0

we have couple DELL physical servers with the same problem about - FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK

from dmesg we can see the following

[2982241.758445]  [<ffffffff81697709>] system_call_fastpath+0x16/0x1b
[2982252.738962] sd 0:0:2:0: [sdc] tag#14 megasas: target reset FAILED!!
[2985405.797192] sd 0:0:2:0: [sdc] tag#102 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
[2985405.797199] sd 0:0:2:0: [sdc] tag#102 CDB: Read(10) 28 00 d7 00 2c e8 00 00 08 00
[2985405.797205] blk_update_request: I/O error, dev sdc, sector 3607112936
[2985405.797214] sd 0:0:2:0: [sdc] tag#104 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
[2985405.797217] sd 0:0:2:0: [sdc] tag#104 CDB: Read(10) 28 00 d7 00 2c f8 00 00 08 00
[2985405.797219] blk_update_request: I/O error, dev sdc, sector 3607112952
[2985405.797477] sd 0:0:2:0: [sdc] tag#97 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
[2985405.797480] sd 0:0:2:0: [sdc] tag#97 CDB: Read(10) 28 00 d7 00 2c b8 00 00 08 00
[2985405.797482] blk_update_request: I/O error, dev sdc, sector 3607112888
[2985405.797493] sd 0:0:2:0: [sdc] tag#103 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
[2985405.797496] sd 0:0:2:0: [sdc] tag#103 CDB: Read(10) 28 00 d7 00 2c f0 00 00 08 00
[2985405.797498] blk_update_request: I/O error, dev sdc, sector 3607112944
[2985405.797508] sd 0:0:2:0: [sdc] tag#96 FAILED Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
[2985405.797511] sd 0:0:2:0: [sdc] tag#96 CDB: Read(10) 28 00 d7 00 2c b0 00 00 08 00
[2985405.797513] blk_update_request: I/O error, dev sdc, sector 3607112880
[3443407.164780] sd 0:0:2:0: task abort: FAILED scmd(ffff881ff7b43100)
[3443433.877426] sd 0:0:2:0: tag#49 megasas: target reset FAILED!

according to redhat the root cause is: ( https://access.redhat.com/solutions/438403 )

• DID_ERROR host status is set for an individual io request within the driver when it detects an otherwise unreported hardware issue.

and the resolutions is:

  1. Engage the storage vendor for assistance.

  2. Please review the systems hardware, switch error counters, etc. to see if there is any indication of where the issue might lie.

  3. The driver is reporting that it is receiving odd, unexpected or invalid information back from storage.

  4. DID_ERROR host status is set for an individual io request within the driver when it detects an otherwise unreported hardware issue. This includes communication and firmware issues within the storage controller, among other hardware based problems.

  5. FC Adapters: Within the fibre channel response frame from storage, conflicting data is present. This indicates an issue within the SAN/storage rather than within RHEL or its driver.

  6. LSI Adapters: scsi io command failed to complete (hung) within the controller. This is different from a scsi failed with returned status or even a clean io timeout issue where the controller was able to abort the io. This typically leads to the loss of the device and follow-on errors of DID_BAD_TARGET which are returned when the device status returned by the HBA is either LD_OFFLINE (logical device present, but not answering commands) or DEVICE_NOT_FOUND (device no longer present within hardware).

so based on above we got lost , because we not sure if we need to replace the HW machine itself , or only the sdc disk or something else

I will; appreciate to get any advice about our case ,

1 Answer 1

1

The sdc disk in your server is dying.

smartctl -a /dev/sdc will let you identify it by its serial number.

Since you might be running RAID, precautions must be taken in order to remove it and install a replacement.

2
  • so you not think that we need to check the machine HW itself ? or maybe replace the machine ?
    – yael
    Commented Jan 26, 2022 at 9:35
  • There's no need to check the machine or replace it. If you have the money to throw at the problem, of course you could replace the entire server. You've got a single disk failing, not anything else. Commented Jan 26, 2022 at 9:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .