MrSampsonite

arrrg. it's continually frustrating to have to have separate templates based on polling frequency and now 32/64. We end up having a long list of application templates per a node that are all similar except for these two things. For example some IIS or SQL things should be queries more frequently than others. But now we…

is the 64bit switch per component, per template or global? please please tell me it's per component.

Hi Guys, I've totally dropped the ball on this one. I need to test these and get them working. Just been swamped with issues to resolve. Will touch base soon. Thanks!

thanks mcbridea.

Thanks Myanta. Because this server isn't crazy critical I think if I have the time I'm going to do some performance trending in both scenarios (or possibly all three). 1st - RAID10 2nd - RAID5 3rd - RAID1, RAID1 Keep in mind if the OS/Pagefile RAID1 set has great drive stats I can move parts to there and take some of the…

I realize this is old, but I too would like this. We need a way to see disk space growth and have a date when we will run out so we can order more 30 days in advance. Any chance this is available and I just don't know it? I'm on 10.1.1 and apm 3.5.

Thanks. I'll start with that and watch the perfmon stats. Then if things get crazy I can move the database to C, split the RAID 10 into 2 RAID1 sets, move it back and see how the stats are then. Thanks!

Thanks. The lack of ability of WMI to go over 4GB makes it unreliable IMO. We have processes that use a large amount of memory. I have yet to see any real benefit of tracking virtual memory usage, so I guess for now we'll use SNMP for anything inside our network and RPC for the few items outside our network through a NAT…

Thanks for the suggestion, but that assumes there's only one resource we want. I'm not so much looking for that as I am looking for a way to include visual distinguishing bars/graphs into custom resources. In your example I'd only be able to print my report with data (no visual distinction) or I could print SW report with…

nope, that answers it. I was hoping each component that's assigned to a node has a polling frequency (albeit that it inherits from the template the first time) and that via SQL Update query I could tell a particular component assigned to a node to change it's polling frequency, without having to create a whole new…

I am thick. I just saw the preview button before you wrote back and saw that it had the ID in it. I copied the URL. Then I went to the menu bar for Admins since that's what' I'm using and added an external site with this URL. Then I realized I had to log out and back in to see the results. Thanks!

Thought I'd share this. For those who are using APM as well, you can monitor the System: System Up Time perfmon counter to get the value in seconds. This is what I'm doing now. I just have to find a way to change our alerts for reboots to be based on that instead of snmp service.

so you're suggesting with it comes to NPM/APM sharing spindles but getting 2x performance is better than splitting out the pieces of the database per SQL best practices (like tempdb, tlogs, etc)? just checking.

I think you forgot to post the link.

Here's a good example too: Since there's no way to view graphs of multiple components at a time (either one graph per or one graph with all on top of each other), I created a view with a bunch of custom HTML resources on it. I load the view and scroll through a list of graphs all with the same type of component over the…

Thanks, but that's still editing them as a whole right? Meaning I can override the polling settings for all the monitors in a template on a particular node. What I'm looking to do is edit the polling frequency of particular monitors within a template (or if I have to editing just those on each node). For simplicity sake,…

Thanks. Do we use that # somehow to watch the progress?

Thanks for the quick reply. One question though. Do you know how to put SystemUptime into a local variable and then use that in conversion instead of looking it up each time for each part? Multiple that by the number of nodes for a report and you can see how this might help. I'm no SQL person though. That just came from my…

Below is the query we're using for pages per second during our peak time (7pm to 5am) for the last 10 days). Adjust accordingly. SELECT TOP 10000 Nodes.NodeID AS NodeID, Nodes.Caption AS NodeName, APM_AlertsAndReportsData.ComponentName AS Component_Name, AVG(APM_ResponseTime.StatisticData) AS AVERAGE_of_StatisticsData,…

Thanks. I guess we'll just have to wait and see. I'll probably buy a server with enough drive bays to add more in case I need to. I may be naive but I can't see us growing to the 2000+ node level, though possibly at the 2000 APM level. we're using the default polling settings.

by lighter weight do you mean less resources on the orion server or server being monitored? thanks.

Thanks V. forgive me but I've never created such types. Do I set up a script APM monitor first and just type "sql" in the box where it defaults to "vbscript"? Or something else. Thank you so much for your help!

sorry I must be thick. I created a view and added resources. I went to the menu bars admin but it does not show my view as available. I also don't know how to find out the ViewID. SQL query some table?

No vmware here yet. thanks for the heads up though.

Thanks. I realize I could do a report and such. I'm sure there are ways to get what we're looking for, but the graph is plotting these points and average/min/max are common enough that it wouldn't be too hard to do this calculation and put it in. They're already getting the data in the query and they're putting other data…

Right, but pages have tons of data/reports on them. Mine are clustered with information that is helpful. But for the storage team to just get a report on space, it's a bit silly to email them a snapshot of a whole webpage. And then if you do the report size is small as a percentage of the whole page. And if the storage…

we're not running that software on our systems.

Thanks. I thought I'd mention also that many of our graphs are not performance related but rather just stat gathering. Things like file sizes, event logs, and other various counters. There are a myriad of uses for average (along with min and max). To simply put that in text somewhere on each graph would be a tremendous…

Thanks. What do you mean "have no need". How does one determine if they would benefit from it?

how do you do that for linux devices? I've not done that before. we have them, but I don't know them very well yet I'm supposed to monitor them. thanks!