Jim Lawson
2012-Mar-21 13:53 UTC
[Dovecot] dovecot 2.0.19 Panic: file mail-index-sync-ext.c: line 209 (sync_ext_reorder): assertion failed: (offset < (uint16_t)-1)
Had a user who couldn't access his INBOX:> Mar 21 09:21:17 penguina dovecot: imap([USER]): Panic: file > mail-index-sync-ext.c: line 209 (sync_ext_reorder): assertion fai > led: (offset < (uint16_t)-1) > Mar 21 09:21:17 penguina dovecot: imap([USER]): Error: Raw backtrace: > /usr/lib/dovecot/libdovecot.so.0 [0x342683c660] -> /usr > /lib/dovecot/libdovecot.so.0 [0x342683c6b6] -> > /usr/lib/dovecot/libdovecot.so.0 [0x342683bb73] -> > /usr/lib/dovecot/libdovecot > -storage.so.0 [0x3426c966a8] -> > /usr/lib/dovecot/libdovecot-storage.so.0(mail_index_sync_ext_intro+0x240) > [0x3426c979c0] -> / > usr/lib/dovecot/libdovecot-storage.so.0(mail_index_sync_record+0x401) > [0x3426c99151] -> /usr/lib/dovecot/libdovecot-storage.s > o.0(mail_index_sync_map+0x245) [0x3426c99c55] -> > /usr/lib/dovecot/libdovecot-storage.so.0(mail_index_map+0x71b) > [0x3426c8afbb > ] -> /usr/lib/dovecot/libdovecot-storage.so.0 [0x3426c85d8b] -> > /usr/lib/dovecot/libdovecot-storage.so.0(mail_index_open+0x1c > e) [0x3426c8617e] -> > /usr/lib/dovecot/libdovecot-storage.so.0(index_storage_mailbox_open+0xb5) > [0x3426c4d865] -> /usr/lib/dov > ecot/libdovecot-storage.so.0 [0x3426c75eab] -> > /usr/lib/dovecot/libdovecot-storage.so.0 [0x3426c31006] -> > dovecot/imap [hdtod > d 10.245.30.58 SELECT](cmd_Stack trace made it look like it was the INBOX, so I deleted the index files for his INBOX and everything was OK. doveconf -n:> # OS: Linux 2.6.18-274.18.1.el5 x86_64 Red Hat Enterprise Linux Server > release 5.8 (Tikanga) > auth_gssapi_hostname = penguina.uvm.edu > auth_krb5_keytab = /etc/krb5.keytab.dovecot > auth_master_user_separator = * > auth_mechanisms = plain login gssapi > base_dir = /var/run/dovecot/ > default_process_limit = 250 > first_valid_uid = 50 > lock_method = flock > login_trusted_networks = [REDACTED] > mail_location = mbox:~/mail:INBOX=/var/spool/mail/%1u/%1.1u/%u > mail_max_lock_timeout = 30 secs > mail_max_userip_connections = 100 > mbox_read_locks = flock > mbox_write_locks = flock > mmap_disable = yes > namespace { > inbox = yes > location > prefix > separator = / > type = private > } > namespace { > hidden = yes > list = no > location > prefix = mail/ > separator = / > type = private > } > namespace { > hidden = yes > list = no > location > prefix = ~/mail/ > separator = / > type = private > } > namespace { > hidden = yes > list = no > location > prefix = ~%u/mail/ > separator = / > type = private > } > passdb { > args = /etc/dovecot/passwd.masterusers > driver = passwd-file > master = yes > } > passdb { > driver = pam > } > service imap { > process_limit = 4096 > } > service lmtp { > client_limit = 1 > inet_listener lmtp { > port = 24 > } > } > ssl_cert = <[REDACTED] > ssl_key = < [REDACTED] > userdb { > driver = passwd > } > verbose_proctitle = yesAny questions/suggestions welcome. Jim
Timo Sirainen
2012-Mar-21 14:02 UTC
[Dovecot] dovecot 2.0.19 Panic: file mail-index-sync-ext.c: line 209 (sync_ext_reorder): assertion failed: (offset < (uint16_t)-1)
On 21.3.2012, at 15.53, Jim Lawson wrote:> Had a user who couldn't access his INBOX: > >> Mar 21 09:21:17 penguina dovecot: imap([USER]): Panic: file >> mail-index-sync-ext.c: line 209 (sync_ext_reorder): assertion fai >> led: (offset < (uint16_t)-1)I kind of remember that this was fixed by http://hg.dovecot.org/dovecot-2.1/rev/b4d8e950eb9d but I'm not entirely sure. I guess I should have included in the commit the error message it fixed.> Stack trace made it look like it was the INBOX, so I deleted the index > files for his INBOX and everything was OK.If it happens again, get a copy of the indexes.