branfarm ✭✭✭✭✭

I see it now.... how does the filter work exactly? Do I use regular expressions to match the process? Here's what I have: 3 java instances with different values in the java options. So all three processes are java, but I need to be able to select based on a "-Dframework.name=" value.

Roger, <sigh>... I was hoping you wouldn't say that. Can I rant for a minute? First, the underlying issue -- I can't monitor individual processes on linux out of the box. I can't imagine that I'm the only one that sees this as a problem. APM works great for Windows processes, and I'm sure it works great for wanting to…

 Could you please let me know what you come up with? I'm very interested to know how to get this working again. Regards, --Brandon

In case anyone runs into this in their own environment -- F5 has acknowledged that this is a bug in 11.6. You can open a case with support and they will provide an engineering hot fix for you. I imagine this will be fixed in the next release.

 Did anyone ever find a solution for the spikes on the charts? My data looks very similar. Could it be related to the flow timeout settings on the netflow source? I"m running the latest NTA v3 SP1.

LGarvin wrote: ... it's likely an academic question at best ... Let's not get academics involved here I don't think this has to be that complicated -- if it logically appears as one node, then that's what it is. When you "stack" switches they become one logical node. If you were to unstack those switches, then you would…

Really? That's exactly what I'm looking for -- I want to be able to have my scheduled discovery run, but not import any interfaces or volumes, even for new nodes.

Is it necessary to enter config mode for this command? It is normally executed from enable mode.

I've been having a similar issue, and I have a support ticket opened for this. In my case, I'm seeing it mostly with Cisco CSS's and tech support indicated that the startup config contained extra spaces that the running config didn't have. I think they are still working through a way around it. This might be the same thing…

Which poller are you using? And are you seeing results for every element? Here's what I get when I assign the temperature UnDP to a Universal Device Poller chart:

 This would be a nice feature to add in future versions.

I'm using NPM 9.1, so I'm not 100% sure if all of the features are backwards compatible. I've also never used a Cisco 7204, so I don't know how compatible these commands are. First, I set the device to use my NPM server as both a syslog server and a trap receiver. On my Catalyst 4507 I use: logging trap notifications (sets…

Ahh! Thanks for the help -- I'm glad it was something easy.

Happy to help -- just let me know.

They are free clocks provided by timeanddate.com. The box they’re in is a custom HTML resource, and I added basic HTML table code to spread them out evenly. Here’s the link for generating your own clocks: www.timeanddate.com/.../free.html

 I ran the following query against the Node details table for the new node and it only returned one row. I'm not much of a SQL guru, so I might be missing something: select * from NetFlowDetail_7 where TotalBytes < 0; The one column that returned had a TotalBytes of -2.117455E+09. I just submitted a support request for…

I'm familiar with it, but I'll admit I haven't had great luck with it over the years. Either the content was too generic for my issue, or was outdated, or the suggested procedures just didn't work. I don't have any recent examples though, and if the KB has improved lately, then maybe I should be giving it another look...

 HI there, I am currently monitoring the physical disk status of my servers with OID 1.3.6.1.4.1.232.3.2.5.1.1.6 (cpqDaPhyDrvStatus). I have the poller setup to use the raw values, mapped to return a result of either Unknown, Ok, failed, or predictive failure. I have it setup as a UnDP on all of my servers, and I wrote a…

That didn't turn out very well. Here it is again: SQLServer:Buffer Manager | Buffer cache hit ratio | 2300 | 537003264 SQLServer:Buffer Manager | Buffer cache hit ratio base | 2300 | 1073939712 SQLServer:Buffer Manager | Page lookups/sec | 676493749 | 272696576 SQLServer:Buffer Manager | Free list stalls/sec | 0 | 272696576

BTW, that trick works from the Node view page as well, without having to click on the Chart title first.

The report is based off the HP proliant UnDP's found here: http://thwack.solarwinds.com/docs/DOC-89891 I've uploaded the report and you can find it here: http://thwack.solarwinds.com/docs/DOC-167482 Hope that helps!

I actually heard back from support regarding my case, and here's what they said: Here is developments summary of the problem found during comparison of net-snmp snmpwalk and SolarWinds SNMP library behavior: net-snmp implementation of SNMPv3 handshake: 1. send "get-request" with msgFlags set to Reportable,…

My apologies -- I guess I misinterpreted how the view limitation feature would work. Like you said, it seems to rely first on the detected device type. I tried to find a way around it, but it seems like this is not a supported behavior. This thread here: http://thwack.solarwinds.com/message/154800#154800 seems to verify…

I don't believe environmental monitoring is available out of the box on any NPM supported Cisco device. You still have to configure UnDP's for them.

Hmm...sounds like the Riverbed is working fine. Have you examined the session trace for the connection attempt through NCM? Just in case you haven't done that before: 1. Go to NCM Settings > Advanced, and Enable Session Tracing. 2. Test the login from NCM 3. Hop on your NCM server and open up the session trace for that…

I don't think NCM knows Perl (someone please correct me if I'm wrong), so I think you'll have to recreate your script using either command scripts or command templates. Do you intend these to run automatically, or do they require any user input at runtime?

Can you tell me if this was remedied in the last APM release?

Thanks, it's good to know others have tried the same thing with no luck. Similar to you, I'm not seeing large numbers of open connections (around 1500 for me). 

I'm running into a very similar issue with a jailed environment, except I'm getting the error 'Syntax Error in BIND configuration'. It works fine if I use the root user. What I've found is that when you attempt to test the BIND credentials from the setup wizard, IPAM will login to your box as the specified user, and runs…

You should see it within the Edit Resource window for the multiple object chart.