January 29, 201610 yr Hello folks, I foolishly decided to upgrade my unraid box earlier, from 6. what ever it was, to the latest, only jumped about 5 minor releases. But after the upgrade I am back in the same position I was after I last upgraded and had to completely rebuild. Basically the shares disappear. Before the upgrade, I had TV, Films, PVR, Macbackup directories etc. and I can still see them all there on the disks.. but when the server reboots all it comes back with is an empty TV share and an Empty Films share, it's not mounting any of the shares at all.. I've looked all over the place and the only thing I can think of that's left from my original config is the super.dat could it be a screwed super.dat file that's breaking the proper smb share process ? I can see in the log file that something odd is happening to the shares :- Jan 29 22:14:32 NAS-server logger: realtime =none extsz=4096 blocks=0, rtextents=0 Jan 29 22:14:32 NAS-server emhttp: shcmd (57): mkdir -p /mnt/cache Jan 29 22:14:32 NAS-server emhttp: shcmd (58): set -o pipefail ; mount -t xfs -o noatime,nodiratime /dev/sdb1 /mnt/cache |& logger Jan 29 22:14:32 NAS-server kernel: XFS (sdb1): Mounting V4 Filesystem Jan 29 22:14:32 NAS-server kernel: XFS (sdb1): Ending clean mount Jan 29 22:14:32 NAS-server emhttp: shcmd (59): sync Jan 29 22:14:32 NAS-server kernel: mdcmd (43): check CORRECT Jan 29 22:14:32 NAS-server kernel: md: recovery thread woken up ... Jan 29 22:14:32 NAS-server emhttp: unclean shutdown detected Jan 29 22:14:32 NAS-server kernel: md: recovery thread checking parity... Jan 29 22:14:32 NAS-server kernel: md: using 1536k window, over a total of 3907018532 blocks. Jan 29 22:14:34 NAS-server emhttp: shcmd (60): :>/etc/samba/smb-shares.conf Jan 29 22:14:34 NAS-server avahi-daemon[5218]: Files changed, reloading. Jan 29 22:14:34 NAS-server emhttp: shcmd (61): chmod 0777 '/mnt/user/Films' Jan 29 22:14:34 NAS-server emhttp: shcmd (62): chown 'nobody':'users' '/mnt/user/Films' Jan 29 22:14:34 NAS-server emhttp: shcmd (63): chmod 0777 '/mnt/user/TV' Jan 29 22:14:34 NAS-server emhttp: shcmd (64): chown 'nobody':'users' '/mnt/user/TV' Jan 29 22:14:34 NAS-server emhttp: get_filesystem_status: getxattr: /mnt/user/Films Operation not supported Jan 29 22:14:34 NAS-server emhttp: get_filesystem_status: getxattr: /mnt/user/TV Operation not supported Jan 29 22:14:34 NAS-server emhttp: Restart SMB... Jan 29 22:14:34 NAS-server emhttp: shcmd (65): killall -HUP smbd Jan 29 22:14:34 NAS-server emhttp: shcmd (66): cp /etc/avahi/services/smb.service- /etc/avahi/services/smb.service Jan 29 22:14:34 NAS-server avahi-daemon[5218]: Files changed, reloading. Jan 29 22:14:34 NAS-server avahi-daemon[5218]: Service group file /services/smb.service changed, reloading. Jan 29 22:14:34 NAS-server emhttp: shcmd (67): pidof rpc.mountd &> /dev/null Jan 29 22:14:34 NAS-server emhttp: shcmd (68): /etc/rc.d/rc.atalk status If I look at the shares.conf file :- oot@NAS-server:/etc/samba# more smb-shares.conf path = /boot comment = unRAID Sever OS boot device browseable = yes # Public public = yes writeable = yes map archive = no map system = no map hidden = no map readonly = yes create mask = 0777 directory mask = 0777 [Films] path = /mnt/user/Films comment = browseable = yes # Public public = yes writeable = yes [TV] path = /mnt/user/TV comment = browseable = yes # Public public = yes writeable = yes So I go into the GUI, delete these shares, reboot, they come back again!!! where is it getting this info from for starters ? I can see all the disks fine :- root@NAS-server:/mnt# pwd /mnt root@NAS-server:/mnt# ls btrfs/ cache/ disk1/ disk2/ disk3/ disk4/ disk5/ disk6/ disks/ user/ I can see the user shares :- root@NAS-server:/mnt# ls user Films/ TV/ but both TV and Films are empty. if I randomly select a disk :- root@NAS-server:/mnt# cd disk3 root@NAS-server:/mnt/disk3# ls Films/ Macbckup/ PVR/ TV/ backup/ I can see the directories fine. Any ideas how I can put this back together ? I really don't want to have to back up TB's worth of data and then rebuild and restore it all again, it will take weeks. Thanks Steve
