Article ID: 116435, created on Jul 12, 2013, last review on May 9, 2014
- Applies to:
- Operations Automation 5.5
Certain provisioning tasks executing jobs on POA-managed Linux nodes start taking too long to complete. On the system where these tasks execute jobs, POA Agent log file (
/var/log/poa.log) is rather large (several gigabytes or more) and is growing quickly.
The problem may affect usual Linux Shared Hosting services provisioning or APS tasks which involve executing operations on Linux servers, like installation, re-configuration or upgrade of APS application.
One of possible causes can be that POA Agent log level on affected Linux shared hosting server is set to the TRACE value. This causes it to fully log all requests coming from POA. On busy systems with limited I/O capacity this can result in server performance degradation.
First, check the current log level value for POA Agent on the shared hosting server that has performance issues (its IP address and/or hostname may be found in the task properties in POA Task Manager):
~# grep loglevel /usr/local/pem/etc/pleskd.props
If log level is set to 'TRACE', you can change it to one of the less verbose options. Possible values are, in decreasing order of verbosity:
If you have POA 5.2 or later, you can use the following command to change the log level of POA Agent:
~# /usr/local/pem/bin/pleskd_ctl -f /usr/local/pem/etc/pleskd.props setLogLevel <loglevel>
~# /usr/local/pem/bin/pleskd_ctl -f /usr/local/pem/etc/pleskd.props setLogLevel INFO
DEBUG [setLogLevel] ===> ENTRY
DEBUG [AA::PEM_Current::context] ===> ENTRY
DEBUG [AA::PEM_Current::context] <=== EXIT [0.000019]
DEBUG [setLogLevel] <=== EXIT [0.082263]
Change is effective immediately and does not require POA agent restart.
If you have POA 5.1 or earlier, log level of POA Agent can be changed by directly editing the loglevel parameter value in the /usr/local/pem/etc/pleskd.props file on a Linux shared hosting server followed by POA Agent restart.
Follow the steps below to change POA Agent log level.