tonybarrett

Thanks. What I've done so far is, after consultation with SW, upgraded the local SQL intance to SQL 2008 R2 Express which has increased the default DB size to 10GB, so I'm running again now. The event log on the NPM server was filled with SQL Application errors complaining about DB size being exceeded. NPM maintenance…

Wow. I can honestly say this is almost exactly what I'm looking for. I selected an unused switch port, and set it to unplugged, then connected a device. The switch port status went to 'Up', and stats came in. I then disconnected the port, and the status went back to 'unplugged'. No down warnings or switch flashing. I just…

I'll give this a try and report back. Thanks. This server with the flagged 'critical' database is NOT part of the DAG. It's in the same site, but a standalone server.

Yep, that's fixed it. Everything looking good again now. I'll make sure that update is skipped for the time being.

thanks alabama. We are indeed running SNMPv3 on the 2530 switches, which is an interesting thought. Could SNMPv3 be the root cause of this? I'd already run the UDT compatibility checker against these switches, and I get a clean bill of health on them - everything shows 'Success' for SNMP Discovery and Layer2/3 discovery…

I can validate this. AP's with no connected clients are not detected properly by SW. All I see when listing resources is 'Wireless Access Point' rather than 'AeroHive HiveOS Access Point by Solarwinds'.

Awesome tip. Thanks a lot. I'll look into this. I've currently changed the retention period for all data held for 365 days to 180 days, and I'm re-running the database maintenance tool, which should give me some time to sort things out. Our Trap Retention time is currently 30 days, and syslog 7 days, so I wouldn't really…

Thanks for confirming this. In our case though, we're still using v2c to manage the AP's so the SNMPv3 problem doesn't stick. It is interesting though that the new NPM beta announcement makes explicit reference to natively supporting HP ProCurve AP's, so I'm wondering if this will resolve the problem.

Anyone?

Thanks for the reply. Those links are useful, but don't actually tell you how to bump hardware health info up to the home page, so if there's a problem, you can actually see it clearly. What I really what to see is when a hardware component enters a warning or error state, for it to appear under 'Nodes with Problems'.

That certainly might explain it. I'd done a Windows update at the same time, so would likely have installed this patch. I'll roll it back and see if it fixes things.

Thanks. That's heading in the right direction. The only problem is if one of the edge ports re-classified as 'unplugged' becomes 'live' again, then I assume I'd still manually need to go into the settings and change the port from 'unplugged' to normal, so it then detects the port is live again. Ideally, you'd want an…

Ok, thanks. I suspected as such, but I thought I'd ask. This might have to go in as a feature request then, as I think it would be a pretty useful feature to have. In companies where edge network connections can change frequently, this could be quite useful.

Maybe to elaborate here. I might have found something, but need it confirmed. If I drill down into <mailbox server> / AppInsight for Exchange / Microsoft Exchange / Database Copies, when I click on one of the DB's and go into the details of this, under Database Details / Component Name (where it shows the database name),…

Thanks for the suggestion. I've had a look at these settings on the DAG mailbox servers, which are as follows; Winrs AllowRemoteShellAccess = true IdleTimeout = 180000 MaxConcurrentUsers = 25 MaxShellRunTime = 2147483647 MaxProcessesPerShell = 15 MaxMemoryPerShellMB = 512 MaxShellsPerUser = 25 I've not changed anything, as…

If you can get native Aerohive support into NPM, that would be great. As has already been said, the 4th gen wifi model Aerohive is very advanced, and leaves the 'traditional' method of wireless management in the dark ages.

Thanks for the info. Sounds like it's not just me then. I would say though, that I was expecting topology information to be discovered at layer 2 as well, as this is where LLDP/CDP operate. I know UDT 1.0 was primarily using ARP information to gather this, and not the proper layer 2 discovery protocols, and it looks like…

Upgrading to NPM 11.0.1 then allowed me to install IPAM 4.2. Still don't know why the errors were displayed though. At least it's now working. Full version string here is; Orion Platform 2014.2.1, SAM 6.1.1, QoE 1.0, IPAM 4.2, NPM 11.0.1, NTA 4.0.3, Toolset 11.0.0, UDT 3.0.2, IVIM 1.10.0

Thanks for the info. I've now managed to get the vCenter server to be detected correctly, and I can now see the cluster information. The Top 10 Hosts by Network Utilization still all show N/A though, so not sure what's going on there. It also appears if the Windows server running vCenter is detected as a vCenter, you…

I noticed a similar problem recently, and asked SW why this was happening. I was told to remotely view the realtime bandwidth guages, you need the Engineers Toolset installed, as the guages are pulled from there. Our Orion server has the ET installed, but my remote client doesn't, which means they work on the management…

Thanks. It's a resource mailbox database with 2 test mailboxes. Only used occasionally, and not one we have in production use, but it's still mounted on an Exchange 2010 server in the site. The 'components with problems' are; Database copies Database File and Transaction Logs Dir Info AppInsight has flagged these as…

Bump. Can anyone from SW comment on this? Will this 'fix' work for anyone experiencing this problem with AP status flapping?

Same problem here with Aerohive AP's. I'm going to log this as a support call.

I certainly don't like the idea of paying for this sensor functionality IF it's been integrated into NPM. This is where it should be of course, in the core product. If it was a separate plugin, then that's different, and we're used to that pricing model. Be very careful here SW. Don't take your customers for a ride. NPM is…

Think we have the same problem here. Layer 2 network with HP Procurve 2530G switches. We have an untagged VLAN (1) that all client traffic goes over, and a tagged VLAN (200) for all management traffic. UDT is polling the switches on the management network. There's no route between the management and client VLAN's, but…

A new HP PoE switch that's identified as 'HP Switch', but not the correct model; HP J9774A 2530-8G-Po​E+ Switch - 1.3.6.1.4.1.11.2.3.7.11.138 Strangely the larger 24 port version (J9773A - 1.3.6.1.4.1.11.2.3.7.11.137) is identified correctly.

I think the versions being reported in the error are wrong. I'm going to download and install the 11.0.1 NPM installer, and upgrade to this first. This is something I hadn't yet done, despite the error versions appeating innacurate.

Recently upgraded to v4.1.258902. Rest of cluster still on 4.0, but moving to vSphere 4.1 soon.

Native Aerohive support please. Been asking for this for ages (so have many others). The current wireless stats in NPM don't show anything for our Aerohive AP's.

Have you tried this yet? We have this problem as well with HP AP's. They are continually reported as going up/down all the time, but clients are connected and working ok. In our case, NPM is reporting >200ms response time, with some of them showing 100% packet loss. All other devices on our network (also HP ProCurve) have…