willhepptemp

Holy Cow!!! The manual maintenance you suggested has cut the DB size in half! Do you know what the difference is between the nightly maintenance and the 'manual' maintenance? Nightly takes 3+ hours and the manual takes tens of minutes. I did notice that the manual is very intensive on the server - it really pounded the CPU…

NVRAM is where the configuration is stored and flash is where the images are stored (either on-board or removable card.) I honestly don't know how much space ifindex persistence may take up but it's a necessity if you want to get good management data. Additionally, we're just talking about an ordered list of interface…

Disappoint? Just telling customers to hit the 'reset' button and waste their time with a reinstall so the SolarWinds support staff don't have to bother themselves with troubleshooting properly is just plain lazy. I've opened a case tonight - and no, I won't be accepting reinstall as a troubleshooting method. Will

Do as you please - on the NVRAM note, any ifindex persistence will have negligible impact on NVRAM so I don't understand where you're coming up with 6513s being some kind of problem. Will

"Reinstall Orion"? You have *got* to be joking. How about actually troubleshooting the problem rather than just crossing your fingers, cuasing the customer a *lot* of pain and effort just in the hopes that the issue goes away?? I've been hearing a lot of "reinstall [product-x]" from support and that's just not a reasonable…

> Your environment is way to large to get stats every second for each element and expect the DB to survive Heh heh - yes, I'm not even contemplating something like that. I'm really more interested in 1 minute intervals for a small subset of the infrastructure and 2 - 3 minutes for the rest if the DB will support it. If…

Is there any way to wildcard the returned prompt? The sequencing in the prompt is always "hostname.[seq]#" so is there a way to tell Cirrus to look for "hostname.*#"? Thanks, Will

> As for the statistics polling interval, that is what's really killing your DB. If you must, then only tweak around 5% of the polls to collect at 1 minute intervals. Yepp - I've got only 3% of the total elements running at low intervals. Those devices are all routers or firewalls so they're completion isn't gated by the…

Oddly enough, I did have blank lines in the script but removed them once things were slow to see what would happen. Unfortunately, it didn't make a difference...

> Just remeber that on charts and reports, you can only get a 5 minute sample period, so polling any smaller than 5 minutes is a bit of overkill. While that's true, the interface will also give you candle graphs when there's more data available than it 'natively' displays. If I could get away with polling every second and…

If it's just unrouted VLAN interfaces, then don't bother managing them (same thing applies to EOBC, NULL, etc.) They have no real meaning/expression even if they did report properly (and they tend not to report properly even when you do get data...) Will

Apologies on the delay - basically, trying to intall service packs on anything but the primary poller is a crapshoot. Some SP/HFs are better than others but, for example, the 8.5.1 SP2 simply doesn't install at all on non-primary pollers in my shop. I'll run the setup, it'll shut down the poller if I haven't already, then…

Even dumber than I thought. I would export the file, manipulate it in Excel, save as csv and then attempt to import. The problem is that Excel would convert to a number (with commas) then save as a text field so it wouldn't match the interfaceid in the original table... dumb, dumb, dumb Will

Hey Haley, Thanks but why would this be happening for only Extreme Networks gear? I think that the return is coming back but it's not what Cirrus is expecting. So, for example, the Extreme XOS has a sequence number that increments for every line of code entered. As a result, the prompt is different after every entry. Also,…

Wow. 90 days is pretty crazy depending upon how chatty your devices are. I'm assuming s/he's not a 'technical' manager?

Apologies - found where the other resources were stashed. Not sure I understand the logic but I can get to where I need to go... Will (Back to muddling through...)

For what it's worth, I wanted to do something similar but couldn't figure it out either (so if going the interface route works, please post back a response!) What I wanted was to see which devices Orion wasn't able read properly so I set a Node alert where: Status Description contains Unknown Node Status is not equal to…

er... I just killed of the interface.nodeid as part of the combined key and it imported just fine. Just a stupid user trick I guess.... Will

It's not the switch, it's the receiving station that can't keep up. I have to disagree here. Out discards happen when the egress bandwidth + egress buffer space is completely utilized (i.e., the switch simply has no capacity to transmit a well-formed frame so has to "discard" it.) The receiving station does not 'cause' the…

The new BigIPs don't have an OID for CPU per se. They've got three different OIDs: CPU idle time, CPU transact time, CPU non-transact time. You'd have to do a custom poller that's calcs the 'negative' of idle time... Will

> ELEVEN Polling Engines!!! What an admin nightmare, why so many? Heh heh, like I said - I inherited. I'm glad to hear your target number of 5 or 6 as that's what I was thinking in my head but wanted some validation before I went nutty. (I knew I was in trouble when I called in for support a while back and the first thing…

Not yet but they're actively working on it...

Since I found an Extreme thread, does anyone know how to make Cirrus "download" configs off of Extreme switches? I believe that since the syntax is different on a Cisco (sh run or sh start) vs. Extreme (show config), that's why I'm experiencing this issue. So I guess I am looking for a way to modify the "download" feature…

Yes - the tutorial basically walked me through the production documentation. Do you know why the universal device poller summary status will not allow more than one custom poller to be chosen? I want to be able to see all of the custom pollers that apply to one node in one resource. I've seen screen shots all over thwack…

A couple of things: * If you're running the CatIOS - some versions will 'lose' their SNMP configuration and you won't be able to poll them consistently * If you're running the CatOS - Make sure that you don't need to enable snmp ifindex persistence. Otherwise, every time you boot after any kind of a card change, all the…

Cool - thanks. (Someone should update the docs: http://www.solarwinds.net/NetPerfMon/SolarWinds/wwhelp/wwhimpl/js/html/wwhelp.htm)

Okay - cool. Thanks Denny. A follow-on question: Can I specify one HSE for a group of pollers in order for it to take over for the 'first' failed poller or do I have to associate one-to-one (i.e., if I have four poller, I have to associate four separate HSEs?) Obviously, if I have multiple failures then I've got other…

Hey Denny, I'm ready to beta whenever!! I've been trying to put something together for IP SLA via the custom poller functionality and it's been a real pain. While we also have Cisco phones, I'm really just looking for the core IP SLA functionality to provide per link latency, jiiter, drops, etc. Will

Would it be possible to put more effort into the QA of the updates? I have yet to run any HotFix/SP that doesn't completely get hosed. Of the 8 pollers I run, I have to re-run the update at 5 times on half of them before they "take". (Right now, I'm trying to patch with SP2 and it simply does not work.) Will