Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

StuartR — Mon, 30 Jun 2008 17:59:09 GMT

With MOM 2005, we can accomplish this quite easily using the following approach:

An SERVER DOWN alert can be generated in response to an internally-generated ping failure event created by a MOM Agent ping script which is part of a MOM Agent connectivity rule.

This rule monitors for the internal failure event and will generate a "Service Unavailable" alert indicating that the Agent Computer is most likely down (or has lost network connectivity).

Note that 12 ping attempts over a 90 second period along with an additional ping after 15 minutes must all have failed before this alert is generated.

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

kevinhol — Mon, 30 Jun 2008 20:01:32 GMT

One note to add - in OpsMgr you will get a distinct alert whenever an agent doest not respond to ping, in addition to the heartbeat failurre alert. What we dont have - is a state view JUST for computers that are down...

You could easily write a custom monitor that runs a ping script - and build your own state view for this in the console... and not need this report. The benefit of the report is being able to schedule it and deliver via email or sharepoint.

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

daviesg — Mon, 28 Jul 2008 16:20:13 GMT

Hi Kevin

Trouble is by creating the monitor you mention you are actually duplicating work that OpsMgr is doing. It sort of highlights the lack of logic in some functionality.

To me, it makes no sense that I have to do a ping script as a monitor when OpsMgr has a much more powerful solution - agent heartbeat with associated ping of servers on which the agent heartbeat has been missed. I just need to get that information into the console .... and the fact that OpsMgr can't is a something of design flaw.

As I mentioned on the newsgroups, I don't think the report is feasible for near real time info in a large environment.

Cheers

Graham

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

daviesg — Mon, 28 Jul 2008 16:23:29 GMT

Ahhh .. didn't read that properly before I posted!! Meant the fact that agent health state couldn't be incorporated into the computer state view is something of a flaw ... realise there are the agent health state views as per my posting in the newsgroup ;-)

A cool way to use a web page view in the console - run a report!

Kevin Holman's OpsMgr Blog — Sat, 08 Nov 2008 01:14:30 GMT

Here is a unique way to use web page views in the OpsMgr console. You can create a web page view in the

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

bradje — Mon, 22 Dec 2008 09:19:57 GMT

Hi,

I do not understand how to IMPORT the report in to the new Custom report folder.

I notice that the reports on my reporting server are *.rpdl but this attachment is *.rdl.

How do I get this report into the new folder?

Thx,

John Bradshaw

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

hnehnes — Tue, 30 Dec 2008 15:00:10 GMT

Hi,

I have a different problem to the same topic. If a server goes down I do not receive any alerts. When I open Health Explorer with the above settings, I see only white bullets under Availability except Local Health Service Availability. Computer not Reachable, ... are disabled in their sealed MP. What is wrong in our configuration and what do I have to change to get an alert when a server goes down?

Thanks

Hendrik

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

Ren — Fri, 23 Jan 2009 14:17:29 GMT

I tried to use this UDL file by following the steps as mentioned in this site. When i run the report getting this error "An error has occurred during report processing.

Cannot create a connection to data source 'ops'.

For more information about this error navigate to the report server on the local server machine, or enable remote errors ".

Please advise.

Ren

re: Whoops!

kevinhol — Fri, 23 Jan 2009 17:13:12 GMT

You are correct - It looks like in this RDL file I named my data source "Ops" instead of "OpsDB".

Simply open the RDL file - edit that, and import..... or simply go to your imported report - edit it - change the data source to your live data source that points to the opsDB.

re: Which servers are DOWN in my company, and which just have a heartbeat failure, RIGHT NOW?

mccreerJ — Sat, 14 Feb 2009 00:16:32 GMT

Do you know of any way to setup subscriptions for only "ping failed" notifications? Right now every time a server fails a heart beat and cannot be pinged we receive two text messages. One for the heart beat failure and one for the Ping failure.

re: Subscribe to server down alerts only?

kevinhol — Sat, 14 Feb 2009 00:36:24 GMT

YES! I do. :-)

In R2 - this is super easy - because we can subscribe to alerts rule by rule - monitor by monitor.

In SP1 - it is doable - just a bit more difficult. Please see my how to post at:

http://blogs.technet.com/kevinholman/archive/2008/10/12/creating-granular-alert-notifications-rule-by-rule-monitor-by-monitor.aspx