"Docker Service failed to start" after a few hours of running

[kevin182]

After this, I run diagnostics, use the reboot command and wait over an hour. Eventually, I decided I need to hard reboot. Comes up for a while then repeats. Please help, thank you.

 

lucifer-diagnostics-20230910-1952.zip

[JorgeB]

Sep 10 11:11:30 Lucifer emhttpd: ST5000DM000_W4J1JZXZ (sdj) 512 8089950

This disk is spamming the log with errors, if it's not being used disconnect it, and post new diags after the issue.

[kevin182]

OK. Thank you. I wish there was some way of logically disconnecting the disk. I was hoping to keep it in there to destroy the contents before I threw it out. I will let you know if it happens again. If it does, it will be in the 24 hrs.

[JorgeB]

5 minutes ago, kevin182 said:

I wish there was some way of logically disconnecting the disk.

You can do 

echo 1 > /sys/block/sdX/device/delete

It will be back after a reboot

[kevin182]

I removed the disk but unfortunately it happened again. Thank you.

 

lucifer-diagnostics-20230911-1114.zip

[JorgeB]

Sep 11 10:39:43 Lucifer kernel: SQUASHFS error: xz decompression failed, data probably corrupt
Sep 11 10:39:43 Lucifer kernel: SQUASHFS error: Failed to read block 0xf6bdc8: -5
Sep 11 10:39:43 Lucifer kernel: SQUASHFS error: xz decompression failed, data probably corrupt
Sep 11 10:39:43 Lucifer kernel: SQUASHFS error: Failed to read block 0xf6bdc8: -5
Sep 11 10:39:43 Lucifer kernel: SQUASHFS error: xz decompression failed, data probably corrupt

 

These were also in the previous log, and AFAIK then can only be form a flash drive problem (possibly RAM), try reformatting the flash drive, or using a different one.

[kevin182]

Error: {"message":"Request failed with status code 403","data":{"error":"Cannot transfer GUID because you've already transferred 2 times in the past year. Please contact support for assistance."}}
I will contact support

[kevin182]

Got a new license key and I believe the flash drive was the problem as the old 1 did have errors on it. Thank you.

[kevin182]

Unfortunately, this has appeared again. Please let me know if you have any insights into what is causing this. Thank you.lucifer-diagnostics-20230921-0741.zip

[JorgeB]

Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 Sense Key : 0x3 [current] 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 ASC=0x11 ASCQ=0x0 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x28 28 00 00 04 25 78 00 01 00 00
Sep 21 02:01:22 Lucifer kernel: critical medium error, dev sda, sector 271736 op 0x0:(READ) flags 0x80700 phys_seg 32 prio class 2

 

This time we can see it really is a flash drive problem.

[kevin182]

Also, when I try to stop the array I continuously get these errors.

[JorgeB]

Post the complete diags please.

[kevin182]

1 hour ago, JorgeB said:

Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 Sense Key : 0x3 [current] 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 ASC=0x11 ASCQ=0x0 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x28 28 00 00 04 25 78 00 01 00 00
Sep 21 02:01:22 Lucifer kernel: critical medium error, dev sda, sector 271736 op 0x0:(READ) flags 0x80700 phys_seg 32 prio class 2

 

This time we can see it really is a flash drive problem.

 

6 minutes ago, JorgeB said:

Post the complete diags please.

That is the complete diagnostics. Albeit just before trying to shut it down.

[kevin182]

1 hour ago, JorgeB said:

Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 Sense Key : 0x3 [current] 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 ASC=0x11 ASCQ=0x0 
Sep 21 02:01:22 Lucifer kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x28 28 00 00 04 25 78 00 01 00 00
Sep 21 02:01:22 Lucifer kernel: critical medium error, dev sda, sector 271736 op 0x0:(READ) flags 0x80700 phys_seg 32 prio class 2

 

This time we can see it really is a flash drive problem.

If that's the case, I have had 2 flash drives go bad in less than 2 weeks. Seems quite unlikely this is my only problem and considering I purchased a very good quality flash drive for this replacement.

[JorgeB]

1 hour ago, kevin182 said:

That is the complete diagnostics.

The zip.

[Wolfy]

I'm having this issue as well. After about a month I had my second flash drive fail. I believe the issue is my syslog was writing to the flash drive. Hopefully now that I fixed it, once I get (another) new flash drive in it'll last a while.