mleinio

I know that there is nothing in the Release Notes about CLR being mandatory, but the upgrade to 9.5 is not mandatory. so why is everyone complaining? Well, NCM 5.5 seems to have corrected some major problems regarding download config jobs (so we will be running NCM 5.5 instead of the previous 5.1.1), but the NPM…

A word of warning for all users having HP ProCurve switches in NCM. If using a Telnet/SSH Prompt Timeout value of 5, the HP ProCurve 2512 and 2524 switches will reboot when downloading the configuration with NCM and using SSH. We've verified this with 120+ individual 2512 and 2524 switches, running various versions of…

I had noticed that it was fixed in 9.5, but after the installation of SP2, it once again began to take forever to shutdown. Interesting to hear that, we jumped from 9.1SP4 directly to 9.5SP2 (skipper 9.5 and 9.5SP1), and the shutdown has been slow all the time. So we didn't have that experience. I hope that observation…

Also, do you know the syntax to pass a particular Putty profile when clicking either the telnet or ssh link? Check the Putty documentation about the command line parameters: http://the.earth.li/~sgtatham/putty/0.60/htmldoc/Chapter3.html#using-cmdline Markku

Email would be fine (NOC here as well), plain text format is enough for us, I guess we can live with HTML as well... Markku

Hi, installed 9.5.1RC2 today, only positive experiences so far! Events page is loading very well now. Thanks SW for the opportunity for using the RC version for fixing this issue! Markku

Database server is separately installed, with no indications that there is any overload going on. Two pollers, 7000+7000 elements. What is most notable that the Events page takes always the same time regardless of the filter (time period, number of events), and the Last 250 events report takes just a couple of seconds.…

We made some changes with 9.5 and the SP's, so hopefully this should be alot better Hi, no it isn't, the netperfmonservice.exe is still taking ages to shutdown when asked. Markku

Thanks Chris for your understanding :-) For now, the setting I was talking about is called "Orion Node Import" in node settings (in NCM). Markku

Thanks Chris. Markku

We are running a 4000+ nodes installation of NPM and having serious issues on importing the nodes automatically to NCM (all nodes import without any manual work involving the custom property): the import job just hangs at some point. I have created a ticket on this, but for some reason I haven't yet had SolarWinds to…

Hi Dave, the support desk gave us an RC version to try, and we are going to install that on Thursday, so let's see what happens. Markku

You who complain the Syslog page is slow, how are your Events pages loading? I mean, for us, since updating from 9.1SP4 to 9.5SP2 our Events page has been taking about 40 seconds to load. Support asked us to update to SP3, we did, it didn't help. Then SP4 came out, again the same request, now we haven't been able to update…

So, just curious, did you ever get your problem fixed? Sorry, I forgot to mention that, yes, we managed to install SP3 after that. All previous installs (NPM 7.x, 8.x, 9.x, service packs) have been done successfully with a domain account (with local admin privileges of course), so either SolarWinds changed something…

Chris, just to make it sure: Have you noticed that also devices usernames and passwords are "defaulted" as well in the import process (just like the transfer settings)? Markku

Just my two cents (without current access to NCM installation): By default NCM web site includes the Thwack.com applet. If you don't have internet connectivity, that applet will cause slowness in NCM web site. You can edit the page layout and delete the Thwack.com component and it will load faster. Markku

We are using the import job.

jp, I also was puzzled by the skipped nodes. The thing is that in NCM all nodes have a property which-name-I-dont-remember-and-dont-have-NCM-by-me-right-now, and that property is normally False. So all nodes that have been newly imported in NCM will have that as False. Now, if you change - say - some nodes SNMP community…

Addition to this: If I add an existing node (to be able to edit the discovered SNMP System Name as a node name), the edited node name field is not saved. NPM actually overwrites the current node name with the device SNMP System Name, regardless of the original node name. Markku

That's correct, the web console of NPM has a bug that makes the "List Resources" function to show incorrectly selected interfaces. This is a common problem for us as well, thanks for starting the discussion :-) Let me give an example: I first have a Cisco 837 router with Ethernet0 interface being monitored in NPM. Then the…

Is there any practical reason to update 9.5.1RC2 to 9.5.1? We are quite pleased with the RC2 (thanks for that!) at the moment. Btw, there is no announce in the announcements forum yet. Markku

I'm not sure I understand the comment "Some routers have several addresses that need to be monitored. By browsing this forum, I have gotten the impression that it is not possible to configure several addresses for nodes" You would normally monitor a device via the loopback address, and then select as many of the interfaces…

We also got that error. The support people told to not to use a domain account but a local user account to do the update. Also to remind those that kill the netperfmonservice.exe manually: please change the Recovery settings of the service before doing that, change the "Restart the service" option to "Take no action".…

Thanks Chris for confirmation on this issue. I wonder why the support case was closed without any communication with me on the issue. Anyway, hopefully a fix will be available (public or private) as soon as possible. Also, as a feature request for this import job, I'd really like to see possibility to use SNMP community as…

I would also see other possibilities for import filtering (Vendor and Machine Types can not distinguish proper nodes, the true/false custom property needs extra work which is costly). Possibility to use SNMP community for selecting the nodes would be great for us. Markku

Let's try again... Hi, I'm missing the HP ProCurve switches and some others (checked with NPM 9.5 SP3):1.3.6.1.4.1.11.2.3.7.11.18 = HP ProCurve Switch 2512 J4812A1.3.6.1.4.1.11.2.3.7.11.19 = HP ProCurve Switch 2524 J4813A1.3.6.1.4.1.11.2.3.7.11.20 = HP ProCurve Switch 5304XL J4850A1.3.6.1.4.1.11.2.3.7.11.23 = HP ProCurve…

Thanks! The support desk promised to keep me updated when there is a fix. I really hope a solution, this has been a major problem every since updating from 9.1 to 9.5 (9.1 was fine with the same amount of devices). Markku

Actually I'm not sure if the NCM syslog server got the correct device IP address in that approach. I'm not near our servers so I cannot test this at the moment. Markku

I haven't yet installed 9.5, but in 9.1SP4 managing the nodes in web is incredibly slow. It is totally unusable. We have 4000 nodes with 14000 elements. SolarWinds, I know you are keen on including new features (we like that as well!), but getting the management performance better has to get more priority on your…

Same problem here (9.5 SP2, just updated), opened case 108422. Markku