Problem creating raid1

[Jocko]

and the results

Code: Select all

Setup task makeraid.sh is now running
Found disk: sda(XXXXXXXX)
Found disk: sdb(XXXXXXXX)
Create raid device md0
=>OK: md0 created (raid1)
Format partition md0 with file system ext3:
=>mke2fs 1.43.4 (31-Jan-2017)
Discarding device blocks: done
Creating filesystem with 487556640 4k blocks and 121896960 inodes
Filesystem UUID: 2e087b55-537f-4ca3-9443-f66d07fa345c
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
102400000, 214990848
Allocating group tables: done
Writing inode tables: done
Creating journal (262144 blocks): done
Writing superblocks and filesystem accounting information: done
set label on the data volume successful (RAID1)
=>Formatting md0 succeeded
Update partition type on sda8:fdisk: WARNING: rereading partition table failed, kernel still uses old table: Device or resource busy
OK
=> Fail (unsupported file system)

The error is:
Disk-b not showed until i shutdown nas.

And, as you can see disk-b is missing on raid.

ERROR_RAID-01-01.png

Maybe i should try to update u-boot and see what happen.

Text in red explains why the setup failed. Because the disk seems to be brought down, the script failed to detect sdb8 as a raid member : " Fail (unsupported file system)" when it tried to update sdb partition table.
But this does not explain why sdb brought down. Did you try with another disk ? and when this happens again please to post

Code: Select all

dmesg

it should have there information.

[hirkkak]

Hi

Please to post these outputs

Code: Select all

cat /etc/mdadm.conf
cat /proc/mdstat
mdadm --detail /dev/md0
mdadm --examine /dev/sd[ab]8
fdisk -l /dev/sda
fdisk -l /dev/sdb

The results in the attachement output.txt

Text in red explains why the setup failed. Because the disk seems to be brought down, the script failed to detect sdb8 as a raid member : " Fail (unsupported file system)" when it tried to update sdb partition table.
But this does not explain why sdb brought down. Did you try with another disk ? and when this happens again please to post

Code: Select all

dmesg

it should have there information

The results in the attachement dmesg.txt, but after rebuild raid and reboot :?

I tried with another pair of disks,swaping it, and replacing sda and sdb with another 2 and always same results.


Now the raid are like in the last jpg, "clean completed: [0KB],, estimated time: ". Are this the normal status?

I could try to make the raid with another 2 disk and post dmesg after err.

P.S. I took a look at the scripts, but I would need a lot more time to fully understand them and preferred to try to rebuild the raid manually.

The same password as the default root password.

Lacie 2 Big Network 2.7z

[Jocko]

Now the raid are like in the last jpg, "clean completed: [0KB],, estimated time: ". Are this the normal status?

No it is not the expected message. This happens after an unclean shutdown or after removing and re-adding a disk. mdadm creates a bitmap (a sort of write cache during synchronisation job) but does not remove it after completing the job... Then you have an additional line in mdstat file ( bitmap: 0/15 pages [0KB], 65536KB chunk).
To remove it do

Code: Select all

mdadm --grow --bitmap=none /dev/md0

As you tried with other disks I think you have a hardware issue on the sata port for sdb according with dmesg output :dry Too such lines like

Code: Select all

[149142.926109] ata2.00: failed to read SCR 1 (Emask=0x40)
[149142.931341] ata2.01: failed to read SCR 1 (Emask=0x40)
[149142.936552] ata2.15: exception Emask 0x10 SAct 0x0 SErr 0x180000 action 0x6 frozen
[149142.944187] ata2.15: edma_err_cause=00000020 pp_flags=00000000, SError=00180000
[149142.951568] ata2.15: SError: { 10B8B Dispar }
[149142.955999] ata2.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[149142.963294] ata2.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[149142.970577] ata2.15: hard resetting link

AS you use your nas with a raid1, it is undetectable for the user and I think sdb8 may be not not really a perfect mirror of sda8
This may happen if the sata cable is damaged or a power issue. Do you have an alternative power supply ?

[hirkkak]

I've been looking for a more powerful PSU but I haven't found any.
I replaced with 12V3A. (Original was 12v4A) It seems to work, at least for now.

I destroyed the raid without problems, when i tryied to make the raid again, same error.

