Page 1 of 1
Scalix-postgres Failure
Posted: Wed Mar 07, 2007 6:32 pm
by GCamp
I have installed Scalix on a SUSELINUX v9 system. On startup I get a failure indication when Scalix-Postgres Service attempts to load.
The pgstartup.log has the following information:
Code: Select all
2007-03-07 15:12:33 LOG: database system was shut down at 2007-03-07 15:12:32 EST
2007-03-07 15:12:33 LOG: checkpoint record is at 0/9B1058
2007-03-07 15:12:33 LOG: redo record is at 0/9B1058; undo record is at 0/0; shutdown TRUE
2007-03-07 15:12:33 LOG: next transaction ID: 536; next OID: 17142
2007-03-07 15:12:33 LOG: database system is ready
2007-03-07 15:12:42 LOG: received fast shutdown request
2007-03-07 15:12:42 LOG: shutting down
2007-03-07 15:12:45 LOG: database system is shut down
2007-03-07 15:12:45 LOG: database system was shut down at 2007-03-07 15:12:45 EST
2007-03-07 15:12:45 LOG: checkpoint record is at 0/9BD498
2007-03-07 15:12:45 LOG: redo record is at 0/9BD498; undo record is at 0/0; shutdown TRUE
2007-03-07 15:12:45 LOG: next transaction ID: 542; next OID: 17143
2007-03-07 15:12:45 LOG: database system is ready
2007-03-07 15:13:16 LOG: received fast shutdown request
2007-03-07 15:13:16 LOG: shutting down
2007-03-07 15:13:19 LOG: database system is shut down
2007-03-07 15:13:22 LOG: database system was shut down at 2007-03-07 15:13:19 EST
2007-03-07 15:13:22 LOG: checkpoint record is at 0/9BD4D8
2007-03-07 15:13:22 LOG: redo record is at 0/9BD4D8; undo record is at 0/0; shutdown TRUE
2007-03-07 15:13:22 LOG: next transaction ID: 542; next OID: 17143
2007-03-07 15:13:22 LOG: database system is ready
2007-03-07 15:16:22 LOG: received fast shutdown request
2007-03-07 15:16:22 LOG: aborting any active transactions
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 LOG: shutting down
2007-03-07 15:16:26 LOG: database system is shut down
2007-03-07 15:20:28 LOG: could not bind IPv4 socket: Cannot assign requested address
HINT: Is another postmaster already running on port 5733? If not, wait a few seconds and retry.
2007-03-07 15:20:28 FATAL: could not create listen socket for "xxx.xxx.xxxxxx.org"
I have searched the forums and have not found anything that helped.
As far as I know I do not have another postmaster running.
Posted: Wed Mar 07, 2007 6:34 pm
by kanderson
Can you type
/etc/init.d/scalix-postgres stop
Then give us the output from
ps auwwx|grep post
as well as
cat /var/opt/scalix/??/postgres/pgstartup.log
Thanks
Kev.
Posted: Wed Mar 07, 2007 8:26 pm
by GCamp
Kev.
Thanks for the quick response.
Here is the response to
Code: Select all
mail:~ # /etc/init.d/scalix-postgres stop
Stopping scalix-postgres service (mail) : failed
Then from the command
I get
Code: Select all
root 15890 0.0 0.0 1712 644 pts/0 R+ 18:58 0:00 grep postfrom
Code: Select all
cat /var/opt/scalix/??/postgres/pgstartup.log I get
Code: Select all
2007-03-07 15:12:33 LOG: database system was shut down at 2007-03-07 15:12:32 EST
2007-03-07 15:12:33 LOG: checkpoint record is at 0/9B1058
2007-03-07 15:12:33 LOG: redo record is at 0/9B1058; undo record is at 0/0; shutdown TRUE
2007-03-07 15:12:33 LOG: next transaction ID: 536; next OID: 17142
2007-03-07 15:12:33 LOG: database system is ready
2007-03-07 15:12:42 LOG: received fast shutdown request
2007-03-07 15:12:42 LOG: shutting down
2007-03-07 15:12:45 LOG: database system is shut down
2007-03-07 15:12:45 LOG: database system was shut down at 2007-03-07 15:12:45 EST
2007-03-07 15:12:45 LOG: checkpoint record is at 0/9BD498
2007-03-07 15:12:45 LOG: redo record is at 0/9BD498; undo record is at 0/0; shutdown TRUE
2007-03-07 15:12:45 LOG: next transaction ID: 542; next OID: 17143
2007-03-07 15:12:45 LOG: database system is ready
2007-03-07 15:13:16 LOG: received fast shutdown request
2007-03-07 15:13:16 LOG: shutting down
2007-03-07 15:13:19 LOG: database system is shut down
2007-03-07 15:13:22 LOG: database system was shut down at 2007-03-07 15:13:19 EST
2007-03-07 15:13:22 LOG: checkpoint record is at 0/9BD4D8
2007-03-07 15:13:22 LOG: redo record is at 0/9BD4D8; undo record is at 0/0; shutdown TRUE
2007-03-07 15:13:22 LOG: next transaction ID: 542; next OID: 17143
2007-03-07 15:13:22 LOG: database system is ready
2007-03-07 15:16:22 LOG: received fast shutdown request
2007-03-07 15:16:22 LOG: aborting any active transactions
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 FATAL: terminating connection due to administrator command
2007-03-07 15:16:22 LOG: shutting down
2007-03-07 15:16:26 LOG: database system is shut down
2007-03-07 15:20:28 LOG: could not bind IPv4 socket: Cannot assign requested address
HINT: Is another postmaster already running on port 5733? If not, wait a few seconds and retry.
