Ouze

Alright. Tests with copying files with midnight commander do not reproduce the error. I've just finished installing a new NIC. After doing so, the problem remains: when I copy over a large folder of small files (like 100 mp3's, let's say) via my windows PC from one share to another on the server, a large percentage (like 24 of those 100) error out and do not copy. Larger stuff seems OK - copying over movies is fine, plex playback is OK, etc. Before I start moving on to re-crimping ends, replacing cables, switching ports, replacing the switch, and all that other good stuff... is it 100% sure the network? Is there some kind of diagnostic or test I can do to be sure it's not some weird cache disk bug or something else? I've tried sending over 150 pings from the server to my pc, they all go through, 0% packet loss, like 0.3ms. I just want to make sure I'm definitely moving in the right direction. ''

Well, I just feel dumb as all hell Thank you so much.

I know this is a super stupid problem but I searched the forum and can't figure out the exact answer. I just set up Sonarr and sabnzbd and have them working on a fresh new share on my server. They talk to each other and it downloads files, no problem. The issue is that the files that it gets have incorrect permissions: I can't rename, delete, or move anything because they are owned by SERVER\nobody. I can copy or execute. When i run the tool to fix permissions, they are OK. However every subsequent download has the same issue. Did I screw up the share (which is new), the docker permissions for sabnzbd or sonarr or what? The share's SMB security is set to "public". In the docker paths for both sab and sonarr, they are all read\write. Neither of them is set to privileged. So there is some box I didn't check or fill out and I'm not sure which one it is. https://i.imgur.com/Ta6d7XT.png https://i.imgur.com/ingCSLi.png https://i.imgur.com/DwP8UUV.png https://i.imgur.com/yGfv3dd.png

Thank you, I will try just that.

I'm having sort of an odd problem and not sure what to start looking at to resolve it - if it's a disk error or NIC or what. I created a new music directory for use in plex. From my windows PC, I am copying files from my old music directory (on the unraid share) into the newly created one (also on the unraid share) , while simultaneously listening to said music from the old directory. So, a share to share copy. 1. I get tons of read errors and files that fail to copy 2.) The music I am listening to stops because it (apparently) briefly can't find the source https://i.imgur.com/PtYC66k.png If I just repeat the copy again (skipping overwrites for files already present) they copy OK. I don't see anything in the syslog at all. Last parity check was 18 days ago and was clean. I have fix common problems installed. The only issues it sees are my CPU won't throttle down, and I have a Marvel controller (but I've never seen the issues associated with the Marvel controller) My diags: https://www.dropbox.com/s/qsfwd98lcekywgp/shiva-diagnostics-20171020-1856.zip?dl=0 Thank you for looking.

Nah, we were posting nearly simultaneously - I think he read the post while I was still editing and attaching that smart log. Thanks for confirming it's bad. If a drive is gonna fail, I at least appreciate when it's polite enough to do so within 10 hours of arriving and before any data has been written to it.

Thanks - thats the drive. So those failures in the smart log don't mean a bad drive, then?

I found a similar thread on the forums. I ran this command: root@Shiva:~# cp /var/log/syslog /boot/syslog.txt and then rebooted. My rootFS is now at 22%. My log is here: https://www.dropbox.com/s/vffbouij7j6kg7z/syslog.txt?dl=0 and I've got some of this action going upon reboot: Should I be starting the preclear again? edit: OK, I think the drive is bad. So I suppose this is what happened: Started the preclear, ran for 10 hours, then the drive broke. The script or (whatever) kept trying, and logging failures, until the root fs filled up. Is that right? Smart log for the drive is here: https://www.dropbox.com/s/p2yi7ht6b8d057i/ST3000DM008-2DM166_Z504RTGE-20170802-1829.txt?dl=0 Is that more or less accurate?