Setup task makeraid.sh is now running
Found disk: sda(XXXXXXXX)
Found disk: sdb(XXXXXXXX)
Create raid device md0
=>OK: md0 created (raid1)
Format partition md0 with file system ext4:
=>mke2fs 1.43.4 (31-Jan-2017)
Discarding device blocks: done
Creating filesystem with 487556640 4k blocks and 121896960 inodes
Filesystem UUID: 85c20366-92a9-4a92-951b-c54cf8be04c2
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
102400000, 214990848
Allocating group tables: done
Writing inode tables: done
Creating journal (262144 blocks): done
Writing superblocks and filesystem accounting information: done
set label on the data volume successful (Raid1)
=>Formatting md0 succeeded
Update partition type on sda8:fdisk: WARNING: rereading partition table failed, kernel still uses old table: Device or resource busy
OK
=> Fail (unsupported file system)

Setup task makeraid.sh is restarting
Found disk: sda(ZFL3YG6F)
Disk not found (sdb)

fdisk -l not detecting sdb

root@fvdwsl-base:/ # fdisk -l

Disk /dev/sda: 2000.3 GB, 2000398934016 bytes
255 heads, 63 sectors/track, 243201 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/sda1 9 72 514080 83 Linux
/dev/sda2 73 136 514080 83 Linux
/dev/sda3 137 200 514080 82 Linux swap
/dev/sda4 201 243200 1951897500 5 Extended
/dev/sda5 209 312 835380 83 Linux
/dev/sda6 321 328 64260 83 Linux
/dev/sda7 337 400 514080 83 Linux
/dev/sda8 409 243200 1950226740 fd Linux raid autodetect

Disk /dev/md0: 1997.0 GB, 1997031997440 bytes
2 heads, 4 sectors/track, 487556640 cylinders
Units = cylinders of 8 * 512 = 4096 bytes

Disk /dev/md0 doesn't contain a valid partition table

This is dmesg results before reboot. Looks similar.

dmesg_2024_04_24-10_05.7z

This evening i will replace sata cables before trying again.

PS. FYI.
The disks are barracuda compute. (ST2000DM008)
https://www.seagate.com/es/es/products/hard-drives/barracuda-hard-drive/
The other pair are Seagate Constellation. ST32000644.

Both should be compatible.

[hirkkak]

I tried with 10 amp psu and same problem.

Checking the board it may be a problem with the capacitors. There are some that don't seem right.
I'll try to replace them and see if that fixes it.

Any other thing that i could try before?

Cap1.jpg

Cap2.jpg

Cap3.jpg

[Jocko]

Hi

I am not sure there is an issue with capacitors. According to your pics they seem ok

Before replacing some capacitors you can check the sata cable of sdb : swap it with sda but keep the ata port for the disk and see if on dmesg output the disk error are still or not on the same ata port (ata2).

[hirkkak]

Hi,

I am not sure there is an issue with capacitors. According to your pics they seem ok

They are actually a little bulky but not excessively.
I don't know what precision these devices need, but they are not at 100% . Maybe one or two only.

Before replacing some capacitors you can check the sata cable of sdb : swap it with sda but keep the ata port for the disk and see if on dmesg output the disk error are still or not on the same ata port (ata2).

I swapped the disks between bays, it should be the same, isn´t?
I tried using the cables from a 4big, modding connectors, but there was no way to detect disks.
Anyway, it would be the first time I've seen a sata cable fail.

I tryed with another disk (another barracuda compute) but 5400RPM. It finished to make the raid suscesfully, but stays at:

Code: Select all

clean completed: [8KB],, estimated time:

This time the disk was not lost after finish, no needed to shutdown to detect it.

dmesg keeps showing several like this.

Code: Select all

[15241.184642] ata2.00: failed to read SCR 1 (Emask=0x40)
[15241.189879] ata2.01: failed to read SCR 1 (Emask=0x40)
[15241.195008] ata2.15: exception Emask 0x10 SAct 0x0 SErr 0x180000 action 0x6 frozen
[15241.202581] ata2.15: edma_err_cause=00000020 pp_flags=00000000, SError=00180000
[15241.209877] ata2.15: SError: { 10B8B Dispar }
[15241.214220] ata2.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[15241.221430] ata2.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[15241.228637] ata2.15: hard resetting link
[15242.648098] ata2.15: SATA link up 3.0 Gbps (SStatus 123 SControl F300)
[15242.655524] ata2.00: hard resetting link
[15243.018864] ata2.01: hard resetting link
[15243.768126] ata2.00: configured for UDMA/133


i don´t understand last line, (should be sata 3 or 6?)

