old application crashing hosted on 2016 failover cluster file server
Hi guys, I am having issues with a file server cluster and access to an old application.
The application in question is an old 16 bit application, unfortunately we have to use this app, there is no way around it.
I have a back end setup of 2x windows 2016 servers set up in a SOFS cluster. The cluster has been working great now for a while.
As part of my file server failover plans I decided to move from 2 individual file servers , to 2 file servers in a cluster, with DFS-R to replicate to a failover machine, which we replicate to Azure.
The setup for this server cluster is 2x Windows 2016 virtual machines, with shared VHD set (VHDS) for each of my apps / data. There are about 8 in total. For the other 7, they all work great. The replication is in place and I have not have a single complaint.
As the 16bit app is very important, I moved it last, It seems to work without issue during the mornings, however at about 2pm the program starts to crash for everyone for a few minutes. After these few minutes it will work again. During this period people can go into the share that the drive is mapped to without issue. I cant see anything in the event logs.
This has been happening for almost 2 weeks now, I have tried moving to a different server running the SOFS role, a different hyper-v machine running the file server role, moving the file servers between physical hosts. Nothing is making a difference.
Starting to pull my hair out at this one, if anyone has any advice it would be appreciated!
Every OS in this setup is a 2016 with the august 2018 CU.
The application in question is an old 16 bit application, unfortunately we have to use this app, there is no way around it.
I have a back end setup of 2x windows 2016 servers set up in a SOFS cluster. The cluster has been working great now for a while.
As part of my file server failover plans I decided to move from 2 individual file servers , to 2 file servers in a cluster, with DFS-R to replicate to a failover machine, which we replicate to Azure.
The setup for this server cluster is 2x Windows 2016 virtual machines, with shared VHD set (VHDS) for each of my apps / data. There are about 8 in total. For the other 7, they all work great. The replication is in place and I have not have a single complaint.
As the 16bit app is very important, I moved it last, It seems to work without issue during the mornings, however at about 2pm the program starts to crash for everyone for a few minutes. After these few minutes it will work again. During this period people can go into the share that the drive is mapped to without issue. I cant see anything in the event logs.
This has been happening for almost 2 weeks now, I have tried moving to a different server running the SOFS role, a different hyper-v machine running the file server role, moving the file servers between physical hosts. Nothing is making a difference.
Starting to pull my hair out at this one, if anyone has any advice it would be appreciated!
Every OS in this setup is a 2016 with the august 2018 CU.
Cliff Galiher
Just so I understand, you are having users launch thus 16-but app from a file share directly. And it is running locally on a 32-bit client OS?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Sorry I was working off about 4 hours sleep over 2 days when I wrote the question.
It was an old 16 bit app, that a 3rd party has "upgraded" to work on 32bit.
We have excluded all AV scanning on the directory on clients and servers.
It was an old 16 bit app, that a 3rd party has "upgraded" to work on 32bit.
We have excluded all AV scanning on the directory on clients and servers.