vw-kombi

Its an 800W, there ae 7 drives and if anything it has less power use now, as I removed an SSD file cache drive that I merges onto the main cache drive (and later will be replaced with an NVME.

Unraid is certainly keeping me alive and kicking today.... So the 10TB hot spare had errors also (it was a preclear only on the weekend and was all fine then). My IT shop only had 8TB's in stock, so I had to get that. No time to pre-clear as running with no parity. Connected that up - and as its smaller than the 10TB hot spare I tried - it could not be added. Oh! So - tools/new config and rebuilding parity again. 13 hours it says - I would say fingeres crossed, but that did not help last time.......

Well balls. My hot spare is now also showing read errors during the parity rebuild and it has stopped building........

I bit the bullet and pulled this disk and am now rebuilding on the hot spare. Fingeres crossed in 1 day and 9 hours (it says)

I stopped the array, shutdown. Swapped the sata cable with one from the mobo rather than the hba. Started back up. Disk has an X next to it again, and all says contents emulated. New diags attached. Disk log states this : Oct 18 06:40:47 Tower kernel: ata2: SATA max UDMA/133 abar m131072@0xf7580000 port 0xf7580180 irq 62 Oct 18 06:40:47 Tower kernel: ata2: SATA link down (SStatus 0 SControl 300) Oct 18 06:40:47 Tower kernel: ata2: SATA link down (SStatus 0 SControl 300) Oct 18 06:40:47 Tower kernel: ata2: limiting SATA link speed to 1.5 Gbps Oct 18 06:40:47 Tower kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) Oct 18 06:40:47 Tower kernel: ata2.00: ATA-10: WDC WD80EFZZ-68BTXN0, 81.00A81, max UDMA/133 Oct 18 06:40:47 Tower kernel: ata2.00: 15628053168 sectors, multi 16: LBA48 NCQ (depth 32), AA Oct 18 06:40:47 Tower kernel: ata2.00: configured for UDMA/133 Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] 15628053168 512-byte logical blocks: (8.00 TB/7.28 TiB) Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] 4096-byte physical blocks Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] Write Protect is off Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] Mode Sense: 00 3a 00 00 Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] Preferred minimum I/O size 4096 bytes Oct 18 06:40:47 Tower kernel: sdc: sdc1 Oct 18 06:40:47 Tower kernel: sd 2:0:0:0: [sdc] Attached SCSI disk Oct 18 06:41:20 Tower emhttpd: WDC_WD80EFZZ-68BTXN0_WD-CA2D4JSL (sdc) 512 15628053168 Oct 18 06:41:20 Tower kernel: mdcmd (7): import 6 sdc 64 7814026532 0 WDC_WD80EFZZ-68BTXN0_WD-CA2D4JSL Oct 18 06:41:20 Tower kernel: md: import disk6: (sdc) WDC_WD80EFZZ-68BTXN0_WD-CA2D4JSL size: 7814026532 Oct 18 06:41:20 Tower emhttpd: read SMART /dev/sdc tower-diagnostics-20231018-0643.zip

My disk 6 is disabled and being emulated. Array Devices screen says 2048 errors. disk log shows this a few times : Oct 17 16:41:14 Tower kernel: critical target error, dev sdd, sector 15628052928 op 0x0:(READ) flags 0x80700 phys_seg 1 prio class 2 Oct 17 16:41:14 Tower kernel: Buffer I/O error on dev sdd1, logical block 15628052864, async page read Oct 17 16:41:14 Tower kernel: Buffer I/O error on dev sdd1, logical block 15628052865, async page read I was running three sync processes to compare files on it at the time, plug copying down one large file and up one large file. This morning, I moved the HBA card from one PCI to another PCI location also. I have a hot spare disk that I can move into place - but it is a 10TB and the disk with the issue is an 8TB - my last 8TB. The dashboard says healthy against this disk and a tick - which seems to contradict the errors shown in the array devices screen. Is it possible this relates to the HBA move ? Should I just do a reboot - will it come back - or should I bite the bullet and put the hot swap in there. Timing is terrible as I bought a new 10TB yesterday that has read errors during the pre-clear, tried it in another unraid server and the same thing soon as it is attached so not likely a sata cable issue. Will take time to get a repacement and I hate not having one ready to go. tower-diagnostics-20231017-1640.zip

