[linux-elitists] A better cron: large production environments

Karsten M. Self karsten at linuxmafia.com
Sat May 8 13:20:04 PDT 2010


on Fri, May 07, 2010 at 11:32:20PM -0700, George Georgalis (george at galis.org) wrote:
> On Fri 07 May 2010 at 07:57:22 PM -0700, Karsten M. Self wrote:
> >
> >That wrapper itself would be what cron would invoke, the wrapper would
> >run the actual job (and handle the notifications/logging) independently
> >of cron.
> >
> 
> Don't forget it needs to bail if the last invocation didn't finish...

Good point.  Lockfiles.  We're having a little issue with this ATM.

Doesn't particularly help that swap is limited to 128KB.  Yeah,
kilobytes.  We burn through that real quick.

> >This makes migrating numerous legacy scripts fairly transperant -- we're
> >adding one layer of infrastructure, but it's pretty light.
> 
> Seems a lot more bang for the buck to make wrappers for legacy
> scripts so they behave like logging tests that fit into the mon
> invocation / alert schema. IMHO.

Need to look at mon to see how it's doing that, but sounds like a good
foundation or model of rthe monitoring componenet.  We're aready using
Nagios, so I'm leaning that way.


Peace.

-- 
Karsten M. Self <karsten at linuxmafia.com>        http://linuxmafia.com/~karsten
 What part of "gestalt" don't you understand?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://zgp.org/pipermail/linux-elitists/attachments/20100508/e115b317/attachment.pgp>


More information about the linux-elitists mailing list