2017-04-29 03:31 UTC

View Issue Details Jump to Notes ]
IDProjectCategoryView StatusLast Update
0005880CentOS-6kernelpublic2014-08-26 06:01
Reportersles 
PrioritynormalSeverityminorReproducibilityalways
StatusnewResolutionopen 
Platformamd64OScentosOS Version6.3
Product Version6.3 
Target VersionFixed in Version 
Summary0005880: SATA error on SIL 3114 and SSD
DescriptionHello!

Installed intel ssd INTEL SSDSC2CT060A3, 300i on old server with


03:05.0 Mass storage controller: Silicon Image, Inc. SiI 3114 [SATALink/SATARaid] Serial ATA Controller (rev 02)

in non-raid mode


and got many following errors


ata6.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata6.00: BMDMA2 stat 0x50001
ata6.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
         res 51/04:01:00:00:00/00:00:00:00:00/b0 Emask 0x1 (device error)
ata6.00: status: { DRDY ERR }
ata6.00: error: { ABRT }
ata6.00: device reported invalid CHS sector 0
sd 5:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 5:0:0:0: [sda] Sense Key : Aborted Command [current] [descriptor]
Descriptor sense data with sense descriptors (in hex):
        72 0b 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00
        00 00 00 00
sd 5:0:0:0: [sda] Add. Sense: No additional sense information
sd 5:0:0:0: [sda] CDB: Write same(16): 93 08 00 00 00 00 05 0b d8 00 00 00 00 08 00 00
ata6.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata6.00: BMDMA2 stat 0x50001
ata6.00: cmd 06/01:01:00:00:00/00:00:00:00:00/a0 tag 0 dma 512 out
         res 51/04:01:00:00:00/00:00:00:00:00/b0 Emask 0x1 (device error)
ata6.00: status: { DRDY ERR }
ata6.00: error: { ABRT }
ata6.00: device reported invalid CHS sector 0


server is still usable though...
Steps To Reproducejust use I/O.
TagsNo tags attached.
Attached Files

-Relationships
+Relationships

-Notes

~0020735

BugMeNot (reporter)

I've experienced exactly the same errors when using a Crucial m550 1TB SSD on exactly the same Silicon Image Sil 3114 SATA-controller, and it seems that the errors are related to SSD TRIM-commands.

We've noticed these errors appearing exactly when we use lstrim. In all tests, lstrim failed with an error indicating TRIM may not be available and the error appears at the same time.

If you are mounting a (ext4) partition with the discard option, these errors may turn up every time the kernel attempts to issue a TRIM command to the drive.

I don't think this is an error related to CentOS though. It is likely a bug/problem with the Sil 3114 chipset.

I can't say I've got a well-tested solution to your problem, but you may want to try using hdparm's wiper.sh available from http://sourceforge.net/projects/hdparm/files/ - and included in that tar.gz is also a similar script for those who are using ext4 partitions on partitionable software (mdadm) RAID1-devices.

~0020754

sles (reporter)

Thank you!

I decided to not use trim for now, but, as I see there is sil driver patch in archive you mentioned which allows to trim command pass controller, so , I guess, real solution is in driver correction, if it is possible for 3114...
+Notes

-Issue History
Date Modified Username Field Change
2012-08-06 12:45 sles New Issue
2014-08-24 15:31 BugMeNot Note Added: 0020735
2014-08-26 06:01 sles Note Added: 0020754
+Issue History