Path: usenet.cise.ufl.edu!usenet.eel.ufl.edu!www.nntp.primenet.com!nntp.primenet.com!howland.erols.net!news.mathworks.com!uunet!in2.uu.net!news.structured.net!crusty.teleport.com!nntp0.teleport.com!usenet
From: Stunt Pope <[email protected]>
Newsgroups: comp.lang.perl.announce,comp.lang.perl.misc
Subject: Announce: software to monitor remote network services
Followup-To: comp.lang.perl.misc
Date: 8 Nov 1996 18:58:02 GMT
Organization: Private World Communications
Lines: 61
Approved: [email protected] (comp.lang.perl.announce)
Message-ID: <[email protected]>
NNTP-Posting-Host: gadget.cscaper.com
X-Disclaimer: The "Approved" header verifies header information for article transmission and does not imply approval of content.
Xref: usenet.cise.ufl.edu comp.lang.perl.announce:51 comp.lang.perl.misc:7718

A couple of weeks ago I posted to a few newsgroups looking for
software to remotely monitor network services, and fire off an
email message if it found a problem.

i.e. every X minutes it checks ports Y and Z on N hosts
and reports any problems encountered.

Thanks to everyone who got back to me, responses ranged from,
"let me know if you find anything", to "there's a TCL lib
that you could probably use to build one" (it's a network
extension to tcl called Scotty, if interested...)

What i ended up doing is taking my original perl script (basically
a tweaked version of the sockets example out of the camel book)
that was originally run as a cron job, and rewrote it as a daemon
process.

It's been running for almost a week on an offsite server where I
have another shell account and so far, it's been just fine. The
client who needed this has had several real failures in that time
(long story as to why so many of those <gulp>) and this thingy
has detected them all and done the right thing.

I've called this "ptingyd" (until i can think of somehing better),
in honour of my client who always referred to that "pingy thingy".

How it works basically is on start up it composes a list of hosts
and ports to monitor. One of these can be designated the gateway,
so if it finds the gateway is down, it won't bother attempting
to check the hosts behind it until it comes back up.

Each item in the checklist can have it's own threshold (number
of consecutive fails before an error email is sent) and it's
own list of email addr's to report the error to. There is
also a maximum number of messages, so it'll stop sending
messages once that maximum is reached (but it'll keep counting
the fails until it successfully gets through).

It logs all failures, and if you send it a HUP it'll dump a table
that shows the current status of all the checks. If someone
planned to have this notify an email-to-pager gateway (we do)
it can also deal with the difference between alphanumeric and
numeric pagers (i.e. it'll just send a numeric code to the
latter, but the actual text of the error msg to the former).

I've packed it up into a distribution, anyone's welcome to download
it, use it, and contribute with enhancements or bug reports. I am
also interested in hearing how anyone who uses it makes out.

It's available from: http://blammo.shmooze.net/~markjr/ptingyd/

regards, markjr

--
----------------------------------------------------------------------
||| mark jeftovic    SysAdmin  ||  http://www.shmOOze.net/~markjr  |||
||| p r i v a t e    w o r l d ||  http://cool.as.fuck.shmOOze.net |||
----------------------------------------------------------------------
              One-time Winner of the Useless Cat Award!!

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.