We're running a small LUA script in the NAS which attempts get the list of alarms using the alarm.list method. Although there are about 2500+ alarms, the alarm.list method sometimes returns an empty list -- and sometimes returns the complete list. When it fails (about 50% of the time), it's failing silently (no errors in the nas.log). I've tried turning up the log level to the highest setting, but still see no log messages pertaining to this script and problems reading the alarm DB.
The script runs every minute and has a seemingly random failure pattern: success, fail, fail, success, success, fail, fail, fail, succes, success, fail, success, etc. Could this be a DB contention issue? Does alarm.list fail silently if the DB is locked?
Does anyone have any other ideas about what might cause this behavior? And suggestions on how I might debug this would be appreciated.