Ext USB disk plugged out/in repeadtely

[mdi]

It seem i cannot see it from the interface, also following your suggestions. I aldo tried to repartition the disk on another pc, without formatting, but it was no better.

Could it be that it is a GPT disk causing the issue?

[Jocko]

Ok

we go to force the web-interface to display the disk in the disk setup menu:
- check environment

Code: Select all

cat /proc/partitions

You should see your disk and its partition (I assume its kernel name is sdh1)
- check file system on it

Code: Select all

file -bs /dev/sdh1

it must be either ext3 or ext4 or vfat. If not you need to format the partition
- check if advanced usb mount mode is enabled

Code: Select all

grep mountmode /etc/nas_conf_db.xml

it should return 'internal' if not go to the usb disk setup menu and enable the advanced mode
- try to mount manually the partition

Code: Select all

/etc/api/nasapi.php mrvlUdevdAttached sdh1

- check if it is mounted

Code: Select all

mount

you should see a line like

Code: Select all

/dev/sdh1 on /share/200? type ...

If it is ok you should see now the disk in the disk table of the disk setup menu, then click on its related link 'reset'

[mdi]

From your comments I noticed I didn't switch the mounting mode ernal/internal for the disks.

Now i can see both. I tried to initialize them and i received this:

Code: Select all

Setup task reset.sh is now running
Found disk: sdd(DC42893059FF-0:0_WDC_WD20_EFRX-68EUZN0)
Found disk: sdf(DC42893059FF-0:1_WDC_WD20_EFRX-68EUZN0)
Delete the partiton table on sdd: OK
Create a new partiton ms-dos table on sdd:
=>Fail (Disk not found)

Setup task reset.sh is restarting
Disk not found (sdd)


The second disk was initialized correctly.
Still chrcking...

[mdi]

I've been able to reset bot disks with the web interface. Then i tried to run a backup to the first disk.

I hot again a disconnection and a reconnection of the disks.

I tried to unmount the 2nd disk (from the web interface, not physically) and run the backup.

Same issue and both disks were reconnected at the end, automatically.

Maybe it's again an issue with the docking station?
Or maybe the diak is failing. I see in dmesg may errors related to inode 37749211.

[Jocko]

Hi mdi,

Or maybe the diak is failing. I see in dmesg may errors related to inode 37749211.

No this issue occurs only because your disk plugged out badly and then it corrupted the fs on the mounted partition.

Maybe it's again an issue with the docking station?

Maybe we can easily validate or not this hypothesis if you have another usb sata interface.
If you have it you can try to use it disk by disk and see if the single disk has the same behaviour or not.

To add;
In case it connects and is recognized and later it spontanuos disconnects it is not a software issue to my opinion, it could point more likely to a bad connection, either a broken cable or connector with bad contact. Try another cable if it can be exchanged. Also power supply of the usb box could be unstable.

up to now you make no feedback about fvdw suggestions.
Do you check the cable, power supply ?
Do you try the docking station with only one disk (try with a single disk one by one)?

[mdi]

Hi Jocko, yes i tried woth only 1 disk connected, the power supply is brand new, provided with the docking station and everything is connected to a UPS device.

The USB cable seems also ok, but i didn't try a different one.

I can try a sata cable anyway. I will feedback about my tests with this.

For the moment it sems that the disks do not disconnect if they are just initialized, but as i try using them, after some time they give i/o error, disconnect and reconnect.

If the filesystem gets corrupted is because of the automatic disconnection.

[mdi]

One thing I noticed is that, after safe-disconnecting both disks from the web-interface, they have been re-mounted again automatically after some hours.

I then turned off the docking station completely, as I don't want to be bothered by the continuous device disconnected/connected emails.

I will let you know once I do another test with 1 disk connected via a sata-usb cable.

Cheers :idea:

[mdi]

I attached a normal usb-sata cable to the 1st disk and run a backup of a big share. It is still copying files since yesterday, but it seems stable.
The strange thing is that while copying the disk appears as disconnected and it doesn't show under the external disks list. Is this normal?

On the contrary, I've attached the docking station with the second disk to a Raspberry available on my network, set it on the NAS as a remote share and i tried a backup of a nas share to that remote share.
There was a disconnection.

As the usb-sata cable seems reliable, i will try this also with the 2nd disk, but I'm rather convinced that the docking station is not reliable at all.

My initial idea was to use the docking station to backup my nas on one disk, then to copy that disk on the second one to have 2 copies.
I use this solution with my pc at work (without a nas, i have a second pc indtead), but with the nas i tried 2 different docking stations and I wasn't lucky so far.

