System configuration

compared with
Key
This line was removed.
This word was removed. This word was added.
This line was added.

Changes (1)

View Page History
{gliffy:name=Log filesystems|size=M|version=4}

h2. Tuning your system

You will find in this chapter some tweaks of your system to improve performances of Petals ESB container(s) running on.

h3. File systems

Petals ESB does not really require HDD I/O except when:
* installing JBI artifacts: that does not require a dedicated configuration of file systems because the installation of a JBI artifact is exceptional,
* or logging information: several log traces are generated during each service execution, so the write of log traces in files will impact the service execution performances.

The following log traces can be generated, and each have special characteristics that can influence the choose of the file system managing the file containing the log traces:
* global log traces flushes in {{${petals.log.dir}/petals.log}}: few traces are written in a weighty file that can become very large,
* monitoring traces, including payload dumps of messages, are written as small files in a directory that will contain a lot of sub-directorie. So the file-system must be effective with small files and a bug number of sub-directories.




{note}This configuration requires to [change the default logging configuration|petalsesbsnapshot:Logging in Petals].{note}
{tip}Don't forget to monitor your file system sizes with a monitoring tool as Nagios{tip}