Possible problems with the NMS5.6 distsrv - this can be a significant problem for NoD installations

[Anon Anon]

Distsrv will continually restart itself when deploying a probe if archive is fully populated

I have had to remove approximately 2/3 od the probe packages from the archive to get this distsrv to deploy a probe successfully. The archive folder itself, when full, is 2.05 GB in size with 247 files and 5 folders. When it is this size and you drag and drop a probe out, the distsrv will deploy the probe, but then continually restart itself and not deem the deployment successful and remove the job from the <install_list> section of the distsrv.cfg file. I have to deactivate the probe and then manually edit the config file and remove the job/jobs from the section of the cfg file. Save the file, then restart the. To get a successful deployment without the distsrv probe constantly restarting itself, well over half to 2/3s of the archive folder contents have to be removed/copied to another place, perform the deployments, then restore the packages to the archive. If this is NoD, obviously the customer doesn't have the filesystem access to go thru this process and neither do we. Response from engineering: the developer confirms that this is a problem with the distsrv, where a variable name conflict causes crashes after attempting to unzip certain archives. This will be fixed in the next version. In the mean time, try removing topology_agent_1.53.zip from the archive.

[Anon Anon]

Great catch Jim.  There's nothing here specific about NOD.  This is a problem generally with ALL customers.

 

Jim alludes to needing to write to the filesystem but in fact that's not necessary.   Users should use the GUI's rather than fallback to editing config files and so forth.

 

So a windows user (like a NOD user) who is NOT an windows admin can infact perform this work around as follows:

 

This is probably a defect in distsrv and is not NOD-specific. This case should be escalated to R&D not to NOD operations. There's nothing that NOD operations can do about this. needs to be fixed in the product.

Based upon Jim's comments, the issue can be worked-around by

1 -stopping distsrv
2 - use raw configure (via infrastructure manager) to remove the batch jobs (instead of editing the cfg file by hand)
3 - use windows explorer to copy any probe packages one wishes to save over to a temporary windows folder.

4 - Start distsrv 
5 - use distsrv GUI to delete the unwanted/unnecessary probe packages as well as the ones that you may have copied to a temp windows folder

6 - After the upgrade, then the packages that were saved in the temp folder can be imported into distsrv using the infrastructure manager.

All of the above can be performed by the customer via Infrastructure manager. There's no need to engage NOD operations to do this.