dstoe

Hi, I have not been here for a while....6 years 😁,....but I am now looking into building a big file server for larger video files. During the pandemic we have built a video studio in our company and I need a place to store our productions. As I still have an older unraid setup in the garage, I was thinking, maybe I could ressurect it with new HDDs and SSD cache. Here is the old setup: Supermicro MB775 I-3210 X7SBE 2x Supermicro Cont SATA2 AOC-SAT2-MV8 PCI-X Intel Core 2 Duo E5200 4gb ddr2 800 Corsair TX750W Norco 4020 Would it be worth using the old hardware or at least the case? As the data I produce in a shooting day is in the range between 50GB and 2TB, is unraid quick enough for me, if I want to overnight backup? Thanks in advance for your insights! Dominik

Thanks, I will try it now. There is a lot to do

thanks, seems reasonable to change the filesystem. I will reread this thread.

is it worth it to change the file system?

Hi, I recently revived a very old version (4.something) of my unraid to the new age (6.1.6). My server has numerous old hdds, with 500GB each and reisrfs. Now I have some 3TB drives, that I would like use instead. If I understood correctly, just exchanging one drive at a time would bring me reiserfs 3TB drives(after rebuild). If I expand my system with a new 3TB xfs drive, I need to copy the data. What happens with the user shares? e.g disk 1 is 500GB and has partial files of 4 user shares If I copy all that to a new 3TB drive, then stop the array, take out the old 500GB, restart the array. Would I need to rebuild parity for everything? Would the user shares be intact? Thanks a lot for your help! Dominik

YES! Very nice. Thanks for this one!

thanks, i will checkout the powerdown script. For now, I killed the remaining two mover processes, stopped the array and rebooted. All disks showed up fine. Then I shutdown and added 5 more disks.And now i can leave office, knowing the server does something useful overnight.