January 29, 201610 yr Community Expert Can you access the shares over the network? If so but just can't see them in the webUI then you have an adblocker or something interfering with your browser. Lots of posts about this recently. Shares "automatically" come back because all top level folders of cache or array data disks are automatically a share named the same as the folder. Conversely if you create a share in the webUI unRAID will create top level folders on cache or array data disks as needed according to the share's settings. Your syslog snippet indicates a correcting parity check was started due to a unclean shutdown.
January 29, 201610 yr Author No can't see the shares at all :- root@NAS-server:/etc/samba# df -k Filesystem 1K-blocks Used Available Use% Mounted on tmpfs 131072 1992 129080 2% /var/log /dev/sda1 1950348 178360 1771988 10% /boot /dev/sdj 1562813784 157072392 1404365208 11% /mnt/btrfs /dev/loop0 20971520 14485420 5798068 72% /var/lib/docker /dev/md1 2928835740 844409100 2084426640 29% /mnt/disk1 /dev/md2 2928835740 1461117684 1467718056 50% /mnt/disk2 /dev/md3 2928835740 582564632 2346271108 20% /mnt/disk3 /dev/md4 2928835740 473838620 2454997120 17% /mnt/disk4 /dev/md5 2928835740 481133900 2447701840 17% /mnt/disk5 /dev/md6 2928835740 479445852 2449389888 17% /mnt/disk6 /dev/sdb1 499863384 32944 499830440 1% /mnt/cache /user isn't mounted at all.
January 29, 201610 yr Community Expert In Settings - Global Share Settings, is User Shares enabled? Go to Tools - Diagnostics and post the complete diagnostics zip. You can also get diagnostics from the command line. See V6 help in my sig.
January 29, 201610 yr Author Hi, Yeah I checked the global settings :- Enable disk shares yes Enable user shares yes Included disks All Excluded disks none Will take a look at the command line bits, thanks. Steve nas-server-diagnostics-20160129-2325.zip
January 29, 201610 yr Author some extra info.. from /boot/config root@NAS-server:/boot/config# more share.cfg # Generated settings: shareDisk="yes" shareUser="e" shareUserInclude="" shareUserExclude="" shareSMBEnabled="yes" shareNFSEnabled="no" shareNFSFsid="100" shareAFPEnabled="no" shareInitialOwner="Administrator" shareInitialGroup="Domain Users" shareCacheEnabled="yes" shareCacheFloor="2000000" shareMoverSchedule="40 3 * * *" shareMoverLogging="yes" fuse_remember="330" shareAvahiEnabled="yes" shareAvahiSMBName="%h" shareAvahiSMBModel="Xserve" shareAvahiAFPName="%h-AFP" shareAvahiAFPModel="Xserve" root@NAS-server:/boot/config# cd shares root@NAS-server:/boot/config/shares# ls btrfs.cfg* test.cfg* root@NAS-server:/boot/config/shares# more btrfs.cfg # Generated settings: shareComment="" shareInclude="" shareExclude="" shareUseCache="no" shareCOW="auto" shareAllocator="highwater" shareSplitLevel="" shareFloor="0" shareExport="-" shareSecurity="public" shareReadList="" shareWriteList="" shareExportNFS="-" shareExportNFSFsid="0" shareSecurityNFS="public" shareHostListNFS="" shareExportAFP="-" shareSecurityAFP="public" shareReadListAFP="" shareWriteListAFP="" shareVolsizelimitAFP="" shareVoldbpathAFP="" root@NAS-server:/boot/config/shares# more