[Xymon] Several yellow alerts on nongreen.html page, acknowledge alerts lists none

[Bill Arlofski]

On 09/17/14 11:01, Root, Paul T wrote:
> 
> 
>> -----Original Message-----
>> From: Xymon [mailto:xymon-bounces at xymon.com] On Behalf Of Bill Arlofski
>> Sent: Wednesday, September 17, 2014 9:46 AM
>> To: xymon at xymon.com
>> Subject: [Xymon] Several yellow alerts on nongreen.html page, acknowledge alerts lists none
> 
>> I have seen a curious behavior that I think might be a bug.
> 
>> Currently I have several servers with a yellow status for the cpu test. Each
>> of these machines were recently rebooted, so the yellow status is correct, and
>> I know that yellow statuses will stay yellow for several more hours.
> 
> Hours? I believe the default is 1 hour, and to me that was way too long.
> I've set it to 30 minutes, and most of the time still think that's too long. 
> 
> You can adjust this in analysis.cfg with the UP option.

Yes, I am aware of that. :)

I set my default UP to 24 hours in analysis.cfg as a matter of preference.



>> I would like to acknowledge these alerts to stop the distracting blinking
> yellow icons and show the static yellow check mark icon instead.
> 
>> From the nongreen.html page, choosing Administration --> Acknowledge alert
>> brings me to the acknowledge.sh page which says "No active alerts" instead of
>> displaying the form with each non-green host/test listed and the ability to
>> acknowledge multiple hosts.
> 
> I've noticed this, I think sometimes when you enter a new page, you have no context. It maybe a bug, but I just think of it as an annoyance.

Yeah, it looks like the idea of "context" is a good idea as a feature, but in
practice it seems to act more against than with the user - at least for me.
But at least now that I see what is going on, I can jump to main view, and
then to Acknowledge page to see all alerts in the list to ack.

> 
>> If I click on a yellow cpu icon from the nongreen.html page, and then
>> Administration --> Acknowledge alert I have the form to acknowledge the yellow
>> cpu alert as expected.
> 
>> However, there are two problems with this.
> 
>> First, with a large number of non-green hosts/tests, I would have to go back
>> and choose each test manually and individually to acknowledge it
> 
> On the acknowledge page, you can click the check box next to individual alerts, and then at the bottom put in the global message and time for acknowledgment.

This was my initial problem.  I went to the Acknowledge alerts page to
acknowledge all of them at once, but when I got there, there were no alerts
listed - hence my original post this morning. :)

>> Second, I have no red alerts in the system, but when I om on the
>> acknowledge.sh page, the background is red
> 
> I think it's always red. I've never seen it otherwise. 

I think white or blue might make a better choice - or the color that
represents the current worst status in the system.


Thanks for the replies Ryan, Paul and Kris

If it is not a bug, then at least I can quickly get to a list of all non-green
tests in my Acknowledge page.


Bill




-- 
Bill Arlofski
Reverse Polarity, LLC
http://www.revpol.com/
-- Not responsible for anything below this line --