We are facing a weird issue in our Weblogic environments -
OHS crashes every 7th day and won't restart. We need to kill all the processed and start it. This schedule has been surprising to us as the environment was built around 8 months back but the issue started occurring in December.
In December a cronjob was written to delete unwanted log files which read as follows -
find /u01/app/orasec/admin/prod_oam_domain/aserver/prod_oam_domain/servers/AdminServer/logs/ -name "*.out0**" -mtime +5 -exec rm -f {} \;
find /u01/app/orasec/admin/prod_oam_domain/mserver/prod_oam_domain/servers/oam_server1/logs/ -name "*.out0**" -mtime +5 -exec rm -f {} \;
find /u01/app/orasec/admin/ohs_instance1/diagnostics/logs/OHS/ohs1 -mtime +5 | xargs rm -Rf
We noticed that few important files are housed in /u01/app/orasec/admin/ohs_instance1/diagnostics/logs/OHS/ohs1 folder :
htpd.pid
ObAccessClient.xml.lck
oblog.log.lck
polltracking.lck
And per the script these will be deleted once there are 6 or more days older.
I know that a missing httpd.pid/ *.lck when OHS is in running state would prevent startting ohf OHS process.
But I'm not sure on whether this could have caused a 7 day periodic OHS crash - deletion of any of the above files ?