bobross

This is exactly what I was looking for. Thanks a ton!

We've seen similar issues with HPUX and Solaris boxes. In addition to alterego's suggestion above I'd recommend verifying that the server is using a current version of net-snmp.

Thanks for the response alterego. The way templates are currently implemented works fine in many cases; however, there are some instances where stand alone templates can become bulky with poor organization. This issue first came about for us when we had to write script monitors to check website data for a third party…

It shouldn't be too difficult, but you'll need to code in the label names statically. From my testing SAM doesn't play nice with dynamic labels: For example: # Use this...print "Message.Label1: $message1";print "Statistic.Label1: $statistic1";# Not this...print "Message.$label1: $message1";print "Statistic.$label1:…

Go into 'Edit Node' and test SNMP. Usually when we see this behavior its due to problems with SNMP. If the test fails - Restart the service, verify the community string, and make sure SNMP traffic isn't being blocked.

If the Startup Type is available over WMI you might be able to get something working. You'd likely need to create an additional component/template that pulls that information. I don't have much experience with how the alert trigger works though. If you are able to get the information into the database then I think that you…

It turns out that both SNMP and free are pulling data directly from /proc/meminfo which does not contain actual utilization levels. free calculates used space by subtracting free memory from total memory. That is explanation enough for me to give the admin. I'd still like to know why it was decided to use the above formula…

Someone posted on thwack recently that they have software to integrate Solarwinds with Service Now. I can't remember the specifics but a quick search should find the post.

We've dealt with similar situations before. Sean's options are really it. It would be nice if NPM could poll up/down of nodes using SNMP though.

As adeimel suggested, you might consider looking into SEUM. We are just starting to deploy it for some of our mission critical user applications, but it allows for much more detail than what you can get from a basic HTTP(S) monitor. With that said, we have a few nodes that have multiple URL monitors attached. In our…

You should be able to delay the alert action for another polling cycle. This way it will only actually generate an email or whatever if it fails twice in a row. We've had to do this for a few different monitors.

The VBScript that I'm looking at would be run locally on the Solarwinds server at a client site. It would open a telnet connection over port 25 to our externally facing email server and possibly execute a HELO command. I've tested manually on a server that we can receive email from as well as from a server that we can't…

Sorry I missed this, alterego. At this point we've been waiting for answers for about two months without much progress... Although I've sent in at least 10 diagnostics. In addition to this popping up on our main server, we're also seeing the same thing happening on our players. I'm going to look into trying the beta, but…

I see, that definitely clears things up and I'll relay this information to our team. Thanks

Any reason why you aren't relying on NPM to alert on volumes as shown in Andy's post above? I'm aware of NPM showing odd values for *nix drive space, but we haven't encountered any problems with Windows drives on APM.

Middle-click works great for me. I usually have EOC open in one tab and another tab with a custom view on our biggest NPM. When something goes down at another site I just middle-click (open in new tab) on the EOC page. That way I never have to leave the EOC page.

You can change the name of the component to reflect the URL and Search String. This information is also a click away on the Edit Application Monitor screen. The Message field is based on the response of the template. Some templates have messages regardless of state, the HTTP(S) monitors don't. There is also a Feature…

You can use the following to setup a resource that only includes x interfaces: Summary Reports Percent Utilization of All Interfaces --or-- Current Traffic on All Interfaces Then set up an SWQL filter similar to this... NodeID = 'N01' AND InterfaceName Like 'Tunnel1' The Filter help has a list of all of the properties that…

You can simulate this with a script. You'd need to pull the CPU information for all processes, sort the results, then return the process name/ID in the Message field. It won't be all that useful for historical data, but can be a useful tool for quickly checking the biggest offender(s) without needing to connect to VPN,…

This does make a bit of sense from what I recall seeing. I'll give this a try on Wednesday when I get back to the office, but this is definitely promising. 

PM Sent. I'd like to leave this open thread open in the hopes that someone will be able to provide an answer. Primarily because we'd like to try and get something working right now, but also because this question has been asked all over the internet without any real answer. And of course, I'd like an answer so I don't feel…

3 nodes need to be monitored. For each node there will be 16 different components. Script/code will be the same for all with different command line arguments based on node and regex filter. I considered using the script on its own as a template and adding it 16 times to each node, but then management becomes a pain when I…

It sounds like you are trying to dynamically generate a report based on the group page that you are viewing. From what I understand of the report writer you won't be able to do dynamic report generation like that. The best that you could do was create whatever report you want then use a WHERE clause to only show the…

You could probably even take it one step further and include a basic command that returns a known value (uname for instance) and check the exit code and/or response of the command if you want something a little less generic.

We had one of our DBAs come up with the queries for this template: Its worth taking a look at at least... If nothing else it might give you some ideas on how to monitor jobs that are specific to your environment. I really don't know anything about Cognos myself, but we have found that picking the brain of DBA's really…

We have ~50 different Solarwinds servers and none of them exhibit this behavior. I'd be inclined to think its more of a network issue myself. Might try taking a look there for the culprit.

Is this a Windows or *nix server?

Probably the easiest solution would be to have the sysadmin modify the script so that it can be called directly from APM without having to use the HTTP. If you did want to use the web based solution I'd advise using either SEUM, or having the script run as a cron job/scheduled task that writes to an html file somewhere. An…

Regarding the auditing of unmanage windows, I know that auditing in general has been an outstanding feature request for quite a while. I'd suggest tracking down that feature request and lending your support there as well.

Are there any wscript processes that are stuck running along with the iexplore processes? Could be stuck in one of the while loops. I never ran into this problem specifically with InternetExplorer.Application before. I have seen behavior where a script runs fine locally, but doesn't function properly when run from…