Hi, today I have run into a problem where large file (GB sized files) operations kill the SMB service. At first I thought it had to do with the unzip and other compression scripts I was running on these files and how hard they were hammering the disks. But I have realized that even just copying a GB sized file is failing, 4GB in this last case, and breaking the SMB server in the process. I have realized that I am at or near the high watermark setting where files would be getting put on other disks. Some background data: 1 - Unraid version 7.3.4 2 - Windows 11 PC is the client machine I am working from 3 - I am using UNC paths - no mapped drives 4 - Most of the times I'll see the copy fail. I'll check the SMB status and it shows still running. But a tcp-netconnection test on TCP port 445 to the unraid server fails, so I know the SMB service is hosed in some way. Sometimes it recovers after a few minutes. Other times I have to manually restart smbd, or even reboot the server. 5 - I am running 10gig ethernet 6 - There is nothing in /var/log/syslog during the timeframes when it happens. I don't really know what other logs I should be looking at. 7- Other services (web GUI, NFS) continue working even when this issue happens. 8 - The problem appears to be repeatable. I don't know what other data would be helpful. I wanted to check with the community before opening a ticket with support. I know there are probably some debug level logs or tech.support data I can pull but I wouldn't expect the community to go through those. I realize that if this is high watermark threshold related, I could raise the threshold or change to a different mode (most free), but even if that alleviates the issue and points to this mode of disk balancing being the problem, it isn't necessarily a fix. Thanks for taking the time to read. Any help is appreciated. I have already spent(wasted) time t-shooting with Grok. Sometimes, this is better left to the experts. unraid01-diagnostics-20261111-1407.zip