unraid says my flash died (while running) while backing up data off my shares NOT my flash config

[cdoublejj]

server is still on but, the pages on the web gui are blank it says flash corrupted and failed. ain't no chance i can copy the config is there?

 

I DO BELIEVE that with no config setting up a new flash drive will loose all my share in of and all the data on my drives.

 

EDIT: oh yeah, the parity has been having some errors and one of the drives died thankfully that dead drive didn't have any important data on it.

FYI, i'm posting in case there is any chance i can save my config before i power down my server.

like many of my newer flash drives i believe history dictates that it probably died. (i'm flashing to a satadom or SSD this time forget flash drives)

EDIT: bash/command line button on webgui seems to work.

EDIT: SMB shows the flash folder but windows or ..unraid... keeps trying to auth against AD and i never succeeded at binding. i think windows login is being wierd

EDIT:Version: 6.10.0-rc2 

Edited April 6, 2022 by cdoublejj

[cdoublejj]

running ls in boot folder brings nothing

 

var/logs is there and so i see file names when looking in disks under mnt

 

EDIT: can i pull the flash drive out of the running server to see how bad it's fried? 

EDIT: no dockers or any of that who-ha, just data shares

EDIT: i know that my parity drive is the only drive that is 8tb the rest are data.

 

Apr 5 10:16:30 unraid1 avahi-daemon[28440]: Leaving mDNS multicast group on interface lo.IPv6 with address ::1.
Apr 5 10:16:30 unraid1 avahi-daemon[28440]: Leaving mDNS multicast group on interface lo.IPv4 with address 127.0.0.1.
Apr 5 10:16:30 unraid1 avahi-daemon[28440]: avahi-daemon 0.8 exiting.
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4042 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=2s
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4042 Sense Key : 0xb [current]
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4042 ASC=0x44 ASCQ=0x0
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4042 CDB: opcode=0x88 88 00 00 00 00 00 ba fc 68 b8 00 00 00 08 00 00
Apr 5 10:16:30 unraid1 kernel: blk_update_request: I/O error, dev sde, sector 3137104056 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Apr 5 10:16:30 unraid1 kernel: md: disk0 read error, sector=3137102008
Apr 5 10:16:30 unraid1 root: Starting Avahi mDNS/DNS-SD Daemon: /usr/sbin/avahi-daemon -D
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Found user 'avahi' (UID 61) and group 'avahi' (GID 214).
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Successfully dropped root privileges.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: avahi-daemon 0.8 starting up.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Successfully called chroot().
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Successfully dropped remaining capabilities.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Loading service file /services/sftp-ssh.service.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Loading service file /services/smb.service.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Loading service file /services/ssh.service.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Joining mDNS multicast group on interface br0.IPv4 with address 192.168.0.240.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: New relevant interface br0.IPv4 for mDNS.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Joining mDNS multicast group on interface lo.IPv6 with address ::1.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: New relevant interface lo.IPv6 for mDNS.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Joining mDNS multicast group on interface lo.IPv4 with address 127.0.0.1.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: New relevant interface lo.IPv4 for mDNS.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Network interface enumeration completed.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Registering new address record for 192.168.0.240 on br0.IPv4.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Registering new address record for ::1 on lo.*.
Apr 5 10:16:30 unraid1 avahi-daemon[28646]: Registering new address record for 127.0.0.1 on lo.IPv4.
Apr 5 10:16:30 unraid1 emhttpd: shcmd (793855): /etc/rc.d/rc.avahidnsconfd restart
Apr 5 10:16:30 unraid1 root: Stopping Avahi mDNS/DNS-SD DNS Server Configuration Daemon: stopped
Apr 5 10:16:30 unraid1 root: Starting Avahi mDNS/DNS-SD DNS Server Configuration Daemon: /usr/sbin/avahi-dnsconfd -D
Apr 5 10:16:30 unraid1 avahi-dnsconfd[28655]: Successfully connected to Avahi daemon.
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4041 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=2s
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4041 Sense Key : 0xb [current]
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4041 ASC=0x44 ASCQ=0x0
Apr 5 10:16:30 unraid1 kernel: sd 9:0:3:0: [sde] tag#4041 CDB: opcode=0x88 88 00 00 00 00 00 be 5d 34 f0 00 00 00 08 00 00
Apr 5 10:16:30 unraid1 kernel: blk_update_request: I/O error, dev sde, sector 3193779440 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0
Apr 5 10:16:30 unraid1 kernel: md: disk0 read error, sector=3193777392
Apr 5 10:16:31 unraid1 avahi-daemon[28646]: Server startup complete. Host name is unraid1.local. Local service cookie is 2525190548.

 

Edited April 6, 2022 by cdoublejj

[MAM59]

Welcome to the Club 😞

 

Looks like you Flash Drive is really broken, I know this "/boot is empty" thing too well.

 

Get a new Flash, try to get an old USB-2-only one with up to 16Gb (the limetech tool for flashing it has a flaw if used on sticks > 16Gb, this can lead to this stick corruption). This is not easy, UNRAID is picky what kind of sticks it likes to be written on. I now have 12 here from my buyings a few weeks ago that were refused.

 

USB2 is slow, but better because those sticks do not get that hot. The modern ones are fast but not build for longer operations. The heat will distroy them sooner or later.

 

I hope, you have a backup of your old stick. If not, you will have to rebuild the config manually. As long as you know which one was the parity drive this should not be that big problem.

