Gabriel Gunderson:
I wrote a Python script (soooo much easier than perl :-) ) that uses ping but also has a backoff algorithm built in, (only does what it needs to on a 1, then 5, then 10 etc. on up to 40 min. interval). I also just use cron to run the script every minute. With an RPS-10 (about $90) hooked to the COM port on the server, it actually reboots the DSL Modem if the ping fails via an expect script. I also have cron save a log file (with ">>") each time so we can look back and see if and when the connection is going down. If I had time I would probably update it to use a databse. If you wanted to watch for a dead gateway in real time, you would just have to put what I have done in a while loop and run it (no cron). In short, ping is the only way I know of to watch for a dead gateway, unless you want to so some SNMP level programming, (which Python has a module for).
Let me know if you would like me to send it to you,
-Ed
P.S.:
I have some asterisk questions, (IPCentrex, Voice Mail, Call routing, H323<-->SIP translation etc.), any Asterisk experts out there?
Gabriel Gunderson wrote:
I'm setting up a box with two connections going out to different ISPs for a small business in town. The second was just for a backup but they wanted to be able add its bandwidth to that of the first for out going traffic. On a failure the traffic goes out the one connection.
I think I have it all working but I'm still looking for an elegant way to detect a dead gateway. I have a cron job that uses ping to check each gw. It works but it's a little clunky and only checks every minute.
What have you all done to detect dead gateways?
Thanks, Gabe
.===================================. | This has been a P.L.U.G. mailing. | | Don't Fear the Penguin. | | IRC: #utah at irc.freenode.net | `==================================='
.===================================. | This has been a P.L.U.G. mailing. | | Don't Fear the Penguin. | | IRC: #utah at irc.freenode.net | `==================================='