test.cfg # Generated settings: shareComment="" shareInclude="" shareExclude="" shareUseCache="no" shareCOW="auto" shareAllocator="highwater" shareSplitLevel="" shareFloor="0" shareExport="e" shareSecurity="public" shareReadList="" shareWriteList="" shareExportNFS="-" shareExportNFSFsid="0" shareSecurityNFS="public" shareHostListNFS="" shareExportAFP="-" shareSecurityAFP="public" shareReadListAFP="" shareWriteListAFP="" shareVolsizelimitAFP="" shareVoldbpathAFP="" root@NAS-server:/boot/config/shares# syslog attached. nas-server-syslog-20160129-2328.zip
January 29, 201610 yr Community Expert The syslog is included in diagnostics.zip which is why I didn't ask for it. Some of your disks appear in the syslog without the usual drive serial number designation, and the SMART for those drives is missing in the diagnostics. I suspect some kind of problem with the controller those drives are on.
January 30, 201610 yr Author Odd thing is it's been working fine for years... The serial numbers also seem to show ok in the BUI. plus I can see them getting picked up in the syslog :- Jan 29 22:13:11 NAS-server emhttp: TOSHIBA_THNSNC512GBSJ_13NS10GMTM6Z (sdb) 500107608 Jan 29 22:13:11 NAS-server emhttp: WDC_WD40EFRX-68WT0N0_WD-WCC4E3URAJ3Z (sdc) 3907018584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e6baa50 (sdd) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e72920c (sde) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e730720 (sdf) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e730bb0 (sdg) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e6cbe68 (sdh) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35000cca03e728ddc (sdi) 2930266584 Jan 29 22:13:11 NAS-server emhttp: 35001e82002764170 (sdj) 1562813784 Jan 29 22:13:11 NAS-server emhttp: array slots: 24 Jan 29 22:13:11 NAS-server emhttp: cache slots: 1 Jan 29 22:13:11 NAS-server kernel: mdcmd (1): import 0 8,32 3907018532 WDC_WD40EFRX-68WT0N0_WD-WCC4E3URAJ3Z Jan 29 22:13:11 NAS-server kernel: md: import disk0: [8,32] (sdc) WDC_WD40EFRX-68WT0N0_WD-WCC4E3URAJ3Z size: 3907018532 Jan 29 22:13:11 NAS-server kernel: mdcmd (2): import 1 8,48 2930266532 35000cca03e6baa50 Jan 29 22:13:11 NAS-server kernel: md: import disk1: [8,48] (sdd) 35000cca03e6baa50 size: 2930266532 Jan 29 22:13:11 NAS-server kernel: mdcmd (3): import 2 8,64 2930266532 35000cca03e72920c Jan 29 22:13:11 NAS-server kernel: md: import disk2: [8,64] (sde) 35000cca03e72920c size: 2930266532 Jan 29 22:13:11 NAS-server kernel: mdcmd (4): import 3 8,80 2930266532 35000cca03e730720 Jan 29 22:13:11 NAS-server kernel: md: import disk3: [8,80] (sdf) 35000cca03e730720 size: 2930266532 Jan 29 22:13:11 NAS-server kernel: mdcmd (5): import 4 8,96 2930266532 35000cca03e730bb0 Jan 29 22:13:11 NAS-server kernel: md: import disk4: [8,96] (sdg) 35000cca03e730bb0 size: 2930266532 Jan 29 22:13:11 NAS-server kernel: mdcmd (6): import 5 8,112 2930266532 35000cca03e6cbe68 Jan 29 22:13:11 NAS-server kernel: md: import disk5: [8,112] (sdh) 35000cca03e6cbe68 size: 2930266532 Jan 29 22:13:11 NAS-server kernel: mdcmd (7): import 6 8,128 2930266532 35000cca03e728ddc Jan 29 22:13:11 NAS-server kernel: md: import disk6: [8,128] (sdi) 35000cca03e728ddc size: 2930266532 Thanks Steve
January 30, 201610 yr Community Expert Those are note the serial numbers, they look like SAS addresses, never seen that on what looks like a SAS2LP, it may not be related to your share issue, but that’s certainly the reason you’re missing several SMART reports, and probably temp info for those disks.
