MNPS_G

+10!

I have been feeling like somewhere along the way in Ver 9.5 that my polling engine has been behind, but haven't really seen anybody else mentioning it until this thread. I am up to SP4, and somewhere along the way nodes were constantly going up/down in Orion, but if I polled them myself they were fine. Stopping Netflow…

I found the MIB in from the walk that appears to show the number of access points that the controller reports it has connected to it. The controller in question has 60 APs on it, reported both via the web interface and the MIB. However, Orion is reporting 31 APs for the controller in question. Here's the screens: The MIB…

This is definitely something we would like to see as well. We are using some custom pollers get serial numbers off of SFPs in our switches, and we don't need those values polled every 15 minutes along with interface statistics. Thanks, Geoffrey

Marie, Sorry, forgot to put that in my post. I'm running 9.5.1 and will be upgrading to V10 once we have a new SQL server in. Brandon, I'm comparing MIB walks with the controller's numbers and Orion's numbers and will post some info in a few hours. Thanks, Geoffrey

This would definitely be something we would like to see as well.

Peter, Sorry for any confusion but above I said, as far as I understand, that you can only have NTA running on your primary polling engine. If someone knows otherwise hopefully they'll chime in. We've got an APE right now and only run NTA on the primary polling engine. Geoffrey

Okay, so we also use the 7758, and 4500 series switches. We don't use the 4200G or the 5500G here, so I can't say if these will work on them. The CPU is broken into a 1 minute avg, a 5 minute avg, and a 5 second avg - we poll all 3. For memory, you get Free Memory, # of memory allocation failues, no memory available…

Being as I'm not a SQL guy either, some of the details escape me as well, but in short RAID 10 offers performance gains (especially in write operations, apparently) over RAID 5. I'm guessing it's faster since it doesn't have to write the parity bit like RAID 5 requires, but like I said I'm no expert, I'm just learning…

I don't believe there is going to be a way to plug-in graphs from the engineer's toolset, but if someone else knows a way we may do that ourselves. What you could do is setup another server with something like PRTG and use the external linking capability to have it as a resource on your menu bar. However, even that…

I'm going to be in the minority on some or all of these, but figure that they're at least worth putting out there: 1. Update some of the Cisco-only tools to work with other major switching vendors (I'd specifically like HP Procurve and HP's legacy 3Com equipment). That's been happening more and more in other products like…

Hi, The interfaces will still exist even though they no longer show up as a resource on the new node which is expected. Through the admin console go to the manage nodes and change from nodes to interfaces, if these interfaces have a name in common refine your search to those. You will see all the old interfaces as…

This is a big complaint of ours as well. Getting contacted by 10 different people through the year for different products doesn't help us or Solarwinds at all. I recently replied to their e-mail trying to get us to try/purchase UDT and pointed out this very problem to them. I got a reply from someone stating that due to…

We are also running into the same problem. It appears Orion is showing roughly 60% of the APs connected to an individual Aruba controller. Mark, did you ever have any luck sorting this out?

Based on my experience, I believe the following would answer those issues: 1. You should only need the reports on any server running the Orion website. The additional pollers don't need the reports on them for the reports to run properly. 2. If they are advanced alerts you won't need to do anything else for alerts to work…

What model 3Com switches? We're monitoring the CPU and Memory in a couple, I could tell you what you need to monitor to get them if you're using the same model as us.

Hi Peter, Unless something has changed recently, NTA can only reside on your primary polling engine. Other than that, it sounds like you are good to go. Your web server will know about the nodes on your APE because your APE will tie into your existing database. The webserver connects to your DB to get the information for…

RAID 10 can also be stated as RAID 1+0. If you look that up, that may help. Your DBAs should know what that is, if you put it that way. I'm curious as to what your final solution is, we're in a similar boat over here with polling and NTA, it's just gotten to the point that I disabled NTA so I could at least be sure the…

Brandon, Thanks for the word on where all of that is headed. I think that will make management of nodes much easier for many of us, which would be great. Geoffrey

I just manually added the node again to try and see what would happen. When it came to the interfaces, it already had the new ones checked, and our custom fields were filled out appropriately. I think that should be equivalent to doing the discovery you mentioned. However, the old interfaces are the only ones listed. Orion…

Mark, Thanks for the help. I am quite familiar with opening technical tickets, but hadn't had to deal with an issue like this previously. Thanks again.

Senthil, Do you know why the current/new Traceroute doesn't have the SNMP option? It seems like a step backwards to me, so I'm curious what the thinking/reasoning is on that. Thanks for leeting me know where the deprecated version is. Thanks, Geoffrey

Byron, Thanks for the help in pointing it out. I had not used the "Current Status of Nodes, Interfaces, Etc" report category before, so I didn't realize the UnDP custom pollers were in there. That worked perfectly. Sean, thanks for the help as well, I just gave Byron's solution a try first and it worked for me. Thanks…

I believe Andy is right on that one. However, it shouldn't be too hard to create a report you could add to your main page (or elsewhere) that lists your APC units that have high current output. Don't know if that would still get you what you want, but based on the general idea you threw out there I figured it would be…

As with others, I wanted to say you aren't alone. I've never opened up a ticket becuase the fix was always pretty simple, if a fix was even needed.

I believe that it uses SNMP to query the uptime of the device. If the uptime has been reset, as though the device rebooted, then it classifies it as a reboot. That's the best theory we've come up with here, but if that isn't the case I'd like to hear how else it determines if a device rebooted.

I don't have NCM, but I can say that the old SuperStack 3 switches have a very different OS on them than the 4210s. The 4210s are newer and run some version of Comware, whereas the Superstack switches are simple, menu-driven stuff. Again, I don't want to say NCM can't do it since I don't have it, but being part of a 3Com…

If it is one of the recent models referenced in the thread Mav posted, then yes, sFlow works. The configuration guide shows you how, but it is fairly straight-forward. You just need to be sure that you are monitoring the outward facing IP address of the switch (the first IP address that Orion would hit on the switch,…

Depending on how difficult it is to relocate the DB in that version, I would think you would want to do that first. That way, as you upgrade and the DB changes in the upgrades happen, you'll get the performance benefit of having your DB on the separate server from your polling engine. The only other reason I can think of…

Is your polling server showing any serious load on the CPU/Memory (although I would expect CPU if either)? Your hardware for your SQL server is much better than the one we have here and we monitor many more nodes (~7500) and have NTA, but our load times sound a bit better in general. For your NTA, that might be about…