Purple Interval

4 messages in this thread

list John Tullis · Fri, 11 Mar 2016 16:41:20 +0000 ·

Is it possible to shorten the purple for the entire server? We have servers that don't have ICMP access so when the server goes offline, we rely on the purple alert to let us know that the server is down. With the 30 minute delay, I'm not getting those alerts until much after the server needs to be back up. Is it possible to change the delay so I would know sooner if the server is down?

John Tullis

list John Thurston · Fri, 11 Mar 2016 07:54:58 -0900 ·

On 3/11/2016 7:41 AM, John Tullis wrote:

Is it possible to shorten the purple for the entire server?

Modify the xymon environment variable STATUSLIFETIME

 From the man-page xymonserver.cfg :

STATUSLIFETIME
    The number of minutes that a status is considered valid after an update. After this time elapses, the status will go purple. Default: 30 minutes

Alternatively, define an ext script on your inaccessible host to send a 
status message in every X minutes with an appropriate TTL

status[+LIFETIME][/group:GROUP] HOSTNAME.TESTNAME COLOR <additional text>

Something like:
   xymon xyserver.foo.com "status+6 client.foo.com.alive green"

This will put a column "alive" for host "client" which will go purple 
six minutes after the last green message is received.


-- 
    Do things because you should, not just because you can.

John Thurston    XXX-XXX-XXXX
user-ce4d79d99bab@xymon.invalid
Enterprise Technology Services
Department of Administration
State of Alaska

list Ryan Novosielski · Fri, 11 Mar 2016 13:05:39 -0500 ·

I have a second xymonnet instance for just this reason, on a machine that does have ICMP access to that equipment. May not work for you, but it's a thought. 
--
____ *Note: UMDNJ is now Rutgers-Biomedical and Health Sciences*
|| \\UTGERS      |---------------------*O*---------------------
||_// Biomedical | Ryan Novosielski - Senior Technologist
|| \\ and Health | user-46c89e614701@xymon.invalid - 973/972.0922 (2x0922)
||  \\  Sciences | OIRT/High Perf & Res Comp - MSB C630, Newark

▸ quoted from John Tullis

`'

On Mar 11, 2016, at 11:41, John Tullis <user-a6bbfd057f07@xymon.invalid> wrote:

Is it possible to shorten the purple for the entire server? We have servers that don’t have ICMP access so when the server goes offline, we rely on the purple alert to let us know that the server is down. With the 30 minute delay, I’m not getting those alerts until much after the server needs to be back up. Is it possible to change the delay so I would know sooner if the server is down?
 John Tullis

list Japheth Cleaver · Thu, 17 Mar 2016 11:59:10 -0700 ·

On Fri, March 11, 2016 9:54 am, John Thurston wrote:
*snip*

▸ quoted from John Thurston

Alternatively, define an ext script on your inaccessible host to send a
status message in every X minutes with an appropriate TTL

status[+LIFETIME][/group:GROUP] HOSTNAME.TESTNAME COLOR <additional
text>

Something like:
   xymon xyserver.foo.com "status+6 client.foo.com.alive green"

This will put a column "alive" for host "client" which will go purple
six minutes after the last green message is received.

This is probably the most "xymony" way of solving the issue :)

That being said, it wouldn't be too hard to add a default status lifetime
as another per-host option in hosts.cfg, overriding the global default,
but not a per-message TTL.

This would apply to *all* incoming status messages for that host, though,
including those which come from varying sources and (potentially)
intervals. xymond_client and xymonnet intervals should be considered along
with any other source of statuses to make sure they're all recurring under
the TTL, or that the failure modes encountered count as features and not
bugs...


Regards,
-jc

Purple Interval 🔗 link

Purple Interval