What about Hyper-V? We seem to be having issues with all of our VMs after updating the VM Configuration from 5.0 to 8.0.
Thanks!
What about Hyper-V? We seem to be having issues with all of our VMs after updating the VM Configuration from 5.0 to 8.0.
Thanks!
I believe that your additional polling engine needs to be the same OS as your Primary. Even if it's not stated in the documentation I would not recommend them being different, as a difference in windows update levels and .NET framework patch levels can cause errors in the website. Having them on different OS's will almost certainly result in having different patching levels and .NET framework patching levels, which in turn can open a can of worms of potential errors that you can avoid.
Nobody uses hyperv
Anyway coneptually the idea is the same, in the process of updating your vm configuration I suspect that the interface indexes are getting changed. That is the primary indicator Orion uses to correlate the interfaces. One trick I have used in the past is just to go into the database and replace the old interfaceindexes with the new one, causes a seamless transition and orion is non the wiser.
Might be some way to have persistent indexes across upgrades in HyperV, but that's not my wheelhouse so you'd be on your own there.
OK. Figured it out. I got a private messages from a member who suggested these two ideas (I would give credit but since it was a PM, I am assuming he wants to remain anonymous):
SubString or Regex would seem to be options: https://support.solarwinds.com/Success_Center/Network_Performance_Monitor_(NPM)/NPM_Documentation/NPM_Administrator_Guide/0R0_Manage_pollers_using_Device_Studio/070_Formulas_used_for_transforming_Device_Studio_poller_results
This did not work out because I could not figure out the formula.
Have you tried the Truncate option? https://support.solarwinds.com/Success_Center/Network_Performance_Monitor_(NPM)/NPM_Documentation/NPM_Administrator_Guide/0Q0_Monitor_custom_statistics_based_on_MIBs_and_OIDs_with_Universal_Device_Pollers/030_Transform_poller_results
Truncate is great just for routing off the decimal values.
This lead me to figure out using Enumeration. I built out a chart using the temp in C with the unit attached and covering that into F. I could have just used 44C=44 but I just used it to covert the 44C=111. Yes, I could have used the Transformation to covert it to C to F but that pulls the Raw Value of 44C which the value of C throws off the conversion.
This is only impacting ProCurve Switches 2610, 2848, 2910, 2920, 3500, and 5406 running Version 15 or greater OS.
The starting tail of this screen shot above was when I used 44C=44 rather than 44C=111 as for testing.
I've seen in the past where different versions of Orion had bugs that would result in bogus event entries. You may be encountering one such case here, support ticket is definitely the way to proceed.
ayesha.qureshi wrote:
Hello,
My Primary Orion Server is running on Windows 2012 server while my additional polling engine will be on Windows 2016.
When I run the installer for additional polling engine I get following error:
Any thoughts? Following is the detail of the Orion platform.
Orion Platform 2017.1.3 SP3, NCM 7.6, , NPM 12.1, DPAIM 11.0.0, QoE 2.3, NTA 4.2.2, VIM 7.1.0, SAM 6.4.0, NetPath 1.1.0
Do you have plans to upgrade to the latest at any time soon? If so, you may as well migrate your main polling engine to be on Windows 2016 as well. Here's one way to migrate your main polling engine pretty painlessly How to Migrate Orion to a New Server With Zero Downtime that you may be interested in.
4 years later and still not working. We use HTTPS with a web cert and the servers dns name.
duration isn't an option with this trigger, unfortunately (see screenshot).
sorry for the late reply
An very small but possibly useful example for your use case would be:
- Create a custom node property called, say, "IsCriticalServer". For the values of this property, you could just make this a "Yes/No" field
- Then make go to a node and look at this custom property. I believe "Yes/No" custom properties default to "No" so when you create the custom property above , all nodes should be defaulting to "No"
- On all of the "critical servers" you want the special alerts on, set this custom property to "Yes"
When creating alerts in Orion, you select this field as part of the alert. Part of teh issue is you're going to probably want to make it a field in *every* alert. So some made up rules for your "standards for all servers" might read like this:
Alert for nodes that have Volume -> DiskSpace >85% AND Node->CustomProperty->IsCriticalServer = "No" AND (more conditions possibly) .... etc.
