Multiple issues post upgrading to Orion platform 2018.4 NPM 12.4 etc

nafees.khan  wrote: 1. Menu bar disappeared 2. Unable to rollback to old charts 3. Ping stuck on main polling engine 4. Getting server error when opening a node details page 5. CPU and Memory polling stopped post reinstalling job engine and collector service 6. Orion deployment and online diagnostic shows nothing 7. No members menu post enabling ACI polling Had run configuration wizard from cmd(admin) to repair packages, below issue still persists. 2. Unable to rollback to old charts 4. Getting server error when opening a node details page 5. CPU and Memory polling stopped post reinstalling job engine and collector service. 6. Orion deployment and online diagnostic shows nothing 7. No members menu post enabling ACI polling Call logged for each issue.

For # 6 - have you tried the following? The Deployment Health, Updates Available, and Diagnostics tabs are blank and diagnostics cannot be gathered from the web…

Also, was this an in place upgrade or a migration to new server infrastructure?

nafees.khan  wrote: Had run configuration wizard from cmd(admin) to repair packages, below issue still persists. 2. Unable to rollback to old charts Call logged for each issue.

Is this a new installation or an upgrade? For new installs there is no 'rollback' to old charts as the new charts are used by default. If this was an upgrade, did you go through the manual steps to upgrade your chart resources to the new charting library?

If so, what happens when you click 'rollback'?

As a side note, can you tell us why you want to roll back to the older charts? Is there some missing functionality in the new charts that is forcing you to roll back?

- It's an upgrade

- When I click on roll back, it shows migrate options but when I click on CPU & Memory Gauge, It goes to modern chart and not old charts

- Modern charts except for interface are showing blank hence I thought to migrate to old chart and check it but I got to know that statistics are not getting saved in the database due to which charts are empty

- But charts should be able to roll back to old charts

- It's an upgrade

- Support told me to rebuild website, which I am doing it now

Rebuild the Orion Website - SolarWinds Worldwide, LLC. Help and Support

Post rebuilding the website and keeping HTTPS ON at port 443 during configuration wizard with "Generate self signed option", deployment page was working

But when I did website binding to Orion server's FQDN at port 443 and selected our SSL certificate, again deployment page is showing blank

I am having multiple issues as well after the upgrade/migration to 12.4 h1 to a new Windows 2016 server.

-High processor utilization on collector.

-False Down alerts for some devices (seems to repeat same set of devices being down every few hours). Fortunately, it's not many as we have many devices being monitored in our environment.

I have a support ticket for each and so far on the first issue it was running the Configuration Wizard. We'll see how this goes. It's the 2nd time and after about 8 hours the processor utilization will creep up

to a steady 96-100% utilization on all cores. I have also deleted and re-added some of the devices in NPM but still have this weird false alert notifications. Orion will report the device down, then about 2-4 minutes

later report the device back online. Odd thing is I can ping these devices when they are supposedly down. It's not a lost packet or latency issue as these are all plugged into our Cisco 6509 core switch in the same

Data Center.

Anyone experiencing these symptoms as well?

High CPU utilization on collector service and false down alerts seem to be connected as we have had the same scenario. But that was on NPM 12.3 HF6. The issue then was trouble connecting to SQL database because of issues with VM host (Look for alot of SQL connection issues in the logfiles and eventlog)

Also have similar issues, ive tried setting the

bindings to all and rerun the config wizard still no events, alerts menu, also NTA won't install - i get an internal server errorerror de all being ok before the upgrade, I've had to uninstall NTA as it's asking for the sa account! Which is never enabled.

Edit..

With NTA installed (removed and config wizard run I didn't) received a 500 - Internal Server error.

CPU & Memory Gauges are unrelated to new/old charts. These resources now link directly to PerfStack rather than the legacy Custom Chart view. This is true regardless of whether or not you've migrated to new charts or remain on old charts, as the radial gauges themselves are not charts. This is also true of many of the table resources which in the previous release would include links to the legacy Custom Chart view. Those too have been updated to point to PerfStack.

After my upgrade to SolarWindsOrion Platform 2018.2 HF6, VNQM 4.5.0, NCM 7.8, NPM 12.3, NTA 4.2.3, UDT 3.3.1, NetPath 1.1.3 I also had troubles with the Website not displaying.  I remembered this same issue from many upgrades ago and did some digging in my past tickets and found the resolution to the past issue and applied the resolution.  It worked!!  You would have thought that this would have been sorted out by now.

1. Stop all Orion services from the Orion Service Manager. 
2. Open IIS Manager. 
3. In the Connections pane, expand the Server Node and click Application Pools.
4. On the Application Pools page, select SolarWinds Orion Application Pool then click Advanced Settings in the Actions pane.
5. For the Identity property, click the ... button to open the Application Pool Identity dialog box.
6. Select the Built-in account option and select LocalSystem from the list and save the changes.
7. Open CMD as an administrator and run: IISreset.
8. Start all Orion Services from the Orion Service Manager.

Note: If this resolves the issue, you will need to change the AppPool identity from NetworkService to LocalSystem each time that Configuration Wizard is run.

I have a list of Pre and Post Upgrade steps that must be checked and corrected for each upgrade.  This step is now of course on "The List"

I hopes that this helps.