How can we replace Ca-Ehealth with CA-UIM without this dependency because we have Integrated CA-SOI?

[Fabiano Lizzardo]

We installed Nimsoft version 8.5.1 software to replace Ca-Ehealth more by migrating urls monitoring and monitoring by CA-SOI integration - I can not see separate urls from a server that is contained in various services but urls alarm all the services that this server is contained. When viewing each url as it was viewed on the Ca-Health connector it would be possible to perform this configuration. How can we resolve this issue?
will there be this distinction of urls in a next version?

Fabiano D.L         

[Daniel Blanco]

Can you give an example of what your talking about?

 

Are you saying previously your integration with CA SOI & eHealth your one eH connector allowed you to generate alerts from one eH poller that was monitoring many different URL's and when those alerts came it, you were able to place each one in their own Service View to reflect that specific services URL health. 

With UIM, its different depending on how you setup your URL monitoring. If you have just one robot, with the URL_Response probe running, then all alerts from that one robot will seem to be coming from that one box and all instances of that robot in the different service views will be reflected thru all services views it present in. 

I don't use the GA UIM connector but when I looked at my lab instance I do see that my URL monitors each have their own URL CI in SOI so that way you can just place the profile in the service you want to show that specific URL's health in. 

[Shaheen Ahmed]

Hi Fabiano,

As Dan has said it would be great if you provide few example, please upload few screen shots as well.

Thanks

Shaheen 

[Fabiano Lizzardo]

I have a monitoring on a server and this server has the monitoring of several urls and belong to different services. In Ehealth I can visualize url tests on the connector and associate with different services, since UIM can not see url tests on the connector and therefore I can not associate with different services.
In the UIM connector the url test elements are not displayed.

 

 

 

 

When validating on the CAUIM connector, I can not see the tests performed by server n are not displayed separately. The urls tests do not belong to the same service, I need them to be separated according to the Eheatlh connector.

FROM CAUIM >

 

FROM CA_SOI

[Shaheen Ahmed]

Hi Fabiano,

URL_response probe does a nslookup on the URL you are monitoring and whatever result is returned gets added to UIM DB, for example, if you create url_response profile to monitor http://www.msn.com you are going to get a-0003.a-msedge.net not www.msn.com. The result of the nslookup is what gets added to UIM DB and UIM connector publishes it. Niscache folder in nimsoft contains files with .dev extension containing the details of the device nimsoft adds in to the DB. So the first thing you want to do is do a nslookup on the URL you are monitoring or look in to the niscache file to see what exactly is the result of the nslookup.

Thanks

Shaheen

[Fabiano Lizzardo]

What I want to do is associate url tests that are created on a single server with separate CA-SOI services. I can not see the separate tests on the connector as it was on ehealth. it is possible to view the url tests in the CA-SOI + UIM connector folders - As is done in the first figure of my first response.
To be honest, you can not understand the relationship of your last answer. I checked the ".dev" files in the folder:
 
__type 7 2 D __id 7 34 DE62FC55E940BBFF5503594ECE2D40008 remote_dev_ip 7 12 10.96.161.8 remote_dev_hostname 7 29 SPOBRCAEEM02.serasa.intranet remote_dev_key 7 29 SPOBRCAEEM02.serasa.intranet
I'll go another example. I have the following configuration:

   

one Alarm was displayed:

Device was created :

 

When searching for the device or even the url to associate with a CA-SOI service I can not find the created Url or device elements.

[Shaheen Ahmed]

Hi Fabiano,

Not finding www.terra.com.br in SOI is a different problem. Open up a case with CA support. I have also asked Alex Yasuda to talk to you about CI name difference between UIM and SOI.

Thanks

Shaheen

[Fabiano Lizzardo]

01109773

TITLE :   Add different URL on a single server CAUIM+SOI connector. Associating the tests across many services.

[Daniel Blanco]

Shaheen_Ahmed

I am interested in this now to as I tried this out and I am seeing that the url_response probe is messed up when it creates the entries for these URL profiles in the UIM DB. 

I setup 3 profiles as a test:

 

In my raw mode I have that useHostNameForDeviceIdentification = no. Not sure what that does but it was ON in the above setup and we have same issue.

In the Advanced tab I have the Source Used for QoS was initially empty and I got the same results even after I changed it to $profile:

The url_response probe does save the metrics it collects as the profile name:

But what UIM's UMP & what SOI see's is these profiles as they are created in the UIM DB by the url_response probe:

They see some some weird name:

UIM UMP:

SOI:

 

SQL DB:

 

So please let me know what the support case says after they look at this. SOI is just picking up the entries in the CM_COMPUTER_SYSTEM table and they are not created correctly up by the url_response probe. 

[Daniel Blanco]

Okay I think this fixes it. In RAW mode on the url_response probe set the "useProfileNameForDeviceIdentification" = "yes" and then restart probe. You then should delete the existing weird entries in the UIM UMP's Inventory and check when the new entries get created. Those will now be the ones in SOI. For me they got created under "GenericIPDevice" class:

 

Not sure why one of them got created as a "Computer System"

I also added to each profile the $profile  variable to also the Source for Alarms and QoS:

 

i can then create service view with each url_profile accordingly:

[Fabiano Lizzardo]

Hi Daniel Blanco;
I configured it according to your prints - probe url_response, but I can not see one of the "MSN" urls. Until I re-read the "MSN" test, the still does not appear on the connector.

 

[Daniel Blanco]

Did you make this change?
In RAW mode on the url_response probe set the "useProfileNameForDeviceIdentification" = "yes" 

Change the other variable "useHostNameNameForDeviceIdentification" = "no" 

 

Also look at your SOI alert queue and find the URL alert that is currently on the UIM console.Check in SOI to see what CI it is associated with. It might be associated to the robot that's running the url_response probe. 

[Daniel Blanco]

Fabaino did you ever get this working?

What you can try doing is de-activate your url_response profiles.

Go into UMP > SLM (portlet) and then look for the data that this url_response probe created. Delete it.

Then go into USM > Inventroy and find the profiles this probe created and delete.

Stop / Start the data_engine probe.

Start up the url_response probe with the right raw mode cfg settings in place and then activate the profiles.

Wait till you see QoS data in the UMP and then check the SOI CI DB for the profile creations.

[Daniel Blanco]

Just saw this thread here that goes over this:

https://communities.ca.com/message/242122741-re-some-tips-needed-while-creating-usm-groups-from-urlresponse-qos?commentI…