Replication Error
Hi, in our online.log we see sometimes messages like 00:01:00 CDR Schedule Manager initialization complete. 00:01:00 CDR GC: catalog recovery begin 00:01:01 CDRGC: server recovery complete 00:01:11 CDR GC: replicate recovery complete 00:01:18 CDR GC:replicate pending state recovery complete 00:01:18 CDR GC: group recovery complete 00:01:18 CDR GC: group pending state recovery complete 00:01:18 CDR GC: catalog recovery complete 00:01:26 CDR queuer initialization complete 00:01:26 CDR Grouper: replay logical log ID (3768) is not present 00:01:26 CDR Grouper Fan Out thread is aborting. 00:01:26 CDR Grouper FanOut thread is aborting. Does anyone know what the fat lines message mean? During the workday the replication is writing lots of data in the Logs, but no Users are working on the maschine. Also in the night, replication fills the Logs: 00:02:16 Logical Log 4695 Complete. 00:02:29 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4695 Complete." "Logical Log 4695 Complete." 00:02:55 Logical Log 4696 Complete. 00:03:07 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4696 Complete." "Logical Log 4696 Complete." 00:03:36 Logical Log 4697 Complete. 00:03:48 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4697 Complete." "Logical Log 4697 Complete." 00:04:16 Logical Log 4698 Complete. 00:04:28 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4698 Complete." "Logical Log 4698 Complete." 00:04:56 Logical Log 4699 Complete. 00:05:09 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4699 Complete." "Logical Log 4699 Complete." 00:05:37 Logical Log 4700 Complete. 00:05:49 Process exited with return code 151: /bin/sh /bin/sh -c /home2/informix/etc/log_full.sh 2 23 "Logical Log 4700 Complete." "Logical Log 4700 Complete." If noone is working in the machine, where does it come from? Thank you Christian
Christain, You have encountered a log wrap. This means that the current log position has overrun the replay log position. I suspect that this is caused by the size of your logical log files. You are crossing a log boundry about every 45 seconds. With ER you need to configure large log files. There is some administrative work that happenes every time the log file boundry is crossed and some of that work does require usage of the logs themselves.. I suspect that your log files are configured so small that this work is causing the logs to advance to the point that another log file is crossed, which causes more administrative work to be done, which causes another log file to be crossed which... - well you see the point. Generally when I'm asked about configuring a system for ER, I make the following suggestions: 1) Try to avoid crossing a log boundry more than once an hour 2) Try to make all of the log files about the same size 3) Try to have 24 hours worth of log file space 4) Try to put the LTXHWM and LTHEHWM arround 30/60 throuch 40/70. Madison Pruet christianw@tronet.de wrote: > Hi, > > in our online.log we see sometimes messages like > > 00:01:00 CDR Schedule Manager initialization complete. > 00:01:00 CDR GC: catalog recovery begin > 00:01:01 CDRGC: server recovery complete > 00:01:11 CDR GC: replicate recovery complete > 00:01:18 CDR GC:replicate pending state recovery complete > 00:01:18 CDR GC: group recovery complete > 00:01:18 CDR GC: group pending state recovery complete > 00:01:18 CDR GC: catalog recovery complete > 00:01:26 CDR queuer initialization complete > 00:01:26 CDR Grouper: replay logical log ID (3768) is not present > 00:01:26 CDR Grouper Fan Out thread is aborting. > 00:01:26 CDR Grouper FanOut thread is aborting. > > Does anyone know what the fat lines message mean? > > During the workday the replication is writing lots of data in the Logs, > but no Users are working on the maschine. Also in the night, replication > fills the Logs: > > 00:02:16 Logical Log 4695 Complete. > 00:02:29 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4695 Complete." > "Logical Log 4695 Complete." > 00:02:55 Logical Log 4696 Complete. > 00:03:07 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4696 Complete." > "Logical Log 4696 Complete." > 00:03:36 Logical Log 4697 Complete. > 00:03:48 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4697 Complete." > "Logical Log 4697 Complete." > 00:04:16 Logical Log 4698 Complete. > 00:04:28 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4698 Complete." > "Logical Log 4698 Complete." > 00:04:56 Logical Log 4699 Complete. > 00:05:09 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4699 Complete." > "Logical Log 4699 Complete." > 00:05:37 Logical Log 4700 Complete. > 00:05:49 Process exited with return code 151: /bin/sh /bin/sh -c > /home2/informix/etc/log_full.sh 2 23 "Logical Log 4700 Complete." > "Logical Log 4700 Complete." > > If noone is working in the machine, where does it come from? > > Thank you > Christian