Diagnostic Tool for Server?

[delfinparis]

We've had the same CentOS virtual server (from GoDaddy) for around 3 years. I upgrade the mysql and php every so often, but I'm sure things have been installed over the years that cause issues. Once in awhile something will fail, or like today, the SQL server went down for some reason.

I know very little about reading error logs, but is there any piece of software I could install that can just do a general scan of the server looking for obvious issues?

I'm probably not the guy that can fix anything, but then I could find someone to go in and fix what's wrong.

If anyone has suggestions on how to assess the overall health of the server and how it's running, I would greatly appreciate it!

[aks]

Yes it's called a system administrator.

[delfinparis]

Yes it's called a system administrator.

I'm unfamiliar with your "technical jargon" and am headed to the library to look up the definition of "system administrator."

Thanks for the software recommendation!

[gerald_clark]

You need to hire one, not read about one.

[delfinparis]

You need to hire one, not read about one.

Yeah, I was trying to make a joke.

I understand I need to hire an admin. I'm happy to do that. But I'd like to know what general bugs issues exist so I can hold that person accountable.

Not trying to be rude, but how is this unclear? I'm just asking if there's a piece of software that can scan the server looking for errors and issues.

[TrevorH]

Not trying to be rude, but how is this unclear? I'm just asking if there's a piece of software that can scan the server looking for errors and issues.

If there was something like that then half this board would be out of work and whoever wrote it would be very rich

[AlanBartlett]

Well there is logwatch which will do as its name implies, once a day. The log files examined are those in the /var/log/ directory.

[aks]

I guess (and ignoring my initial flippantly), it depends on what your machine is doing. So the majority of errors will be in the logs files in /var/log/ (as Alan suggested). Errors there may or may not need attention (i.e.: some may be red herrings). Additionally, logs from the application(s) may indicate problems. I guess the key thing here is interpreting the errors and taking appropriate measures. That's what a (half decent) system administrator does. I think most businesses only consider the application level stuff - so does the thing I'm selling (or the thing that's supporting what I'm selling), work in the way that (usually) the company's sales staff "think" (heh, like that's not an oxymoron) it should. So if you hire somebody and they can't deliver what you expect, then that's how you'd hold them accountable. This also gets into stuff like have you laid out what's expected, does the recipient actually understand and so on.

If you're really interested in what bugs exist (rather than implementation/configuration problems), just google for CVE.