Hello all, I have kind of an odd issue. I was preclearing a disk, and had the status window open in the background. I opened the window to check if zeroing was finished, and there was an error about the log being full, which I thought was odd but OK. I closed it. I opened the gui via my browser, and it comes up, but literally everything in it is blank - no parity disk, no array, nothing. I knew there wasn't anything seriously wrong since I was just watching a movie. I checked the shares via Explorer, it all looks good there. I tried going back a directory, which usually displays the flash share - that, I couldn't see, just the main array share. I logged in via putty. Came up OK. I did a df -h and am seeing that the root filesystem is full, which I assume caused everything to break: root@Shiva:~# df -h Filesystem Size Used Avail Use% Mounted on rootfs 1.9G 1.9G 0 100% / tmpfs 2.0G 288K 2.0G 1% /run devtmpfs 1.9G 0 1.9G 0% /dev cgroup_root 2.0G 0 2.0G 0% /sys/fs/cgroup tmpfs 128M 32M 97M 25% /var/log /dev/sda1 1.9G 602M 1.3G 32% /boot /dev/md1 1.9T 1.7T 143G 93% /mnt/disk1 /dev/md2 932G 851G 81G 92% /mnt/disk2 /dev/md3 1.9T 1.7T 157G 92% /mnt/disk3 /dev/md4 1.9T 1.8T 97G 95% /mnt/disk4 /dev/md5 1.9T 1.8T 93G 96% /mnt/disk5 /dev/md6 1.9T 1.7T 140G 93% /mnt/disk6 /dev/md7 1.9T 1.7T 174G 91% /mnt/disk7 /dev/md8 1.9T 1.8T 100G 95% /mnt/disk8 /dev/md9 2.8T 2.4T 348G 88% /mnt/disk9 /dev/md10 2.8T 2.2T 589G 79% /mnt/disk10 /dev/sdf1 233G 16G 217G 7% /mnt/cache shfs 20T 18T 1.9T 91% /mnt/user0 shfs 20T 18T 2.1T 90% /mnt/user /dev/loop0 10G 900M 7.3G 11% /var/lib/docker shm 64M 0 64M 0% /var/lib/docker/containers/8672950ae21eb605d57614b0338eac7de53477e27e979d28230733b879257d5a/shm shm 64M 8.0K 64M 1% /var/lib/docker/containers/92474a28b452f53f58e43394b4716c015b4e95645e3e596ccb346d335374162f/shm I'm not sure where to go from here. A.) How do I clean up that filesystem - rebooting? B.) How do I find out how it filled up? I was running a preclear, but I don't know enough to say "that's what did it". I'm just mentioning it. Thank you!

Thank you both for the information, especially that bit about the data density - I assumed that the speed of any drive was more or less based on rpms alone, so any 7200 would be the same as any other 7200. So, I learned 2 things from this thread. I'll leave it alone until smart says otherwise.

OK, kind of a weird question. One of the first drives I threw in my array was a 1TB Seagate Barracuda. This drive has, so far as I have known, never had an issue, and is currently working OK. That drive is like, 6 and a half years old now. Does it ever make sense to replace a healthy drive just because it's old, or is it better to run it until it dies? It's 7200rpm, if it matters - I don't think I'd see a performance difference in replacing it, though of course the replacement would probably be 3tb. Just curious. Thanks!

Hello - I accidentally broke my Plex somehow and I was wondering if it was possible to restore it instead of recreating it. 1.) I attempted to create a new path. When I hit apply, it threw an error and then Plex disappeared from running apps. 2.) I checked docker. PlexPy was present, but Plex was not. 3.) I rebooted the server. 4.) Same as #2 - PlexPy present, Plex is not. 5.) I re-added a Plex docker. I now have a new Plex instance that has nothing set up in it whatsoever. This is probably where I really screwed up. 6.) When I look at the actual appdata share, I have a plexmediaserver folder, and a PlexMediaServer folder. The latter was created at the time as #5, but the former is dated from December and some subfolders are current to about 30 minutes before this happened. 7.) I don't have the CA backup application set up. Is there any way to fix this without recreating everything? Thanks! ----- Please disregard the above. I've fixed it. I just changed the docker image config to point to the old config location with the old spelling, and it came back up. All the mounts were broken, so I just recreated the mounts. Still, that only took a few minutes vs many, many hours. I feel super excited about fixing my own problem for once. Now I'm gonna have to get that CA backup utility set up...

You're right. It wasn't wholly disabled, it was also enabled in a different place (I think chipset options or something). I disabled it there too and the messages have stopped. Thanks!

It's disabled in bios. I just powered down and double checked.