January 30, 201610 yr Author ahaaaa yes your right they are SaaS II discs.. it's getting amusing now, I'm getting calls from brother and parents etc. telling me my plex server is down.. It's got to be something to do with that super.dat, I'm pretty sure that's the only thing I had left from my old pre stable V6 build.. can I jut blast this file and rebuild without loosing my data if I put everything back the way it currently is ? I can save a copy of the file if some one wants to pick it apart later.. Thanks Steve
January 30, 201610 yr Community Expert You can, and you don’t have to rebuild parity. Go to tools a do a new config. This will rename super.dat to super.old, re-assign all your disks, make sure you assign the old parity to the parity slot, check “parity is already valid” and start array.
January 30, 201610 yr Author ok cool, I'll try that see if it brings the beast back to life. Thanks Steve
January 30, 201610 yr Author Aggggg bleeeeeeeepp So I deleted the super.dat rebooted and reset the disks. start the array, back in exactly the same place. Jan 30 19:25:50 NAS-server logger: = sectsz=512 sunit=0 blks, lazy-count=1 Jan 30 19:25:50 NAS-server logger: realtime =none extsz=4096 blocks=0, rtextents=0 Jan 30 19:25:50 NAS-server emhttp: shcmd (69): mkdir -p /mnt/disk6 Jan 30 19:25:50 NAS-server emhttp: shcmd (70): set -o pipefail ; mount -t xfs -o noatime,nodiratime /dev/md6 /mnt/disk6 |& logger Jan 30 19:25:50 NAS-server kernel: XFS (md6): Mounting V4 Filesystem Jan 30 19:25:51 NAS-server kernel: XFS (md6): Ending clean mount Jan 30 19:25:51 NAS-server emhttp: shcmd (71): xfs_growfs /mnt/disk6 |& logger Jan 30 19:25:51 NAS-server logger: meta-data=/dev/md6 isize=256 agcount=4, agsize=183141659 blks Jan 30 19:25:51 NAS-server logger: = sectsz=512 attr=2, projid32bit=0 Jan 30 19:25:51 NAS-server logger: = crc=0 finobt=0 Jan 30 19:25:51 NAS-server logger: data = bsize=4096 blocks=732566633, imaxpct=5 Jan 30 19:25:51 NAS-server logger: = sunit=0 swidth=0 blks Jan 30 19:25:51 NAS-server logger: naming =version 2 bsize=4096 ascii-ci=0 ftype=0 Jan 30 19:25:51 NAS-server logger: log =internal bsize=4096 blocks=357698, version=2 Jan 30 19:25:51 NAS-server logger: = sectsz=512 sunit=0 blks, lazy-count=1 Jan 30 19:25:51 NAS-server logger: realtime =none extsz=4096 blocks=0, rtextents=0 Jan 30 19:25:51 NAS-server emhttp: shcmd (72): mkdir -p /mnt/cache Jan 30 19:25:51 NAS-server emhttp: shcmd (73): set -o pipefail ; mount -t xfs -o noatime,nodiratime /dev/sdb1 /mnt/cache |& logger Jan 30 19:25:51 NAS-server kernel: XFS (sdb1): Mounting V4 Filesystem Jan 30 19:25:51 NAS-server kernel: XFS (sdb1): Ending clean mount Jan 30 19:25:51 NAS-server emhttp: shcmd (74): sync Jan 30 19:25:51 NAS-server emhttp: skipping parity sync Jan 30 19:25:51 NAS-server emhttp: shcmd (75): :>/etc/samba/smb-shares.conf Jan 30 19:25:51 NAS-server avahi-daemon[5222]: Files changed, reloading. Jan 30 19:25:51 NAS-server emhttp: shcmd (76): chmod 0777 '/mnt/user/Films' Jan 30 19:25:51 NAS-server