 

[trurl]

I see you "liked" a number of posts on other threads that are relevant to your problems. Did they give you enough information to get going again?

 

The reason /boot was empty is because the flash drive was no longer mounted.

 

Looks like you may have problems with parity disk.

[cdoublejj]

9 hours ago, Michael Meiszl said:

Welcome to the Club 😞

 

Looks like you Flash Drive is really broken, I know this "/boot is empty" thing too well.

 

Get a new Flash, try to get an old USB-2-only one with up to 16Gb (the limetech tool for flashing it has a flaw if used on sticks > 16Gb, this can lead to this stick corruption). This is not easy, UNRAID is picky what kind of sticks it likes to be written on. I now have 12 here from my buyings a few weeks ago that were refused.

 

USB2 is slow, but better because those sticks do not get that hot. The modern ones are fast but not build for longer operations. The heat will distroy them sooner or later.

 

I hope, you have a backup of your old stick. If not, you will have to rebuild the config manually. As long as you know which one was the parity drive this should not be that big problem.

 

guess i just need to find a 16gb sata SSD and put it in a sata to USB enclosure. i'm done with flash. every flash drive in my ESXi host bit the dust, one by one.

 

1 hour ago, trurl said:

I see you "liked" a number of posts on other threads that are relevant to your problems. Did they give you enough information to get going again?

 

The reason /boot was empty is because the flash drive was no longer mounted.

 

Looks like you may have problems with parity disk.

i got some loose ideas. it sounds like i can image a new flash device and i can remount my existing drives with data? something about boot up will auto start array and wipe the data? don't mount the parity drive?

[MAM59]

14 minutes ago, cdoublejj said:

guess i just need to find a 16gb sata SSD and put it in a sata to USB enclosure. i'm done with flash. every flash drive in my ESXi host bit the dust, one by one.

 

I'm afraid, this won't work. The UNRAID Developer use the numbering scheme of USB Sticks for licence protection. Thats why they only accept certain devices. And surely a SATA SSD is not among them.

14 minutes ago, cdoublejj said:

i got some loose ideas. it sounds like i can image a new flash device and i can remount my existing drives with data? something about boot up will auto start array and wipe the data? don't mount the parity drive?

dont worry, nothing will be wiped unless you urge it to do so.

The only real problem is to recreate the array config and chosing the correct parity drive like before

The "shares" are still there (every folder found on a data drive is automatically considered to be a share). Whats lost are the users and the share permissions if you dont have a backup.

if you really dont know which one was the parity drive before, create a new stick and start with NO disks.

(Hint: it is very helpful to install the plugin "unassigned devices" before because this allows you to see more infos of the not-yet-included drives and even mount them manually)

 

Then just insert ONE, it will be found and UNRAID will tell you if it contains a file system or not. If YES: add it to the array!

If NO it is most likely the old parity drive (cant be recognized, has no valid filesystem). Skip it for now.

Step through all disks until you have (re)found all your data disks.

Create Users and assign Permissions to the shares. Then you are back in business again already.

At the very end, stop the array, assign the left out drive for parity (but really ONLY IF YOU ARE SURE NOW THAT IT IS THE FORMER PARITY DRIVE!!!!) and restart the array.

This drive now really will be wiped out and filled with new parity data.

But this is the only automatic - wiper in the game!

 

Edited April 6, 2022 by Michael Meiszl

[trurl]

12 hours ago, cdoublejj said:

EDIT: can i pull the flash drive out of the running server to see how bad it's fried? 

EDIT: no dockers or any of that who-ha, just data shares

EDIT: i know that my parity drive is the only drive that is 8tb the rest are data.

Shutdown, put flash in PC, let it checkdisk. While there make a backup. The config folder from your flash backup is all you need to get going again with your configuration on a new install to the same (preferred so you won't have to transfer license) or different (if necessary) flash drive.

[cdoublejj]

a checkdisk is only possible if the flash drive powers up and is seen the by the PC. i'll try that first. HOWEVER i have been told i plug my data drives in to a linux PC and pull my data off my drives since they are XFS? This would impliy i can simply get my data THEN blow the whole thing away like a gun fight in an 80s movies and start over as i please!

[trurl]

If you mean to start over with Unraid there is no need to blow anything away. Even if you have to start from scratch with a new install on a new flash, Unraid will recognize the disks with their contents (and shares) when you assign them and start the array.

[trurl]

5 hours ago, trurl said:

I see you "liked" a number of posts on other threads that are relevant to your problems. Did they give you enough information to get going again?

 

4 hours ago, cdoublejj said:

i got some loose ideas. it sounds like i can image a new flash device and i can remount my existing drives with data? something about boot up will auto start array and wipe the data? don't mount the parity drive?

 

Probably best to just proceed with advice given in this thread, starting with

3 hours ago, trurl said:

Shutdown, put flash in PC, let it checkdisk. While there make a backup.

Report back on how that goes and we can decide what to do next.

[cdoublejj]

all my stuff was there minus the passwords and permissions levels, all my shares even showed up as public shares, good enough for the girls i go out with but, month or two later, aka tonight i noticed this: 

 

[trurl]

7 hours ago, cdoublejj said:

all my shares even showed up as public shares

Shares are simply the combined top level folders on array and pools, so shares exist whether or not you have created them again or made any settings for them. They are public because that is the default setting. The other settings for each of your shares is also defaults unless you change them.