.0(i_fatal+0) [0x7fe9c0880a61] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mail_index_sync_keywords+0x828)
[0x7fe9c0c1fdd8] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mail_index_sync_record+0xef)
[0x7fe9c0c2079f] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mail_index_sync_map+0x238)
[0x7fe9c0c21758] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mail_index_map+0x71)
[0x7fe9c0c18de1] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0xd28fd)
[0x7fe9c0c2e8fd] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0xd2f03)
[0x7fe9c0c2ef03] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mail_index_open+0x78)
[0x7fe9c0c2efe8] ->
/usr/lib/dovecot/libdovecot-storage.so.0(index_storage_mailbox_open+0x92)
[0x7fe9c0c08042] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0x5e8d2)
[0x7fe9c0bba8d2] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0x5e9b8)
[0x7fe9c0bba9b8] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0x37b86)
[0x7fe9c0b93b86] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mailbox_open+0x20)
[0x7fe9c0b93d00] ->
/usr/lib/dovecot/libdovecot-storage.so.0(index_storage_get_status+0x20)
[0x7fe9c0c073b0] ->
/usr/lib/dovecot/libdovecot-storage.so.0(mailbox_get_status+0x37)
[0x7fe9c0b940e7] -> dovecot/imap(imap_status_get+0x77)
[0x55d3b4c3f977] -> dovecot/imap(cmd_status+0x1a9) [0x55d3b4c32969] ->
dovecot/imap(command_exec+0x9c) [0x55d3b4c3806c] ->
dovecot/imap(+0x193e2) [0x55d3b4c363e2] -> dovecot/imap(+0x1946c)
[0x55d3b4c3646c] -> dovecot/imap(client_handle_input+0x165)
[0x55d3b4c36825] -> dovecot/imap(client_input+0x82) [0x55d3b4c36db2]
-> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x4c)
[0x7fe9c08efc6c] ->
/usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0x101)
[0x7fe9c08f1101] ->
/usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x25)
[0x7fe9c08efcf5] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x38)
[0x7fe9c08efe98]
Jan 3 05:56:59 -- dovecot: imap(owen): Fatal: master: service(imap):
child 18289 killed with signal 6 (core dumped)
I have deleted the dovecot.* files in the Maildir for all users (did
this while Dovecot was stopped, then started it again), but the errors
persist. They are especially frequent for one particular user. I can't
figure out where the core file is, it doesn't seem to be in my home or
in /var/run/dovecot
Here's my dovecot -n output:
# 2.2.22 (fe789d2): /etc/dovecot/dovecot.conf
# Pigeonhole version 0.4.13 (7b14904)
# OS: Linux 4.4.0-1043-aws x86_64 Ubuntu 16.04.3 LTS
mail_location = maildir:~/Maildir
namespace inbox {
inbox = yes
location =
mailbox Drafts {
special_use = \Drafts
}
mailbox Junk {
special_use = \Junk
}
mailbox Sent {
special_use = \Sent
}
mailbox "Sent Messages" {
special_use = \Sent
}
mailbox Trash {
special_use = \Trash
}
prefix =
}
passdb {
driver = pam
}
protocols = " imap"
service auth {
unix_listener /var/spool/postfix/private/auth {
group = postfix
mode = 0660
user = postfix
}
}
ssl_cert = </etc/letsencrypt/live/mydomain/cert.pem
ssl_key = </etc/letsencrypt/live/mydomain/privkey.pem
userdb {
driver = passwd
}
Thanks in advance for any support you can offer!
