Yes, route works fine.
We always use it when monitoring stuff over VPN, to avoid getting 50 e-mails
if the VPN goes down.
But my experience is that conn for the “routed” hosts go yellow, not clear,
if the “router” doesn’t respond to ping. Could be worth considering if you
have alerts configured for yellow.
/Johan
From: Geoff Hallford [mailto:user-dc9e7f30b1e2@xymon.invalid]
Sent: den 6 juli 2010 21:57
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] Depends Question
I have always used "route" and it works perfectly if conn is all you care
about. I have never actually gotten "depends" to work. It doesn't make the
ping go through any host you put in route and it also chains properly to
make it easier. Example:
10.10.10.10 server1 # conn route:network1
20.20.20.20 network1 # conn route:network2
30.30.30.30 network2 # conn
If the route to server1 is: hobbit -> network2 -> network1 -> server1
The above statement works, that if network2 is down ... network1 and server1
go blank :)
On Tue, Jul 6, 2010 at 2:35 PM, <user-ddebaeecde97@xymon.invalid> wrote:
I have nothing to add to this thread, but it looks like "depends" might be
helpful for a problem I am currently having with a pair of servers, whereby
one at a time will show a process as running, the other not. So as long as
one of these machines is running this app, the color should stay green.
.vp
I've not experienced that behavior. Use this for dsl/cable modems and
our routers for 3+ years.
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
“Success is not final, failure is not fatal: it is the courage to
continue that counts.”
--- Winston Churchill
On Tue, Jul 6, 2010 at 1:54 PM, Patrick Nixon wrote:
I read somewhere that if you use route, it does something weird with
the pings so they go through that host or something?
Let me test it and see how it behaves
--patrick
On Tue, Jul 6, 2010 at 1:46 PM, Josh Luthman
wrote:
I'm not certain of this because I've always used FQDN.
If you have FQDN=true and are not using them, maybe the DEPENDS tag
isn't finding eto correct?
In your case, as it is conn, you can use route
1.2.32.4 customer.router.com # testip route:customer.cablemodem.com
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
“Success is not final, failure is not fatal: it is the courage to
continue that counts.”
--- Winston Churchill
On Tue, Jul 6, 2010 at 1:42 PM, Patrick Nixon wrote:
Josh,
Not sure what you were saying there?
My goal is that if eto's conn goes red, that standby's conn should go
clear if it also fails.
Does my depends statement not appear to work in that fashion?
--Patrick
On Tue, Jul 6, 2010 at 1:32 PM, Josh Luthman
wrote:
In stanby's DEPENDS statement referring to eto - not the testing of
the host.
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
“Success is not final, failure is not fatal: it is the courage to
continue that counts.”
--- Winston Churchill
On Tue, Jul 6, 2010 at 12:49 PM, Patrick Nixon wrote:
They're both testip
On Tue, Jul 6, 2010 at 12:41 PM, Josh Luthman
wrote:
Looks right to me, maybe domain related (FQDN)?
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
“Success is not final, failure is not fatal: it is the courage to
continue that counts.”
--- Winston Churchill
On Tue, Jul 6, 2010 at 12:34 PM, Patrick Nixon wrote:
Hey all,
I'm trying to get depends working, but it's not working the way I
expect it to.
10.18.16.172 standby # conn testip
depends=(conn:eto/conn)
10.18.16.174 eto # conn ssh snmp testip
eto is currently red, but standby, which is down as well, is also
red.
anything specific wrong with my depends statement? Any reason why
conn wouldn't be able to depend on something else?
Thanks!
--Patrick