Upgrade 6.3 (6.3-5.02) to 6.3 (6.3-8.01) stored crashes after enabling ssl
807574Mar 19 2009 — edited Feb 10 2020We are currently attempting to upgrade our Messaging server from:
Sun Java(tm) System Messaging Server 6.3-5.02 (built Oct 12 2007; 32bit)
libimta.so 6.3-5.02 (built 17:05:22, Oct 12 2007; 32bit)
Linux ripley 2.6.9-78.0.17.ELsmp #1 SMP Thu Mar 5 04:58:34 EST 2009 x86_64 x86_64 x86_64 GNU/Linux
to
Sun Java(tm) System Messaging Server 6.3-8.01 (built Dec 16 2008; 32bit)
libimta.so 6.3-8.01 (built 17:01:59, Dec 16 2008; 32bit)
Linux bourne 2.6.9-78.0.13.ELsmp #1 SMP Wed Jan 7 17:45:52 EST 2009 x86_64 x86_64 x86_64 GNU/Linux
(These are two different systems as we ofcourse want to test this in our test environment first)
The reason for upgrading is the message no longer exist bug and a current problem we are having where mail is being bounced on sunday (why only then we have absolutely no clue)
The odd thing is that when I install the patch everything seems to work fine (didnt complete a full test) until I enable ssl.
(After which
Then the stored crashes and the watcher daemon restarts the processes and gives up after two attempts (within 600 secs).
The watcher log is as follows:
[03/19/09 10:52:59] Connecting to watcher ...
[03/19/09 10:52:59] ERROR: ens failed twice in 600 seconds, will not perform restart
[03/19/09 10:55:00] Received request to restart: store imap http
[03/19/09 10:55:00] Connecting to watcher ...
[03/19/09 10:55:00] http server is not running
[03/19/09 10:55:00] imap server is not running
[03/19/09 10:55:00] Stopping store server 15212 ....[03/19/09 10:55:00] Received request to restart: store imap http
[03/19/09 10:55:00] Connecting to watcher ...
done
[03/19/09 10:55:00] Starting store server .... 16483
[03/19/09 10:55:00] Checking store server status ... ready
[03/19/09 10:55:01] Starting imap server .... 16485
[03/19/09 10:55:01] Starting http server .... 16509
[19/Mar/2009:10:55:01 +0100] bourne [16477]: General Error: cannot connnect to localhost:25: Connection refused
[03/19/09 10:55:01] ERROR: store failed twice in 600 seconds, will not perform restart
Our ssl specific configutil settings are (dont want to flood with a complete dump):
./configutil |grep ssl
encryption.rsa.nssslpersonalityssl = webmail2-acc.amc.nl
service.http.enablesslport = yes
service.http.sslcachesize = 0
service.http.sslport = 443
service.http.sslusessl = yes
service.imap.enablesslport = yes
service.imap.sslcachesize = 0
service.imap.sslport = 993
service.imap.sslusessl = yes
service.pop.enablesslport = no
service.pop.sslcachesize = 0
service.pop.sslport = 995
service.pop.sslusessl = no
Our test environment is a unix copy of the production environment which was modified to fit the correct ip/hostnames in the config files.
Is this a known bug or is there a problem with our test set up ?
Kind regards and much obliged,
Jeffrey
Ps. We still use Messenger Express (eventhough its depricated), is it known when this will be actually removed ?
[edit] We don't use pop at all as such it is not properly configured[edit]
Edited by: jxmulder on Mar 19, 2009 3:52 AM (see [edit][edit]