Two Disabled Drives (Again)

[teedge77]

So, for the second time in a week, I had two drives suddenly become disabled.

Last time, I made the mistake of rebooting and then getting diagnostics. This time, I knew better and have the diagnostics appropriately ready.

I was able to get them going again last time by stopping the arry, disabling them, starting/stopping the array, re-enabling them, and starting the array to rebuild. It took like 28 hours.

Any advice on what to look for and any insight into what you can see wrong is greatly appreciated.

 

Thanks for any help.

zion-diagnostics-20190927-2310.zip

[teedge77]

So, I noticed the following and assume that is the problem, I don't exactly understand why it suddenly can't read it right.

 

Sep 27 17:35:19 ZION kernel: sas: sas_form_port: phy2 belongs to port6 already(1)!
Sep 27 17:35:21 ZION kernel: drivers/scsi/mvsas/mv_sas.c 1435:mvs_I_T_nexus_reset for device[2]:rc= 0
Sep 27 17:35:21 ZION kernel: sas: --- Exit sas_scsi_recover_host: busy: 0 failed: 0 tries: 1
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] Read Capacity(16) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] Sense not available.
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] Read Capacity(10) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] Sense not available.
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] 0 512-byte logical blocks: (0 B/0 B)
Sep 27 17:35:21 ZION kernel: sd 12:0:6:0: [sdv] 4096-byte physical blocks
Sep 27 17:35:21 ZION kernel: sdv: detected capacity change from 8001563222016 to 0
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Read Capacity(16) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Sense not available.
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Read Capacity(10) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Sense not available.
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Write Protect is off
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Mode Sense: 00 00 00 00
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Read Capacity(16) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Sense not available.
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Read Capacity(10) failed: Result: hostbyte=0x04 driverbyte=0x00
Sep 27 17:35:21 ZION kernel: sd 12:0:5:0: [sdu] Sense not available.
Sep 27 17:35:21 ZION kernel: program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO
#

Is this a case of the crappy controller just "losing" drives?

[JorgeB]

Again, SAS2LP is not recommend and likely the source of the problem.

[teedge77]

Again, it's what I have now. "Likely the source of the problem" isn't actually all that helpful.

I've had the same controllers since I started using unraid in 2010. This problem just started. It may be the controllers, it may not.

Do you have a suggestion for a replacement? I would love to replace them with something that works great and can handle more drives. Then, I can move everything off of the unraid box and into boxes of drives.

[JorgeB]

Any LSI with a SAS2008/2308/3008/3408 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, 9400-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

[Zonediver]

On 9/29/2019 at 11:26 AM, johnnie.black said:

Any LSI with a SAS2008/2308/3008/3408 chipset in IT mode, e.g., 9201-8i, 9211-8i, 9207-8i, 9300-8i, 9400-8i, etc and clones, like the Dell H200/H310 and IBM M1015, these latter ones need to be crossflashed.

I would say, lets pin this suggestion on top of "General Support" so everyone can see it - otherwise you will give this information for the next 50 years again and again 😉

Edited October 3, 2019 by Zonediver

[JonathanM]

53 minutes ago, Zonediver said:

I would say, lets pin this suggestion on top of "General Support" so everyone can see it - otherwise you will give this information for the next 50 years again and again 😉

I would say put it in your signature, but with last year's forum update sigs are pretty well screwed, so no joy there.