emhttp: shcmd (77): chown 'nobody':'users' '/mnt/user/Films' Jan 30 19:25:51 NAS-server emhttp: shcmd (78): chmod 0777 '/mnt/user/TV' Jan 30 19:25:51 NAS-server emhttp: shcmd (79): chown 'nobody':'users' '/mnt/user/TV' Jan 30 19:25:51 NAS-server emhttp: get_filesystem_status: getxattr: /mnt/user/Films Operation not supported Jan 30 19:25:51 NAS-server emhttp: get_filesystem_status: getxattr: /mnt/user/TV Operation not supported Jan 30 19:25:51 NAS-server emhttp: Restart SMB... Where is it getting those 2 mount points from ? if I delete the shares they come back after reboot.. Short of completely formatting the flash drive and rebuilding I am at a loss.. I don't want to loose my docker containers though, it's taken me ages to get them set the way I want them Any advice welcome... pleaaaassssseee
January 30, 201610 yr Community Expert All top level folders on a disk are automatically turned into User Shares. I seem to remember there being an issue when getxattr failed in earlier releases of unRAID and there being a command that could be run to fix this. However I cannot find it off-hand and my memory may be faulty.
January 30, 201610 yr Author hahahaha this thing is doing my nut in..it's like the array is possessed So I stopped the array, deleted the TV and Films share again, turned off the shares in settings global shares settings and rebooted. I rebooted the the array, started it.. The shares are back and there, even though the global shares setting is still turned off.. it's magic..only not in a good way.
January 30, 201610 yr Author OK I decided to pay for this software and now I'm starting to think I've been done over... not in a bad way but a slightly annoying way.. surely the developers must know whats gone wonky here.. I can see all the data is still there, it's just that the software has decided my shares don't exist.. what the f**k is stopping the array from picking them up ? I am seriously thinking I need to switch back to freenas.. but I don't want too I can still see all the data is there :- root@NAS-server:/mnt/disk4# ls ../disk1 American\ Horror\ Story/ Downloads/ Films/ TV/ test/ root@NAS-server:/mnt/disk4# ls ../disk2 Downloads/ Films/ Macbckup/ TV/ backup/ btrfs@ root@NAS-server:/mnt/disk4# ls ../disk3 American\ Horror\ Story/ Films/ Macbckup/ PVR/ TV/ backup/ root@NAS-server:/mnt/disk4# ls ../disk4 American\ Horror\ Story/ TV/ root@NAS-server:/mnt/disk4# ls ../disk5 American\ Horror\ Story/ TV/ root@NAS-server:/mnt/disk4# ls ../disk6 American\ Horror\ Story/ TV/ root@NAS-server:/mnt/disk4# BattleStar\ Galactica/ Childeren\ of\ Dune/ Fargo\ (2014)/ House\ MD/ The\ 100/ The\ Wire/ Black\ Adder\ (1983)/ Continuum/ Friends/ Revolution/ The\ Big\ Bang\ Theory/ Under\ The\ Dome/ Boston\ Legal/ Dune/ Game\ Of\ Thrones/ Spooks/ The\ Sopranos/ Breaking\ Bad/ Falling\ Skies/ Heroes/ Star.Trek.The.Next.Generation.Complete.NTSC.DVD.DD5.1.x264-JCH/ The\ Walking\ Dead/ root@NAS-server:/mnt/disk2/TV# everything is there...the software has gone nuts some where.
