Unsolved
This post is more than 5 years old
3 Posts
0
278740
Unexpected sense. SCSI sense data: Sense key: B
Hi,
I am receiving tons of alerts for 2 drives on a PowerEdge server like these:
Unexpected sense. SCSI sense data: Sense key: B Sense code: 4B Sense qualifier: 4, Controller 0, Connector 0, Physical Disk 0:1:13 [Event Code:2095]
They come as Normal severity, but I would like to know if it is an issue or if it is possible to stop them.
Already exported a controller log and all I see is the following:
07/03/14 18:22:23: EVT#455089-07/03/14 18:22:23: 52=Background Initialization progress on VD 0e/d is 24.37%(18258s)
07/03/14 18:22:44: EVT#455090-07/03/14 18:22:44: 113=Unexpected sense: PD 0a(e0x20/s10) Path 5000c50042771491, CDB: 88 00 00 00 00 00 1d 0b 9b 80 00 00 00 80 00 00, Sense: b/4b/04
07/03/14 18:22:44: Raw Sense for PD a: 72 0b 4b 04 00 00 00 20 00 0a 80 00 00 00 00 00 1d 0b 9b a1 03 02 00 00 80 0e 00 00 00 00 00 00 00 00 00 00 00 00 00 00
07/03/14 18:23:00: EVT#455091-07/03/14 18:23:00: 52=Background Initialization progress on VD 0a/9 is 30.59%(18295s)
07/03/14 18:23:12: EVT#455092-07/03/14 18:23:12: 52=Background Initialization progress on VD 0b/a is 29.59%(18307s)
07/03/14 18:23:13: EVT#455093-07/03/14 18:23:13: 113=Unexpected sense: PD 00(e0x20/s0) Path 5000c50042734229, CDB: 88 00 00 00 00 00 92 9e db 80 00 00 00 80 00 00, Sense: b/4b/04
07/03/14 18:23:13: Raw Sense for PD 0: 72 0b 4b 04 00 00 00 20 00 0a 80 00 00 00 00 00 92 9e db cb 03 02 00 00 80 0e 00 00 00 00 00 00 00 00 00 00 00 00 00 00
07/03/14 18:23:28: EVT#455094-07/03/14 18:23:28: 52=Background Initialization progress on VD 05/4 is 28.60%(18323s)
07/03/14 18:23:43: EVT#455095-07/03/14 18:23:43: 52=Background Initialization progress on VD 0d/c is 26.36%(18338s)
07/03/14 18:23:46: EVT#455096-07/03/14 18:23:46: 113=Unexpected sense: PD 02(e0x20/s2) Path 5000c500427b8c6d, CDB: 88 00 00 00 00 01 1d 83 91 80 00 00 00 80 00 00, Sense: b/4b/04
07/03/14 18:23:46: Raw Sense for PD 2: 72 0b 4b 04 00 00 00 20 00 0a 80 00 00 00 00 01 1d 83 91 b7 03 02 00 00 80 0e 00 00 00 00 00 00 00 00 00 00 00 00 00 00
07/03/14 18:23:56: EVT#455097-07/03/14 18:23:56: 52=Background Initialization progress on VD 07/6 is 29.84%(18351s)
07/03/14 18:24:01: EVT#455098-07/03/14 18:24:01: 52=Background Initialization progress on VD 0f/e is 21.63%(18356s)
Could you please advice?
jalarcon
3 Posts
0
August 26th, 2014 19:00
It's a PowerEdge R720 with OS SLES 11(x86_64), the controller is a PERC H710P Mini (Embedded), the disks are original from Dell, and this has been happenning since we replaced a bad hard drive
DELL-Josh Cr
Moderator
Moderator
•
8.6K Posts
0
August 26th, 2014 19:00
Hi jalarcon,
That sense key is for an aborted command, data phase error. It sounds like it might not be reading off the drive well. If you have openmanage server administrator installed you may want to try and run a consistency check and see if that helps.
jalarcon
3 Posts
0
August 26th, 2014 19:00
Hi Josh,
This server is on a production environmet, will the consistency check affect the server functionality?
Thanks
theflash1932
9 Legend
9 Legend
•
16.3K Posts
0
August 26th, 2014 19:00
What system?
What controller?
What OS is installed?
Original Dell disks or after market? If after market, what make/model?
Is the system behaving normally or is there a coinciding issue?
DELL-Josh Cr
Moderator
Moderator
•
8.6K Posts
0
August 27th, 2014 10:00
It depends on the load of the server if the consistency check is noticeable, it is usually about a 30% performance reduction while it is in progress. If it happened after a bad drive, there could be an issue with the array and the consistency check should help. If you need to wait until downtime it shouldn’t be a problem to wait.
papka2
3 Posts
0
September 29th, 2015 10:00
Josh,
Where can i get a list of those sense codes and explanations of what they are? I have many of your machines and receive those Unexpected Sense codes way too much, so i'd like to have that list for myself, so i wouldn't have to bother you guys each time.
Peter
theflash1932
9 Legend
9 Legend
•
16.3K Posts
0
September 29th, 2015 20:00
https://en.wikipedia.org/wiki/Key_Code_Qualifier
Remember that some sense keys are normal, even the "unexpected" ones. You can also address many of them by keeping your system and HDD firmware up to date.
papka2
3 Posts
0
September 30th, 2015 12:00
Thank you, theflash1932.
What can you make out of this log? We have some servers with same/similar setup who give us a lot of Sense 3, Sense Code 11, qualifier 0.
papka2
3 Posts
0
September 30th, 2015 12:00
I've attached log file. It's not looking good at all, while my OMSA health is 100% green. I have multiple boxes like that.
How would you read that log file and what would be your recommendations based on that log file?
Thanks in advance!
1 Attachment
lsi_0930.zip
theflash1932
9 Legend
9 Legend
•
16.3K Posts
0
September 30th, 2015 19:00
From the log you just posted, it could mean you have another bad disk and that data needed to rebuild another disk is unreadable or corrupt. It could also mean that the array was already corrupted and that the logical data has been lost.
It looks to me like you probably have the potential for data loss. I would run diagnostics on the drives, keep the good ones, build new RAID, update all firmware (starting with iDRAC/BIOS, then PERC (driver first), HDD, etc.) and drivers, and restore data to it.
theflash1932
9 Legend
9 Legend
•
16.3K Posts
0
September 30th, 2015 22:00
Sorry, I was going off the snip of log you posted. Looking at the full log you posted, you do have a corrupt array (look for "puncture"). Your disks may be healthy, or you may have a faulty one (could have even been responsible for the corruption). Test the drives, delete array, build anew, restore data.