Jeff Waugh <[EMAIL PROTECTED]> wrote: > <quote who=3D"Ken Foskey"> > >> I have a lot of servers and data is flying everywhere in a nice >> untraceable manner. I was thinking of a logging daemon on every machin= > e >> that caches messages from each and then forwarded those messages to a >> central daemon that will report on what is happening. > > syslog is awesome for doing this, but on top of that...
I was thinking it might be fun to play with backing syslog with tdbrepl, but only because I worked on the latter code once before and it was the first thing that popped into my head. http://tdbrepl.inodes.org/ >> Error tracking like nothing has happened for 2 hours, something broke. >> Simple status reporting. Where are things now and should I be worried. >>=20 >> Of course netowrk links are not perfect. A truck and a Telstra box met >> taking out one site for a day for example. So I want the daemons to ca= > che >> messages and forward them when they are able to. >>=20 >> Now I was thinking that this is probably one of those workflow middlewa= > re >> tools that handle this sort of stuff. Anyone have any ideas about these >> things. > > Pulling it all together with beauty and ease - Splunk <www.splunk.com>. > > (This is not very coders though.) -- Horms http://www.vergenet.net/~horms/ _______________________________________________ coders mailing list [email protected] http://lists.slug.org.au/listinfo/coders
