probable bug in BPI

This support forum board is for support questions relating to Nagios XI, our flagship commercial network monitoring solution.
Locked
smoren
Posts: 62
Joined: Tue Sep 29, 2015 7:24 am

probable bug in BPI

Post by smoren »

Hello,

I think I have found a bug in BPI component. Here's my configuration:
I have primary group with one group member(this is also a group). This group member is configured as essential member. When this group member changes state to warning, 'parent' group changed status to critical.
I'd expect this behaviour if group member is NOT an essential member(this is where health threshold apply). In my case (when group member IS configured as EM), I'd expect 'parent' group to have same state as essential group member(warning).

Both group are monitoring using check_bpi. You may find state history of both services in attachment.
Please note missing essential member in Information column (parent_group.png).
I'm using Nagios XI 5.2.8.

Thanks.
You do not have the required permissions to view the files attached to this post.
User avatar
lmiltchev
Bugs find me
Posts: 13589
Joined: Mon May 23, 2011 12:15 pm

Re: probable bug in BPI

Post by lmiltchev »

Well, this has always worked this way - it's even explained in the "help" menu:
example01.PNG
I believe the state of the "main" BPI group should match the state of the "sub-group" (child) if it is an essential member. That's why, some time ago, I posted an internal bug report (TASK ID 5643). The bug was partially fixed.

For example, when you add a host/service to a BPI group as an essential member, and this host/service goes to "WARNING", the "main" BPI group would go to "WARNING" as well.
However, when you add a BPI group to another BPI group as an essential member, this no longer works the same way. The "main" group's status will change to "CRITICAL".

I will be re-opening the bug report, and including a link to this post in it. I already discussed the issue with our developers.
You do not have the required permissions to view the files attached to this post.
Be sure to check out our Knowledgebase for helpful articles and solutions!
smoren
Posts: 62
Joined: Tue Sep 29, 2015 7:24 am

Re: probable bug in BPI

Post by smoren »

Thanks for explanation. This really clear things up. So to sum it up:
Current behavior is:
For example, when you add a host/service to a BPI group as an essential member, and this host/service goes to "WARNING", the "main" BPI group would go to "WARNING" as well.
However, when you add a BPI group to another BPI group as an essential member, this no longer works the same way. The "main" group's status will change to "CRITICAL".
Behavior after that bug will be completely fixed could be written as:
For example, when you add a host/service/group to a BPI group as an essential member, and this host/service/group goes to "WARNING", the "main" BPI group would go to "WARNING" as well.

Do I understand this correctly? :) If so, do you have any time range when this should be fixed(maybe in 5.3.0)?

By the way, do you know why there is empty string in information column for critical states(see image parent_group.png)? (e. g. CRITICAL - Essential member "" is in problem state....). Another bug?
User avatar
lmiltchev
Bugs find me
Posts: 13589
Joined: Mon May 23, 2011 12:15 pm

Re: probable bug in BPI

Post by lmiltchev »

Do I understand this correctly? :) If so, do you have any time range when this should be fixed(maybe in 5.3.0)?
Yes, this is correct. I don't have an ETA. It is possible that this is going to be fixed in 5.3.0 but I cannot be 100% sure. It depends on the length of our developers "TODO" list.
By the way, do you know why there is empty string in information column for critical states(see image parent_group.png)? (e. g. CRITICAL - Essential member "" is in problem state....). Another bug?
Yes, I noticed that prior to filing the bug report, and already have it described in TASK ID 5643, so both issues will be fixed.
Be sure to check out our Knowledgebase for helpful articles and solutions!
Locked