I checked the instructions, I confirmed it was set to vhost0 and had a mac adress (the upgrade did all that I guess). I host an nfs share on this that UD connects too - hence I notice is same offline. Ping say not reachable. Re network settings, no bonding, no bridging - as per the instruction to keep using macvlan on this version (I have unifi USG and want to keep all that management working).

So those commans I got from the log for when UD does the mount do not work. The mount command throws this - mount point /mnt/remotes/192.168.1.6_Media does not exist I checked the help page but that is all do do with /dev devices - not the UD NFS mounts I am not a linux guy and am way out of my depth here.

400 pages on this - crazy. I have a user script that wakes a backup server, then rsyncs my media to it. I find I have to manually un mount the nfs connection to it of there are errors/warnings in the logs as the server has gone to sleep. I would like to automate the mounting and unmounting as part of the script with suitable delays. I looked at the logs on unraid after a mount, but it does not help me with the actual commands. Can anyone help me with the moount and unmount commands. From the interface - The source is 192.168.1.6:/mnt/user/Media The mount point is 192.168.1.6_Media I think I have it : /sbin/mount -t 'nfs' -o rw,soft,relatime,retrans=4,timeo=300 '192.168.1.6:/mnt/user/Media' '/mnt/remotes/192.168.1.6_Media' /sbin/umount -t nfs '/mnt/remotes/192.168.1.6_Media'

Third check and correct said 0 errors but strangely the email was a repeat of that last one - exactly. Same over 1.5 million errors. seems that email is stuck and sends the same thing. That’s new. added a scn shot of the parity status and the email that was sent.

No disk errors shown against any disk. So I ran a correcting parity check and that finished showing g over 1.5 mil errors. im travelling at the moment so trying to use tailscale and my iPhone to get info, so no access to diagnostics. I’m back in a week.it’s as if it did not write corrections to parity? There is nothing in the sys log except the emails being sent. iv set it off again after ticking off and on the write corrections.

OK - trying that now - same tests : set to high setting (500bps) spun down all disks (if that matters) started the movie 12:08. Checked back at 12:30 - still playing move - OK Checked back at 13:00 - still playing Checked back at 13:30 - still playing Movie ended at 13:50 - at the end of the credits - unraid went to sleep. Its not very forgiving there - should have some sort of time delay or something ? No time to even select something else to watch!

I just proved it def does not work monitoring disks again. I have set 15 minutes now for disk spin down delat in disk settings and the sleep plugin settings also. Wake it up, spin all disks down, start a movie from emby, see one disk active, it then goes to sleep (according to the logs about 40 minutes later). If I get time, I may do this a bit more scientifically later.

Testing the S3 Sleep plugin - I setup a user script to wake it my backup unraid server (new) , do an rsync to it, and then let it sleep again according to the settings on the plugin. For some strange reason, it was sleeping too early. i saw a thread that maybe the array disks were not being read correctly, so I added all the array disks manually to monitor as per that thread. Then I thought I would test it out as I noticed some abnormal sleeping going in. I have the settings set like this : I started the emby container and simulated a failed main server - so wife has moved to the backup emby server (for when I am travelling). The disks all spun down except the one that was streaming the media, I was playing a movie, came back to it later, and movie frooze, unraid backup had gone to sleep..... Its like it is ignoring the active disk ? This is with 6.12.4 os also.

This whole episode has made me re-thing my backup scenarios. The unassigned disk was only reading at 25ish MBps which would have taken ages - I am hoping that was just due to it planning on failing as it did. I was able to copy from my backup unraid server at 110MBps - This makes me think I should have a 'backup server' offsite, and not individual unassigned disks. So - primary unraid server onsite - copied to a cold unraid server onsite (for when I am away and the wife/kids cant fix the main server), then an offsite unraid server just in case house burns down or a break in steals both the current unraid servers.

