"system.log" filling log disk - SQL dead?

Hi everyone,

ever since I have upgraded to 8.301 and then 8.302 (shortly after oneanother) I have been having problems with high CPU utilization for postgresql. I have given up on fixing it, but it tells me something is certainly wrong.

Over the past week another problem has presented itself - every now and then I will suddenly find the 16GB log disk completely full - where normally the disc usage is below 15%. Interesting thing is that lookijng at the e-mail warning messages the log disk fills up in a matter of about 2 to 3 hours...

WHen this happens it turns out that it is the "system.log" file located in /var/log that fills up the disc. It happened again today, after I only restarted the system yesterday...

The "system.log" file had grown to fill the disc - over 14GB in size. I have captured the last 2000 lines using "tail --lines=2000 /var/log/system.log >/var/storage/2000lines.txt so that I could see what was going on. The log file is flooded with repeats of the following lines:

2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: 1. main::sqlexecute:284() adbs-maintenance.pl
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: 2. main::stepwise_cleaning:232() adbs-maintenance.pl
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: 3. main::clean_mailanon:98() adbs-maintenance.pl
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: 4. main::main:53() adbs-maintenance.pl
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: 5. main::top-level:460() adbs-maintenance.pl
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: |================================================= ========================
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]: statement failed: no connection to the server
2012:05:07-12:09:08 plumbum adbs-maintenance[4326]:

So, looks to me that the adbs-maintenance process is loosing its connection to the server - but what server? postgresql?

Anybody with some tips on how to fix this one?

Martijn.