We are running cyrus-imapd-2.1.14 under Solaris 9. Twice recently, a week apart, we've had an incident where mail delivery stopped, with messages like this being logged:
Mar 2 11:34:29 electra sm-mta[15336]: [ID 801593 mail.crit] j22HYRtu015310: SYSERR(root): Could not connect to socket /var/run/imap/lmtp: Connection refused by localhost Other than those two incidents, the system has been running smoothly for months. In both incidents, the normal number of lmtpd processes (30 to 40) were running. Stopping and restarting Cyrus fixed the problem. This time, I was able to examine some of the processes. Most of them were waiting on a mutex. Here's a typical stack trace: fef9f7a8 lwp_mutex_lock (fee20000) 000c75a0 __db_pthread_mutex_lock (0, fee20000, fec47f68, 2b13b0, 0, 0) + 64 000af170 __db_r_detach (2b1210, 2b1518, fee20000, fee23e00, 0, 2b1210) + 2c 000c6e50 __txn_close (2b1210, 2b1518, 2b1508, 0, 0, 0) + f0 000ac170 __dbenv_refresh (2b1210, 0, 0, 2b1210, 0, 0) + 1c 000ac114 __dbenv_close (2b1210, 2b1210, 2b1210, 0, 0, 0) + 50 0009aac8 done (0, 0, feffa760, 0, 0, 0) + 50 00071d94 duplicate_done (0, 0, 1, 28c990, 0, 2893a0) + 7c 0003e440 shut_down (0, 2b2860, 2b2ac0, 0, 0, 0) + 8 0003a7fc service_abort (0, 2adb10, ffbffaa4, 1a9e0, 3a078, 1) + c 0003a2c0 main (1, ffbffa9c, ffbffaa4, 289000, 0, 0) + d98 00038f20 _start (0, 0, 0, 0, 0, 0) + 108 Does this look familiar to anyone? Is there a solution to this problem? -- -Gary Mills- -Unix Support- -U of M Academic Computing and Networking- --- Cyrus Home Page: http://asg.web.cmu.edu/cyrus Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
