Our users have been running into issues with pages intermittently disappearing in their PDF editor.
Their Documents folder is redirected to a network share that’s been mounted by GPO.
I have switched some users to another program to rule out a software bug - they still run into the same issue: pages disappearing.
I have a hunch this is a network related issue (possibly temporary loss of file server connectivity) but have no idea where to begin looking. Any help would be greatly appreciated, and a solution will be rewarded with coffee!
And that’s probably where you should start since you already have a hunch. You could do something as simple as a continuous ping overnight from your workstation to see if there is packet loss. Or run an iperf test from the server after hours. Also I’d check the interface statistics on your file server’s network switchport to look for interface errors and resets.
This is where telemetry and instrumentation shines But yes begin testing your compute, memory, storage, and networking to rule each one out or get closer to the root cause.
Most of this is above my paygrade so please excuse any misunderstandings.
Okay well a prelim check of the file server reminds me that we had deployed DFS at some point and turned it off due to it causing issues.
The file server is consistently hitting me with the Warning “The DFS Replication service has been repeatedly prevented from replicating a file due to consistent sharing violations encountered on the file. The service failed to stage a file for replication due to a sharing violation.”
Could this be relevant?
Edit: Nobody is actually using DFS for their home folder - the redirect points directly to the share location. Could DFS still be affecting this?
The errors could be related to network instability but I wouldn’t make the leap to link your PDF issues an replication errors to the same cause. To get an idea if your replication errors are momentary or ongoing you’ll want to run a health report.
Do you have a network admin that can assist your with troubleshooting? They will be able to check the switch port stats of the server.
At this point in the company… I am the network admin lol. I do have access the the switch the fileserver is connected to. Will the Netgear web dashboard have relevant information? Just check the port of the file server to see if there’s packet loss?
I’m not familiar with Netgear’s interface but yes, almost every managed switch will give you access to interface statistics. And just to be clear, I should had said to “check for errors” rather than packet loss in my last post. All you are doing is looking for hardware errors on the network side that could be causing packet loss. But there could still be other reasons for packet loss that aren’t related that particular switch port or hardware at all.
Edit: Got confused and thought I said something I didn’t. Sorry
Is there a convenient way to see which port the server is plugged into? (The server room is quite chaotic)
Also, the file server is running as a Hyper-V VM. Would the host machine be able to supply any pertinent information before I start looking towards the switch?
Sorry, wouldn’t be able to help you out there. Every environment is different and I know better than to make generic recommendations on something like that.
Probably a running ping from your workstation would be a good start. I’d run 3 simultaneous pings to your file server, hyper-v host, and another reliable host such as your router. That’s the most basic place to start if you’re looking for packet loss.
Thank you for your help! I think I’ve honestly hit the limit of what’s possible considering the complexity of my environment.
If I am ever blessed enough to solve this issue I will be sure to report back.
I would highly recommend seeking out a professional. Between a chaotic server closet, consumer grade networking equipment, and server virtualization, if you don’t know what you are looking for or how to use the tools available, you risk creating more issues in the long run.
Also if …
And …
I would try to seek out the person that originally set this stuff up. Even recommend to the big bosses to pay this person to come in and either sort things out or at least document the current configuration so that you can pay someone to create a sustainable and maintainable solution.
But yes begin testing your compute, memory, storage, and networking to rule each one out or get closer to the root cause.