Hello, I'm not sure if I'm correct to ask this here, sorry if it's not. I use Debian 11 with 5.10.0-8-amd64 #1 SMP Debian 5.10.46-4 (2021-08-03) x86_64 GNU/Linux as Kernel on a ThinkPad T440. I also have a network attached storage device which I access over nfs. I load it via //192.168.0.103/fileserver /home/norman/fileserver cifs auto,rw,rsize=32768,wsize=32768,uid=1000,gid=1000,defaults,iocharset=utf8,username=<HERECOMESMYUSERNAME>,password=<HERECOMESMYPASSWORD>,vers=1.0 0 0 in my fstab. It all works fine if I start using it while I am at my docking station (with LAN-cable to my router and IP 192.168.0.109), but it fails to access (and even re-mount and umount, when not using umount -a -t cifs -l) on wifi (having 192.168.0.102 there). But when I umount before unplugging, and then unplug and re-mout, it works just fine. When I am in the situations that it doesn't work, all windows keep stuck and I cannot interact with them anymore, if have anything at all to do with my NAS (even ls ~, because it's mounted in ~). I am not a kernel programmer, so I cannot look into the source code to tell why this happens, but I can speculate. I believe the NFS-driver is somehow linked to the interface it was first connected on. So that, if you connect on enp0s25 first, then plug it out, and try to access the same host (the NAS) when enp0s25 is down, it fails to do something, and there seems to be no timeout set by default that prevents the system from crashing. So in short: use the IP so long as it works, but if it doesn't, use the MAC-address to arp-scan the network, maybe you're on a different port now. Is this reasonable? If this is the wrong mailing list, I'm very sorry. I'd like to be corrected on where to send this to be dealt this properly. Thanks, Norman
