Slash Boxes
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

use Perl Log In

Log In

[ Create a new account ]

shockme (2685)

  reversethis-{gro.dnuowtixe} {ta} {kcohs}
AOL IM: stephenhargrove (Add Buddy, Send Message)

If you really must have more information about me, see my home node [] at PerlMonks [] or my journal [] at

Journal of shockme (2685)

Tuesday February 19, 2002
10:54 AM

Event Logs and Ear Infections

[ #2955 ]
Yesterday. Sheesh. First, the little guy pops up with an ear infection. So, between the hours of 11:00 and 14:30, I'm running him to the doctor, picking up prescriptions, etc. He's fine, and while I hated to miss work (yes, I love my job), I figure it's better to be safe than sorry. Plus, since I'm salaried and Lori isn't, it'd cost us alot more than the doctor bill for her to take him....

He breathed a huge sigh of relief when he discovered that the doctor wasn't going to give him a shot. It was hilarious.

Once back at the office, I started digging around in Win32::EventLog. I can't believe Activestate hasn't made libwin32 available through PPM yet. Anyway, it was short work getting the module to compile.

I spent the evening writing a script to remotely access our NT/2K servers and report back some statistics on the Event Logs, as well as error and warning messages via Mail::Sendmail. It works like a charm. Beautiful stuff. But, with almost 75 servers, it's information overload. I'm going to have to identify certain error and warning codes that can safely be ignored. That would greatly decrease the amount of information contained in each email, and make "real" problems more obvious.

The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
More | Login | Reply
Loading... please wait.
  • I think there's two kinds of logs: things like daemon logs that you keep around only if you need to read them to diagnose when something goes wrong -- and log analyses, for human consumption. And the log analyses must be very concise.

    One of my few victories at my last job [] was in successfully insisting that the log analyses start out small and comprehensible (rather than comprehensive), and they mostly stayed that way. When "people" asked for pie-charts, and reports on every single user session (times a

    • Heheh. Good point.

      Actually, this project is of my own undertaking. We've got 75 servers supported by 3 people, and I'm hoping that by utilizing this script, it'll help us stay focused on the "important" stuff and, at the same time, possibly give us heads-up on any impending doom. ;)


      If things get any worse, I'll have to ask you to stop helping me.