I tried to stop the array, while mover was pushing files from the cache drive. Unraid seems to have stopped the other drives which now all show up as unformatted. I searched a bit around the forum and found hints, that this is some sort of bug. The shutdown button does not show up, as the array could not be stopped properly. Important learning from the search : I make sure not to push the format button! Should i now wait for the mover to finish, before trying to stop the array again? Should i somehow try to kill the mover app? If so, how would I do that? I actually wanted to shutdown, in order add some more disks to the array. Here the last part of the syslog. I started the mover manually. Mar 26 15:19:07 Tower logger: mover started Mar 26 15:19:14 Tower logger: `./Projekte/Vademecum Minze Phantom RAW/Nicole.small.mov' -> `/mnt/user0/./Projekte/Vademecum Minze Phantom RAW/Nicole.small.mov' Mar 26 15:19:14 Tower logger: removed `./Projekte/Vademecum Minze Phantom RAW/Nicole.small.mov' Mar 26 15:36:11 Tower emhttp: shcmd (83): /usr/sbin/hdparm -y /dev/sde >/dev/null Mar 26 15:53:54 Tower shfs(1): duplicate object: /mnt/disk4/Projekte/Vademecum Minze Phantom RAW/f25.cine Mar 26 15:57:08 Tower shfs(1): duplicate object: /mnt/disk4/Projekte/Vademecum Minze Phantom RAW/f25.cine Mar 26 15:58:08 Tower last message repeated 2 times Mar 26 16:10:11 Tower emhttp: shcmd (84): /usr/sbin/hdparm -y /dev/sdd >/dev/null Mar 26 16:11:12 Tower emhttp: shcmd (85): /usr/sbin/hdparm -y /dev/sdb >/dev/null Mar 26 16:42:14 Tower emhttp: shcmd (86): /etc/rc.d/rc.samba stop >/dev/null Mar 26 16:42:14 Tower emhttp: shcmd (87): /etc/rc.d/rc.nfsd stop >/dev/null Mar 26 16:42:15 Tower emhttp: Spinning up all drives... Mar 26 16:42:15 Tower emhttp: shcmd (88): sync Mar 26 16:42:57 Tower emhttp: shcmd (89): umount /mnt/user Mar 26 16:42:57 Tower emhttp: shcmd (90): rmdir /mnt/user Mar 26 16:42:57 Tower emhttp: shcmd (91): umount /mnt/user0 Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (91): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd (92): rmdir /mnt/user0 Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (92): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd (93): umount /mnt/disk1 Mar 26 16:42:57 Tower emhttp: shcmd (93): umount /mnt/disk2 Mar 26 16:42:57 Tower emhttp: shcmd (93): umount /mnt/disk3 Mar 26 16:42:57 Tower emhttp: shcmd (93): umount /mnt/disk4 Mar 26 16:42:57 Tower emhttp: shcmd (93): umount /mnt/cache Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (93): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd (94): rmdir /mnt/disk4 Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (94): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (93): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd (94): rmdir /mnt/cache Mar 26 16:42:57 Tower emhttp: shcmd: shcmd (94): exit status: 1 Mar 26 16:42:57 Tower emhttp: shcmd (94): rmdir /mnt/disk2 Mar 26 16:42:57 Tower emhttp: shcmd (94): rmdir /mnt/disk3 Mar 26 16:42:57 Tower emhttp: shcmd (94): rmdir /mnt/disk1 Mar 26 16:42:57 Tower kernel: mdcmd (1188): stop Mar 26 16:42:57 Tower kernel: md: 2 devices still in use. Mar 26 16:42:58 Tower emhttp: shcmd (93): cp /etc/exports- /etc/exports Mar 26 16:42:58 Tower emhttp: shcmd (94): mkdir -m 700 /mnt/user Mar 26 16:42:58 Tower emhttp: shcmd (95): /usr/local/sbin/shfs /mnt/user 1 2000000 Mar 26 16:42:58 Tower emhttp: shcmd (96): mkdir -m 700 /mnt/user0 Mar 26 16:42:58 Tower emhttp: shcmd: shcmd (96): exit status: 1 Mar 26 16:42:58 Tower emhttp: shcmd (97): /usr/local/sbin/shfs /mnt/user0 0 Mar 26 16:42:58 Tower emhttp: shcmd: shcmd (97): exit status: 1 Mar 26 16:42:59 Tower emhttp: shcmd (98): /etc/rc.d/rc.samba start >/dev/null Mar 26 16:42:59 Tower emhttp: shcmd (99): /etc/rc.d/rc.nfsd start >/dev/null Mar 26 16:43:23 Tower emhttp: shcmd (100): /etc/rc.d/rc.samba stop >/dev/null Mar 26 16:43:23 Tower emhttp: shcmd (101): /etc/rc.d/rc.nfsd stop >/dev/null Mar 26 16:43:24 Tower emhttp: Spinning up all drives... Mar 26 16:43:24 Tower emhttp: shcmd (102): sync Mar 26 16:43:41 Tower emhttp: shcmd (103): umount /mnt/user Mar 26 16:43:41 Tower emhttp: shcmd (104): rmdir /mnt/user Mar 26 16:43:41 Tower emhttp: shcmd (105): umount /mnt/user0 Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (105): exit status: 1 Mar 26 16:43:41 Tower emhttp: shcmd (106): rmdir /mnt/user0 Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (106): exit status: 1 Mar 26 16:43:41 Tower emhttp: shcmd (107): umount /mnt/disk4 Mar 26 16:43:41 Tower emhttp: shcmd (107): umount /mnt/cache Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (107): exit status: 1 Mar 26 16:43:41 Tower emhttp: shcmd (108): rmdir /mnt/disk4 Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (107): exit status: 1 Mar 26 16:43:41 Tower emhttp: shcmd (108): rmdir /mnt/cache Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (108): exit status: 1 Mar 26 16:43:41 Tower emhttp: shcmd: shcmd (108): exit status: 1 Mar 26 16:43:41 Tower kernel: mdcmd (1191): stop Mar 26 16:43:41 Tower kernel: md: 2 devices still in use. Mar 26 16:43:42 Tower emhttp: shcmd (107): cp /etc/exports- /etc/exports Mar 26 16:43:42 Tower emhttp: shcmd (108): mkdir -m 700 /mnt/user Mar 26 16:43:42 Tower emhttp: shcmd (109): /usr/local/sbin/shfs /mnt/user 1 2000000 Mar 26 16:43:42 Tower emhttp: shcmd (110): mkdir -m 700 /mnt/user0 Mar 26 16:43:42 Tower emhttp: shcmd: shcmd (110): exit status: 1 Mar 26 16:43:42 Tower emhttp: shcmd (111): /usr/local/sbin/shfs /mnt/user0 0 Mar 26 16:43:42 Tower emhttp: shcmd: shcmd (111): exit status: 1 Mar 26 16:43:43 Tower emhttp: shcmd (112): /etc/rc.d/rc.samba start >/dev/null Mar 26 16:43:43 Tower emhttp: shcmd (113): /etc/rc.d/rc.nfsd start >/dev/null Mar 26 16:59:13 Tower emhttp: shcmd (114): /usr/sbin/hdparm -y /dev/sdb >/dev/null Mar 26 16:59:13 Tower emhttp: shcmd (115): /usr/sbin/hdparm -y /dev/sdd >/dev/null Mar 26 16:59:14 Tower emhttp: shcmd (116): /usr/sbin/hdparm -y /dev/sde >/dev/null regards Dominik