[mdi]

I'm back from work and dmesg says:

Code: Select all

CIFS VFS: No writable handles for
 inode


I rebooted the NAS, the disk was remounted and i re-run the backup. At the end ingot the following dmesg:

Code: Select all

...
[   22.164248] sd 2:0:0:0: [sdb] 3907029168 512-b
yte logical blocks: (2.00 TB/1.82 TiB)
[   22.173183] sd 2:0:0:0: [sdb] Write Protect is
 off
[   22.177970] sd 2:0:0:0: [sdb] Mode Sense: 03 0
0 00 00
[   22.178802] sd 2:0:0:0: [sdb] No Caching mode
page found
[   22.184173] sd 2:0:0:0: [sdb] Assuming drive c
ache: write through
[   22.203370]  sdb: sdb1
[   22.209183] sd 2:0:0:0: [sdb] Attached SCSI di
sk
[   30.126829] udevd (957): /proc/957/oom_adj is
deprecated, please use /proc/957/oom_score_adj in
stead.
[   30.136232] udev: starting version 130
[   34.722995] EXT4-fs (sdb1): recovery complete
[   34.728251] EXT4-fs (sdb1): mounted filesystem
 with ordered data mode. Opts: user_xattr,acl,nod
elalloc
[   42.201977] EXT4-fs (sda8): mounting ext3 file
 system using the ext4 subsystem
[   84.412669] EXT4-fs (sda8): warning: maximal m
ount count reached, running e2fsck is recommended
[   84.563755] EXT4-fs (sda8): recovery complete
[   84.621554] EXT4-fs (sda8): mounted filesystem
 with ordered data mode. Opts: user_xattr,acl
[   87.403168] mv643xx_eth_port mv643xx_eth_port.
0 eth0: link down
[   87.409328] IPv6: ADDRCONF(NETDEV_UP): eth0: l
ink is not ready
[   90.201157] mv643xx_eth_port mv643xx_eth_port.
0 eth0: link up, 1000 Mb/s, full duplex, flow con
trol disabled
[   90.211031] IPv6: ADDRCONF(NETDEV_CHANGE): eth
0: link becomes ready
[  100.444426] capability: warning: `proftpd' use
s 32-bit capabilities (legacy support in use)
[  396.831361] nf_conntrack: automatic helper ass
ignment is deprecated and it will be removed soon
. Use the iptables CT target to attach helpers in
stead.
[  510.991555] usb 1-1: reset high-speed USB devi
ce number 2 using orion-ehci
[  541.961809] usb 1-1: USB disconnect, device nu
mber 2
[  541.967944] sd 2:0:0:0: Device offlined - not
ready after error recovery
[  541.991649] sd 2:0:0:0: [sdb] tag#0 UNKNOWN(0x
2003) Result: hostbyte=0x01 driverbyte=0x00
[  541.999829] sd 2:0:0:0: [sdb] tag#0 CDB: opcod
e=0x2a 2a 00 48 70 3f 88 00 00 f0 00
[  542.007393] blk_update_request: I/O error, dev
 sdb, sector 1215315848
[  542.013882] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916582)
[  542.026610] Buffer I/O error on device sdb1, l
ogical block 151900516
[  542.032954] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916583)
[  542.045673] Buffer I/O error on device sdb1, l
ogical block 151900517
[  542.052008] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916584)
[  542.064733] Buffer I/O error on device sdb1, l
ogical block 151900518
[  542.071067] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916585)
[  542.083793] Buffer I/O error on device sdb1, l
ogical block 151900519
[  542.090128] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916586)
[  542.102853] Buffer I/O error on device sdb1, l
ogical block 151900520
[  542.109198] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916587)
[  542.121922] Buffer I/O error on device sdb1, l
ogical block 151900521
[  542.128257] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916588)
[  542.140982] Buffer I/O error on device sdb1, l
ogical block 151900522
[  542.147316] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916589)
[  542.160041] Buffer I/O error on device sdb1, l
ogical block 151900523
[  542.166376] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916590)
[  542.179100] Buffer I/O error on device sdb1, l
ogical block 151900524
[  542.185436] EXT4-fs warning (device sdb1): ext
4_end_bio:315: I/O error -5 writing to inode 3774
9903 (offset 0 size 0 starting block 151916591)
[  542.198161] Buffer I/O error on device sdb1, l
ogical block 151900525
[  542.216239] sd 2:0:0:0: [sdb] tag#0 UNKNOWN(0x
2003) Result: hostbyte=0x01 driverbyte=0x00
[  542.224449] sd 2:0:0:0: [sdb] tag#0 CDB: opcod
e=0x2a 2a 00 48 70 40 78 00 00 b0 00
[  542.232006] blk_update_request: I/O error, dev
 sdb, sector 1215316088
[  542.262533] JBD2: Detected IO errors while flu
shing file data on sdb1-8
[  542.269163] Aborting journal on device sdb1-8.
[  542.310322] JBD2: Error -5 detected when updat
ing journal superblock for sdb1-8.
[  542.325205] EXT4-fs error (device sdb1): ext4_
journal_check_start:56: Detected aborted journal
[  542.325214] EXT4-fs (sdb1): Remounting filesys
tem read-only
[  542.325223] EXT4-fs (sdb1): previous I/O error
 to superblock detected
[  542.347332] JBD2: Detected IO errors while flu
shing file data on sdb1-8
[  552.564747] EXT4-fs error (device sdb1): ext4_
put_super:837: Couldn't clean up the journal
root@nas:/ #


[Jocko]

Hi mdi

I attached a normal usb-sata cable to the 1st disk and run a backup of a big share. It is still copying files since yesterday, but it seems stable.
The strange thing is that while copying the disk appears as disconnected and it doesn't show under the external disks list. Is this normal?

No it is not normal.
If the disk is unmounted then that would mean the backup stores the synchronized files on the mount point and not in the disk !

With your last tries, I think more and more the issue is on the disk.

I think a better test would be keep only one disk (1st disk) as according to your dmesg output it is only sdc1 which returns a fatal error and occurs an re-mounting on both disks.

So do the same test with your 2d disk