Yep. I went through stuff i found here online. Forst to try and change uuid at command line, it said it could not do that and to run another command. Did that, then after mouonting, all gone - just a few lost and found folder with little in them.

Well - This has all turned into a right pisser! While my movies and TVShows are saved to another unraid server, that never had enough space for the Sports, Documentaries and StandUpComedy. But that was all saved on my offsite backup disks. This 8TB drive was going really slow to read (about 8MB/s). I stopped it, did an unmount, failed to unmount as its in use. Nothing was using it that I could see, so I did a force umount on it. I connected to another hole in my server, could not mount. went though all the recovery stuff I found on here, in the end, it mounted, but no data!!!!! Moral of the story is I guess - have multiple copies of your data backed up!!!!!! Luckily I copied the homemovies off it, and not copying that to multiple places in the house.

OK - I read up on this now - thanks @trurl. I guess I have been using unraid for so long (8+ years) without ever having seen this, so I rushed into it. I was supposed to stop the array, start it in maintenance mode (never had to do that before), and run the check and repair of the xfs disk. As I have a backup unraid server with the same files that is refreshed weekly, I have not really lost anything, plus my offsite backup disks were still onsite from when run a week ago - so there was that too - I slotted that disk in (unassigned) and am coping it all back. One I get over all this shock, i will use my backup server as a testbed to play with these check/repair commands and document them for any future issues. I imagine the disk2 I replaced with the hot swap disk will be fine if it passes the smart tests, and is re-formatted for use as a new unassigned hot swap disk.

Sorry - posts passed in the night - diagnostics here - but I guess its after the fact now as the 'failed' disk is now no longer part of the array. tower-diagnostics-20230911-1221.zip

So - I aborted the parity build - thinking it was not going to help, then allowed it to format the drive (which was my hot swap drive now), it still had a triangle on it and said emulated. So I stopped the array, tools, new config, let it start the parity rebuild of that, and while its doing that, I am restoring my movied to that disk 2. Looking for some direction in what I 'should' have done here, but I guess I panicked as I dont have a load of time as I am going overseas soon and needed this back up.

Will this partity rebuild just rebuild the same unmountable disk issue ?

All my shares dissapeared. seached here and it said to do a restart of unraid to get them back Clean shutdown, restart - shares showing. Disks however show my disk2 saying Unmnountable: wrong or no filesystem. disk2 for me is part of media share with the movies only. Check share - no movies, only TVShows - everything on non TV shows disk is gone. No problem I thought - I have a hotspare 10TB drive in the system, stopper array, moved that into place, started aray, parity rebuild. But - horribly, it is building the 'faulty disk' data to it - i.e the emulated contents are nothing....... What did i do wrong here ? Should I have done something different ? I have a backup of all this stuff, but still its going to be a day to restore it all.......

Is there ever a chance that this plugin will be incorporated into the unraid system itself ? I have been using it since I started on unrain many years ago and I have so many scripts that it is getting difficult to find them. I just embarked on a script delete, rename and combine process to make it a bit better, but its still in a bit of a mess. Ideally, an ability to sort them in order, or be able to drag and drop like the VM's and Dockers would be ideal.

Just posting that yesterday I went up to 7.3.83 tag. No issue in the upgrade and I messged around with a few settings and all is working fine. Each update on the AP's throws an error about an uplink cant be removed or something like that - I dont have any uplinks so now sure what it is meaning. Everything is working to date........

I am a bit worried about this. I like what macvlan gives me in the unifi controller. I dont want to lose that. I will be holding back on the 6.12.3, many people still reporting reboots on reddit. Pity - as the latest space invader vids on zfs are something I was lookng forward too. Are there people on macvlan and 6.12 ? is it just hardware differences causing this ?