Alert for nodes that have Volume ->DiskSpace >90% AND Node->CustomProperty->IsCriticalServer = "No" AND (more conditions possibly) .... etc.
And for your critical servers it might read like this:
Alert for nodes that have Volume -> DiskSpace >=75% AND Node->CustomProperty->IsCriticalServer = "Yes" AND (more conditions possibly) .... etc.
Alert for nodes that have Volume ->DiskSpace >85% AND Node->CustomProperty->IsCriticalServer = "Yes" AND (more conditions possibly) .... etc.
As long as you're careful, the 2 top alert rules will only alert on non critical servers when the disks cross the thresholds, and the bottom 2 only the critical servers.
The good part of doing it this way is you have very granular control of alerting conditions, and can easily make alerting exceptions for servers, volumes, CPU, etc (and these custom properties can be used in a lot of things like reports, groups, alerts, filters in the web UI, etc)
The bad part of it is now you have to manage all of these settings, and if you make a "badly named" one (say, later on someone decides that it's not the *server* that's critical, it's the *application*....well, your custom property is called "IsCriticalServer", not 'IsCriticalApplication' and all your alerting rules are now using the "IsCriticalServer" custom property. In this case you can simply create another custom property 'IsCriticalApplication', and set it on all the nodes...but now you may have to go through all the rules and update them .
For what it's worth, this is why choosing custom property names/values should be done thoughtfully. Another issue is you may end up with a *lot* of them to manage
Also: I like to think of custom properties as "my personal Orion database" so I only add them when I really need them (like for alert conditions) and the data is either not in any other system OR I can pull in data from other systems (via scripts) into "my database" so I don't have to maintain the actual data but I can use it. I rarely add them for "just some data I want to track" , like notes (Orion nodes now have a "notes" area anyways). The reason I bring this up is that if you're also needing to maintain small pieces of data for servers, nodes, etc. (for things that can be other than alerting), custom properties can also be a great place to put it vs using a spreadsheet or text file.
If the Volume could not be reached it would incur a status change. The node, volume and any other entity should remain in the DB/system since it was not deleted and simply change status. graphs and charts would be affected for the entity, not the overall count of, or entities themselves.
If you are on a VM and the label changes I do believe the system would see it as a new volume; that you would have to perform a list resources, and then remove the old volume still present.
Hello Justin,
Take a look at this architecture guide, especially the part for centralised deployment either with or without additional polling engines: https://www.solarwinds.com/-/media/solarwinds/swdcv2/licensed-products/orion-scalability-engines/resources/whitepapers/o…
Best regards,
Steffen
Same result, blocked installation. After more research, it was due to running a newer installer than what the main polling engine is.
Once that was resolved we completed the install. Would like to try this in MS server 2019 to see if it will work.
Hello,
am looking for formula to calculate In and Out percentage.
Can you please help me in getting formula based on below conditions
Device: 1-1, In% - 22.74 & Out% - 22.61
Device: 1-2, In% - 66.11 and Out% - 36.35
Expected result based on 2 devices - In% - 37.79 and Out% - 10.41
Regards,
Chandrashekar B
Hello ALL,
Can someone please help me how to pull the concurrent users for Bluecoat proxysg 900? I need the right OID that can show it. Thanks in advance.
Hi kavana33 ,
Something like below might work as a variable, I don't believe there is an actual column that is "Hardware" (at least none that I could find in the appropriate table). I think they are transforming the NULL's from the Orion.AssetInventory.ServerInformation table 'Virtual Host' that's what the below query does for you. If it's not NULL it returns the raw value
${N=SWQL;M=SELECT isnull (Manufacturer, 'Virtual Host') as Manufacturer FROM Orion.AssetInventory.ServerInformation}
Here's a screenshot
Let me know how that works
Installer on mains server and APE needs to be in the same version to avoid any miscommunications. In case the one used on APE is in lower version - it should get auto-updated automatically, opposite scenario should trigger a big yellow message with option to redownload correct installer from the main server. If neither of those scenarios describes what you experienced, then there must be an issue with communication between service on the main server and installer on APE.
You could try this OID to see if it helps you.
.1.3.6.1.4.1.3417.2.11.3.1.3.2 ----> BLUECOAT-SG-PROXY-MIB:sqProxyHttpClientConnectionsActive