2007-03-07 15:20:28 FATAL: could not create listen socket for "mail.int.wmpress.org"
Posted: Wed Mar 07, 2007 8:38 pm
by kanderson
If you restart postgre is there anything in the log then?
/etc/init.d/scalix-postgres start
Kev.
Posted: Wed Mar 07, 2007 8:58 pm
by GCamp
Kev
When I started Scalix-postgres I got the following in the log file:
Code: Select all
HINT: Is another postmaster already running on port 5733? If not, wait a few seconds and retry.
2007-03-07 15:20:28 FATAL: could not create listen socket for "mail.int.wmpress.org"
2007-03-07 19:52:14 LOG: database system was shut down at 2007-03-07 15:16:26 EST
2007-03-07 19:52:14 LOG: checkpoint record is at 0/9BD518
2007-03-07 19:52:14 LOG: redo record is at 0/9BD518; undo record is at 0/0; shutdown TRUE
2007-03-07 19:52:14 LOG: next transaction ID: 552; next OID: 17143
2007-03-07 19:52:14 LOG: database system is ready
Posted: Wed Mar 07, 2007 9:01 pm
by GCamp
Kev,
Could the problem be a timing issue during the startup sequence causing the error?
Posted: Thu Mar 08, 2007 10:34 am
by GCamp
Problem Resolved.
For the sake of anyone else that may run into this, here is what I did to resolve the problem.
During the bootup sequence I noticed that the scalix-postgres failure occurred before the network was brought up. Somewhere in my reading about postgres it was stated that "of course the network has to be active for postgres to work"
I discovered the following in /etc/init.d/rc5.d: (not a complete listing)
Code: Select all
S01random ...
S02scalix-postgres ...
S02scalix-tomcat ...
S05network ...
S06syslog ...This seemed to correspond to the sequence that I was seeing in the startup sequence.
Assuming that the numbers were used by the script to launch the modules, I made the following modifications:
Code: Select all
S01random ...
S02network ...
S03scalix-postgres ...
S03scalix-tomcat ...
S06syslog ...Upon reboot the network started followed by scalix-postgres which started without error
I have reinstalled SUSE and Scalix multiple times and have seen the postgres failure each time. Can someone answer the questions 1. was what I did the proper thing to do or will it cause problems elsewhere; 2. why was this necessary?
Posted: Thu Mar 08, 2007 11:13 am
by kanderson
What you did should work just fine.
Kev.
Posted: Thu Mar 08, 2007 11:17 am
by GCamp
Kev,
Thanks for your help. Sorry that I wasn't as quick in responding to your replies as you were to mine.
Posted: Thu Mar 08, 2007 12:06 pm
by kanderson
No Prob. Glad you figured that out, it would be brutal to troubleshoot from here. :)
Kev.
Posted: Mon Apr 16, 2007 10:18 am
by GCamp
A new twist to the original problem has surfaced since I first posted this. I have had instances after rebooting that the changes that I made to the /etc/init.d/rc5.d file have reverted to the original order. It doesn't happen every time but does seem to be related to whatever GUI or process overwrites this file. It would be nice to NOT have to re-modify this file after a reboot.