January 30, 201610 yr Author Seriously I pay for this software.. can't someone from the unraid team at least respond.. it can't be that hard to figure out surely ? Something has gone wonky with the mounting..I want my plex server back there is bugger all on TV
January 30, 201610 yr Just to point out that it is the weekend and Limetech are not obliged to read every post - try emailing.
January 30, 201610 yr As Peter noted, it is the weekend ... and one other point: This forum is user-supported ... any help you're getting here is from folks who simply provide free help to fellow users. It's a "community forum" ... not a Limetech support forum. LimeTech does monitor it occasionally and respond, but the "official" way to get help is to e-mail a request to [email protected]
January 31, 201610 yr Author fair point folks, guess I was just getting a little frustrated with having nothing to watch
January 31, 201610 yr Author I now have this working again.. I stopped the array.. Deleted the shares via the GUI. started the global share option again. started the array, bingo the shares all mounted again... So I have a work around at least shame it doesn't do this properly at boot up any more.
January 31, 201610 yr A couple of thoughts ... => Now that you deleted the shares and let them be re-generated with global shares enabled, have you tried re-booting to confirm whether or not all is now resolved? ... If that's not the case, you may want to (a) save your key file; (b) reformat the flash drive and start with a fresh install of v6.1.7 => be sure you know which drives are assigned to non-data functions [parity & cache] so you can assign all the drives correctly. You'll have to reinstall Plex if you do that ... but it sounds like you've had something corrupted for some time (since you indicated you had similar issues the last time you upgraded as well) => Are you using the Plex Docker?
January 31, 201610 yr Author Hi thanks for the reply Gary, Now that I know I can rebuild the array from scratch I will look at it.. I haven't rebooted because I want to watch something, but I will see what happens.. The only slightly annoying thing with another complete wipe would be to rebuild my pleX server etc again.. but they are all on a separate SSD which I mount as a btrfs disk, so I hope if I rebuild I can pull the docker apps back from the SSD..(is that possible ? ) it's taken about a year to get plex, sickrage, transmission etc. all to play nicely with each other.. (and before I am bashed for downloading stuff, I buy every single TV program & film I watch on DVD, as soon as they come out on DVD, friends find it amusing that all my DVD's are still wrapped in their cellophane ... I just don't like to wait). Thanks
January 31, 201610 yr ... and before I am bashed for downloading stuff ... Somehow I doubt that will happen -- the number of folks who use sickrage, transmission, coach potato, etc. to gather material for Plex is clearly pretty high. FWIW, I don't download any media [i buy movies on DVD, and record TV shows I want on my HTPC, which has 9 tuners to ensure there are never any conflicts if multiple things are on at once] ... but there are a LOT of folks who have clearly built massive media collections through "other means"
January 31, 201610 yr Community Expert ... so I hope if I rebuild I can pull the docker apps back from the SSD... The recommended and easy way to recover docker apps is just to reinstall them using the template for each docker that was saved to your flash when you created the docker. If you go to the Docker page and Add container, you will get a blank form. The dropdown at the top will let you reload that form from any template that was saved. These will all have a name beginning with "my-", such as "my-transmission". The unRAID Add Container page is just a form to let you specify the parameters used by the docker run command, such as volume mappings. All of your dockers "forms" have already been saved and can be selected to let you create the docker again with all of the parameters you set before. Then the docker code will be downloaded again, and the docker will be run with the settings from the form so it will use all of the volume mappings and other settings you made before, and it should all work just like you had it.
Archived
This topic is now archived and is closed to further replies.