Correct me if I'm wrong, but won't this cause it to no longer be a real-time solution? The machine I'm running MySQL on is fairly vulnerable to attack (which, for the time being, cannot be changed...although I have secured it as much as is possible), which is why I'm logging everything remotely. If the solution is not real-time, theoretically someone can gain unauthorized access and remove traces of their entry before the logs are batched up and sent. This is what I want to prevent.

Stephen Touset

Harald Fuchs wrote:

In article <[EMAIL PROTECTED]>,
Stephen Touset <[EMAIL PROTECTED]> writes:



However, I have a different need. I'm setting up several machines on a
network, and wish for them all to send logs through stunneled syslog
connections to a central logging server, where I can run logcheck et al
to generate reports of system abuse and ensure that all services are
running smoothly. However, with MySQL logging to it's own file, this
presents a problem to me. Some solutions I can see are simply rsync'ing
the logs over daily or mounting the directory over NFS and copying the
logs. However, both of these require me to set up new services, and/or
change already-existing firewall rules, and write new scripts to do the
fetching/retrieval, all of which takes time and effort--not to mention,
needlessly complicates the system.



How about making the error log a named pipe and reading from it with logger?







--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]



Reply via email to