So, this is an annoyance rather than a problem, I think. I got a weird message that my array tested OK, which I thought odd because I hadn't triggered any testing. I checked my syslog and discovered my syslog is absolute full of the following verbiage: Mar 13 07:33:13 Shiva kernel: blk_update_request: I/O error, dev fd0, sector 0 Mar 13 07:33:13 Shiva kernel: floppy: error -5 while reading block 0 Mar 13 07:34:14 Shiva kernel: blk_update_request: I/O error, dev fd0, sector 0 Mar 13 07:34:14 Shiva kernel: floppy: error -5 while reading block 0 Mar 13 07:35:14 Shiva kernel: blk_update_request: I/O error, dev fd0, sector 0 Mar 13 07:35:14 Shiva kernel: floppy: error -5 while reading block 0 Mar 13 07:36:13 Shiva kernel: blk_update_request: I/O error, dev fd0, sector 0 Mar 13 07:36:13 Shiva kernel: floppy: error -5 while reading block 0 I couldn't find a thread on this. Google seems to indicate this is the system looking for a non-existent floppy drive. Is there any way to track down what application or plugin keeps generating these requests and resolve it? I've attached my diags. shiva-diagnostics-20170313-0735.zip

OK, thank you. Here are my logs: https://www.dropbox.com/sh/v7b2s0aegbshk5z/AAAi_xFT3i6FtrsLPClmwkz9a?dl=0 I also put FCPsyslog_tail.txt in there in case you need that. Thank you again for looking!

Ok, my server has again taken a dump. Here is where I am: 1.) I got a warning from Windows that a backup failed because the network storage was unreachable. 2.) Server is inaccessible in Windows Explorer. 3.) Cannot access the webgui dashboard. 4.) pings are OK. 5.) logging in over telnet is OK. 6.) The Plex gui is responsive, but I can't play any media. Plexpy loads. 7.) Stopped and started Samba via telnet console. root@Shiva:~# samba stop root@Shiva:~# samba start Starting Samba: /usr/sbin/nmbd -D /usr/sbin/smbd -D /usr/sbin/winbindd -D root@Shiva:~# This did not resolve the partial connectivity - pings OK, console OK, Plex gui loads but nothing works. I do have Fix Common Problems running in troubleshooting mode, but not sure what you will want me to attach here. I just entered "diagnostics" into the console and then entered "powerdown".

Thank you, will do!

My Unraid 6.2 server was hung this morning. I'm not sure what happened - not pinging, not able to putty in, no web gui available. I had to jam it down... twice. It was again hung (apparently) the second time. So, my main question is - I don't think I can really troubleshoot this or know what happened without a syslog, right? If so, is there any way to occasionally archive or write the syslog for troubleshooting? This server has been up and stable since 2011, so this isn't really a recurrent issue (knock on wood). I'm running Fix Common Problems and the fast check didn't find anything noteworthy, trying the longer one now. Just curious about the syslog thing, mostly.

This seems mostly resolved. I had one movie that was identified as damaged but it was a terrible movie so, whatever. The files (a handful of iso files for video games) that were stuck in the cache drive seem to have disappeared - they're not on cache anymore, but they never wound up on the array, either. The directories were created, but are empty. I checked the lost+found folder but it's empty as well. Not really a problem there either really, I know which files they are and can restore them from elsewhere. Thanks again for the help.

Thanks

I'm running this now. Should I be replacing this drive, regardless of how the rebuild goes?

Ok, that was fairly straightforward, thanks. It looks like I need to do a rebuild tree: I'm not sure I have the syntax correct. The device in question is md3, and I'm on unraid 6.2 using reiserfs, so what i want to do is: reiserfsck --rebuild-tree /dev/md3 is that correct? Thank you again for the advice.

Ok, here they are. Thanks!

Hello all, I've got a directory that looks like it's permissions have gotten screwed up, or something like that - it looks like the destination directory went read only somehow. When I try to move some disc images to the server, they get stuck on the cache drive and won't leave. I don't really know how it broke or how to resolve this. Here's the relevant section of my syslog: Running Unraid 6.2 if that matters. I assume I need to change the attributes of the directory in question but I don't know the exact commands to do so.