Ok, nevermind. I took a different approach and setup a separate XP system in the office. On that one I just run Total Commander, which so far works flawless. regards Dominik

Ok, thanks for the input. Unmenu starts without me doing anything. The Go Script line works nice (remember I am a linux newb). I also installed a cache disk. This will speedup things a bit. But there is still something I can not figure out. Lets say I mount a disk full of data, that I want to copy to the array. Now I want to select some, but not all of the data on the disk to copy that to the array. How can I batch copy several files with midnight commander? If I select all, only files, but no folders are marked. I would really like a tool like total commander on the Unraid machine. Any ideas? regards Dominik

To get unMenu to boot on startup you need to add a line to your go script that reads: /boot/unmenu/uu To speed up transfers to the system you could also disable the parity drive. That will allow you to transfer the files and then you can reassign the parity drive. Ok, but how do I disable the parity drive?

a little update. I finally received all the bits and pieces for my office system. Getting the mainboard locally was taking more time, than shipping the Norco 4020 from the US to germany. Anyway, I build the system consisting of this hardware: Supermicro MB775 I-3210 X7SBE 2x Supermicro Cont SATA2 AOC-SAT2-MV8 PCI-X Intel Core 2 Duo E5200 4gb ddr2 800 Corsair TX750W Norco 4020 Running Unraid 4.4.2 For the start I use a 1,5TB Seagate as parity drive and 3x 500G Samsungs. While filling up, I will add more disks, but I wanted to try out the system first and get used to it. So far the only software addition has been Unmenu. Although I have not figured out how to start it automatically on boot. Yesterday I copied some 400GB to the array and right now I do a parity check. Unmenu shows 70,000KB/s as speed after the first 15%. Some observations: Had to change the NTP server, as it would not show the correct time. The Norco is far too loud. I would have to swap the divider plate with a custom one and go for slow 120mm fans, if I wanted a similar system at home. Copying to the array is slow, I will think about a cache drive. Unmenu is very useful and should be a part of the original package. I want to thank everybody who has given me advice. Thanks! regards Dominik

Does anybody know of a mainboard with 3 PCI-E x4 slots? regards Dominik

Thanks, that did it. Sometimes it is so easy

I just received my Norco 4020 case and I can not get the lid of. Has anybody, who knows this case, got an idea? Thanks Dominik