[Jocko]

Hi

I swapped the disks between bays, it should be the same, isn´t?

Not really as the former sda (which are now sdb) still use the current cable of the ata2 port you can not valid if the cable is ok if you have the same behaviour. Be careful, according to a previous post the partition table of the disks are not the same : on the former sdb you have only 5 partition (4 to 8) which does not allow to use it as a system disk in the left slot.

I tryed with another disk (another barracuda compute) but 5400RPM. It finished to make the raid suscesfully, but stays at:

Code: Select all

clean completed: [8KB],, estimated time:

So that would mean there were again some issues when it created the raid and then need to set a bitmap to complete the job.

i don´t understand last line, (should be sata 3 or 6?)

Lacie NAS are old products and then their board does not support such modes for ata port. Here the dmesg outpout you should have when the OS detects an ata port. In this output it is a device with 4 ports where 2 disks are present (bring up the device, reset the ports, detect the disk and set ata mode). Later you should not see other information about the ata device if there is no issue.

Code: Select all

[   12.424786] ata2: SATA link up 3.0 Gbps (SStatus 123 SControl F300)
[   12.431742] ata2.15: Port Multiplier 1.2, 0x11ab:0x4140 r193, 4 ports, feat 0x5/0x1f
[   12.446847] ata2.00: hard resetting link
[   12.785555] ata2.01: hard resetting link
[   13.295042] ata2.02: hard resetting link
[   13.805044] ata2.03: hard resetting link
[   14.164821] ata2.01: ATA-10: WDC WD30EFRX-68N32N0, 82.00A82, max UDMA/133
[   14.171581] ata2.01: 5860533168 sectors, multi 16: LBA48 NCQ (depth 31/32)
[   14.194835] ata2.01: configured for UDMA/133
[   14.204998] ata2.02: ATA-9: WDC WD30EFRX-68EUZN0, 82.00A82, max UDMA/133
[   14.211670] ata2.02: 5860533168 sectors, multi 0: LBA48 NCQ (depth 31/32)
[   14.225046] ata2.02: configured for UDMA/133
[   14.229535] ata2: EH complete


[hirkkak]

Hi i´m still fighting vs that nas :fists

Replacing cables, disks. But always:

Code: Select all

ata2.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen

As you told, it must be a hardware issue.

If no other tips, i gonna try to replace the worst capacitors and see if helps.
I will start with the closest one to sata2 power port (Cap2.jpg) and we will see.

There is no chance that a firmware update or something like that can fix it?
Can you send me the old firmware to make a try before.

UIMAGE-414133-KIRKWOOD-1

No way to find on forum for download.

[hirkkak]

I replaced all bad capacitors but two, i don´t have any more. All of them are rated 470uf but only show 50uf-100uf on tester. So they must was broken.

The problem is still there. No way to create raid 1. I tried with raid 0 and maked without problems. But still showing same errors on dmesg.

Raid0.png

Code: Select all

root@fvdwsl-base:/ # dmesg | grep -E "error|exception|failed"
[   30.337599] ata2.00: failed to read SCR 1 (Emask=0x40)
[   30.342724] ata2.01: failed to read SCR 1 (Emask=0x40)
[   30.351395] ata2.15: exception Emask 0x10 SAct 0x0 SErr 0x180000 action 0x6 frozen
[   30.371132] ata2.00: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[   30.378508] ata2.01: exception Emask 0x100 SAct 0x0 SErr 0x0 action 0x6 frozen
[   42.775013] synth uevent: /gpiochip0: failed to send uevent
[   42.775032] gpio gpiochip0: uevent: failed to send synthetic uevent
[   42.868722] synth uevent: /gpiochip1: failed to send uevent
[   42.868741] gpio gpiochip1: uevent: failed to send synthetic uevent
[  106.541417] ata2.00: failed to read SCR 1 (Emask=0x40)
[  106.546567] ata2.01: failed to read SCR 1 (Emask=0x40)


Finally i got "UIMAGE-414133-KIRKWOOD-1" and make the test with that with same bad results.

Any other tip?