JorgeB

Unraid driver crashed as the rebuild was starting: Nov 27 09:27:41 Tower kernel: mdcmd (35): start RECON_DISK Nov 27 09:27:41 Tower kernel: unraid: allocating 107910K for 1280 stripes (21 disks) Nov 27 09:27:41 Tower kernel: sysfs: cannot create duplicate filename '/devices/virtual/block/md1' Nov 27 09:27:41 Tower kernel: CPU: 15 PID: 8790 Comm: emhttpd Not tainted 5.19.17-Unraid #2 Nov 27 09:27:41 Tower kernel: Hardware name: Dell Inc Compellent SC8000/0VRCY5, BIOS 2.9.0 12/06/2019 Nov 27 09:27:41 Tower kernel: Call Trace: Nov 27 09:27:41 Tower kernel: <TASK> Nov 27 09:27:41 Tower kernel: dump_stack_lvl+0x44/0x5c Nov 27 09:27:41 Tower kernel: sysfs_warn_dup+0x56/0x62 Nov 27 09:27:41 Tower kernel: sysfs_create_dir_ns+0x9b/0xc7 Nov 27 09:27:41 Tower kernel: kobject_add_internal+0xc3/0x1cb Nov 27 09:27:41 Tower kernel: kobject_add+0xa6/0xcf Nov 27 09:27:41 Tower kernel: ? preempt_latency_start+0x2b/0x46 Nov 27 09:27:41 Tower kernel: device_add+0x34f/0x843 Nov 27 09:27:41 Tower kernel: ? dev_set_name+0x53/0x6e Nov 27 09:27:41 Tower kernel: device_add_disk+0xf3/0x30d Nov 27 09:27:41 Tower kernel: md_proc_write+0x1142/0x1579 [md_mod] Nov 27 09:27:41 Tower kernel: ? mntput_no_expire+0x4f/0x1f2 Nov 27 09:27:41 Tower kernel: ? terminate_walk+0x48/0x6e Nov 27 09:27:41 Tower kernel: ? path_openat+0x9cc/0xaa9 Nov 27 09:27:41 Tower kernel: ? asm_sysvec_apic_timer_interrupt+0x10/0x20 Nov 27 09:27:41 Tower kernel: ? do_filp_open+0x8e/0xb8 Nov 27 09:27:41 Tower kernel: ? virt_to_slab+0x5/0x19 Nov 27 09:27:41 Tower kernel: ? memcg_slab_free_hook+0x4b/0xf9 Nov 27 09:27:41 Tower kernel: proc_reg_write+0x88/0xa3 Nov 27 09:27:41 Tower kernel: vfs_write+0xbc/0x129 Nov 27 09:27:41 Tower kernel: ksys_write+0x76/0xc2 Nov 27 09:27:41 Tower kernel: do_syscall_64+0x6b/0x81 Nov 27 09:27:41 Tower kernel: entry_SYSCALL_64_after_hwframe+0x63/0xcd Not sure what this sysfs: cannot create duplicate filename '/devices/virtual/block/md1' error is about but a reboot should fix the current issue.

Please post the diagnostics.

See if you can access it locally, usually due to many different permissions inside appdata you might not be able to access it over SMB.

At least some of issues appear to be caused by free space cache v1, on the console type: btrfs check --clear-space-cache v1 /dev/sdh1 then post new output of btrfs check

If the share is set to cache=yes it will start copying to the array after it hits the share's minimum free space or cache floor, which ever is higher.

Is that from the syslog server and does it cover a crash? It's mostly spammed with modprobe messages.

Also make sure write cache is enabled in the server BIOS, it's disabled by default.

Oct 19 08:26:05 Zhengs-Unraid init: Switching to runlevel: 0 Something is telling the server to shutdown, could be a plugin or a for example a badly functioning power button.

Looks like you were not running in safe mode, do you have the S3 sleep plugin installed? If yes try without it.

Looks like 22:20? Multiple call traces before the crash, but no clue to me of what caused them.

You are issues with the cache2 Samsung disk, check/replace cables then post new diags after array start.

Disk dropped offline and reconnected, check/replace cables and post new diags after array start.

Don't you have the original structure in the recovered drive? If you use the disk share it won't use cache.

Nothing obvious to me in the log, do you know the time when it last happened?

Disk7 dropped offline, check/replace cables and post new diags after array start. You can also then check filesystem on disk5.

Not that I'm aware, it would be wherever you saved it.

Run an extended SMART test on that disk.

Should be, backup current flash drive and you can also run FCP's update advisor.

Scrubbing once a month is generally enough, also take a look below for better pool monitoring: https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=700582

It will work assuming the pool is default RAID1, and you can start with step 5.

MegaRAID models cannot usually be flashed to IT mode.

Please post the diagnostics.

Do you have "Enhanced macOS interoperability" enabled? If yes see if disabling it makes any difference (Settings -> SMB Settings)

Try disabling the bridge for eth1, if it doesn't help try deleting/renaming /boot/config/network.cfg and then reconfigure the settings.

Should be fine for a single disk in terms of performance, there could be the typical Marvell issues, like dropped disks etc, but they do work for some.