cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 9

Orion 2020.2 Missing Menu Bar

We've noticed that we have 2 of our admin's that experience missing Orion menu's and the hosting server recently completely froze up with so many DCOM errors that the error log overwrote most of the older entries.

Here's a look at the error we see:

DCOM was unable to communicate with the computer 192.xxx.xxx.xxx. using any of the configured protocols; requested by PID c47c (C:\Program Files (x86)\Common Files\SolarWinds\JobEngine.v2\SWJobEngineWorker2x64.exe), while activating CLSID {8BC3F05E-D86B-11D0-A075-00C04FB68820}.

Currently running Orion 2020.2

While the DCOM errors are a concern, the larger concern is fixing the experience of missing menu's for our infrastructure group who live in Orion along with the menu's.

Labels (2)
0 Kudos
4 Replies

Most of the DCOM Errors will not affect your Orion application - maybe polling issues once in a while here and there by JobEngine but then i don't see a major impact there and for sure i dont think thats the reason for issues that you see on your Orion.

We've noticed that we have 2 of our admin's that experience missing Orion menu's and the hosting server recently completely froze up with so many DCOM errors that the error log overwrote most of the older entries.

There could a lot many reasons to this, but you gotta dig more to get to the root cause (Generic reasons are mentioned below)

- DB latency or Orion cant talk to DB, DB should be healthy

- Check if the host on which Orion Application is running is healthy and you have no capacity issues like CPU, Memory, RAM

- Check if IIS is eating up a lot of memory on the server

- Check when does this issue happen are there too many concurrent users and is there a pattern or time at which this issue happens regularly (any activity happening on Orion at that time).

- Any Network latency in your environment etc...

Hope it helps

I'm convinced more than ever that the DCOM errors have nothing to do with the issue but SW support pushed the agenda so I went down that road.  Limiting the number of port connections only worked for about a day but did slow down the amount of DCOM errors.  We have 6 other users that rarely if ever have the issue but the 2 are constantly experiencing it.

0 Kudos

2 are constantly experiencing it, this is strange to me. Can you tell me few things:

1. Where do they login into the Console, is it outside the perimeter from what you do ?

2. How many concurrent users do you have at any given point of time ? Is it more than 40 ?

3. How is your Orion architecture placed ? You just have a Main poller or you have a AWS and these folks login into AWS ?

4. Any specific time at these 2 users login ? Especially do they see this issue at 2:15 AM when data archive happens on Orion ? Are they facing issue at that time ?

5. Lastly how is access enabled to these 2 users ? Is it via AD or Orion App account ?

@allnyabi78 often times DCOM errors are caused by incorrect polling methods. Such as a server being monitored via SNMP, with an application template that pulls WMI statistics, and the assigned application monitor is told to inherit the credential from the node. What happens is SAM will then reach out to the device and attempting to use an SNMP which does not match the assigned fetching method for the assigned application template. I would take a look at some of the IP's it lists in the event log and see if this is the case. A method I've used in the past is to look at all the application monitors with an "Unknown" status and set them to "Unmanaged" and then take a look at the event log and see if those log entries start to reduce in frequency.