We are getting the following server log file and it seems like based on Google search, the process is still writing to the old log file. Why would it still be writing to an old log file if we have it set up to Verbose logging and how can we disable this so the appserver process does not still stay connected to the log file? How does this happen?
[17/10/30@14:32:05.797-0500] P-008380 T-006720 1 AS – (Procedure: ‘handleError jms/impl/errorhndlr.p’ Line:165) javax.jms.JMSException: The broker is not accepting messages at this priority
[17/10/30@14:32:05.797-0500] P-008380 T-006720 1 AS – (Procedure: ‘si/sim130-out.p’ Line:2200) Multi-Company Server Process is implementing flow control. (2)
It looks like by default the server log files are on a cyclic logging process without verbose logging enabled. What is the size limit of a server log before it creates a new cyclic log file and how many log files does it generate before it starts over and writes over the first log file? Where are these settings set up to view?
Assuming you are on E9. These settings are in the Management Console for Progress. You select your appserver and then click on configuration and then the edit button. You can modify the size of the files and the number of files it creates.
Our settings are set to 15MB with a max server log files of 10. What happens when the max number of server log is reached and it starts the cycle over by writing to an old server log? I think this is what’s causing the issue when the max number of logs is reached and then it writes to an existing old log which then generates the error messages. Is this correct or “working as designed”?
aidacra
(Nathan your friendly neighborhood Support Engineer)
5
Correct–overwrites the oldest when max log files is reached.
NOTE: If a process starts to write to a log file, it will always write to that log file even if that log file has reached the threshold size–that is why the threshold size isn’t strictly honored. It simply prevents new processes from writing to it, and a new log file is generated.
We are noticing that when it overwrites the oldest log file, it will contain the error messages below. Anyone noticing this happening too or this limited to multi-company?
[17/10/30@14:32:05.797-0500] P-008380 T-006720 1 AS – (Procedure: ‘handleError jms/impl/errorhndlr.p’ Line:165) javax.jms.JMSException: The broker is not accepting messages at this priority
[17/10/30@14:32:05.797-0500] P-008380 T-006720 1 AS – (Procedure: ‘si/sim130-out.p’ Line:2200) Multi-Company Server Process is implementing flow control. (2)
aidacra
(Nathan your friendly neighborhood Support Engineer)
7
The errors for that Sonic process didn’t exist previously until the log files started overwriting the old log files. It only exists in the overwritten old log file. Is this related to the log file process or the Sonic process? I’m not sure how to associated the issue if this is a Sonic issue or a server log issue due to being overwritten?
aidacra
(Nathan your friendly neighborhood Support Engineer)
11
Our E905 Support team would definitely be able to assist you with that.
Are there any concerns if we reduce the max number of server log files down from 10 to 3 without restarting the task agent? If we reduce the log size on Friday and restart the task agent on Sunday over the weekend, any concerns? Ideally we’d like to make this change at the same time we are restarting the task agent.
aidacra
(Nathan your friendly neighborhood Support Engineer)
14
The change will not take effect until the restart, but, you can make the change now and restart later.
