Tweet
Hi, I have been using both the 64 GB and 256 GB Falcon SSD on an intel motherboard (ICH7) using SLES 10 SP2. I have 4 of these setups, and have been experiencing a problem where the 256 GB drive stops responding.
The first time I this happened, a soft reboot resulted in a missing drive, but a power cycle brought the drive back. I assumed it may have been a firmware bug, so I updated all of the firmware on all 8 drives.
With the firmware update, the drives are no longer appearing after a power cycle once they have failed. Almost as soon as a write test starts, there are messages in dmesg indicating stuff like this:
SCSI device sdc: 500118192 512-byte hdwr sectors (256061 MB)
sdc: Write Protect is off
sdc: Mode Sense: 00 3a 00 00
SCSI device sdc: drive cache: write back
ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata2.00: cmd ea/00:00:00:00:00/00:00:00:00:00/a0 tag 0 cdb 0x1e data 0
res 51/10:08:2d:60:5f/00:00:00:00:00/a0 Emask 0x81 (invalid argument)
ata2.00: configured for UDMA/133
ata2.01: configured for UDMA/100
ata2: EH complete
Eventually, the drive will stop responding and will not be back after a power cycle. The first time this happened I RMA'd the drive and got a replacement, but the replacement also shows the error messages above in dmesg. Two more drives have failed since then and will also not come back after a power cycle, for a total of 3 failed drives.
It is curious that the 64 GB version has not given any problems.
I am not able to reproduce the write problems using ubuntu 9.04, but I MUST use SLES 10 SP2.
what do I do???
I can provide any additional information required to troubleshoot this. Thanks.
The first time I this happened, a soft reboot resulted in a missing drive, but a power cycle brought the drive back. I assumed it may have been a firmware bug, so I updated all of the firmware on all 8 drives.
With the firmware update, the drives are no longer appearing after a power cycle once they have failed. Almost as soon as a write test starts, there are messages in dmesg indicating stuff like this:
SCSI device sdc: 500118192 512-byte hdwr sectors (256061 MB)
sdc: Write Protect is off
sdc: Mode Sense: 00 3a 00 00
SCSI device sdc: drive cache: write back
ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata2.00: cmd ea/00:00:00:00:00/00:00:00:00:00/a0 tag 0 cdb 0x1e data 0
res 51/10:08:2d:60:5f/00:00:00:00:00/a0 Emask 0x81 (invalid argument)
ata2.00: configured for UDMA/133
ata2.01: configured for UDMA/100
ata2: EH complete
Eventually, the drive will stop responding and will not be back after a power cycle. The first time this happened I RMA'd the drive and got a replacement, but the replacement also shows the error messages above in dmesg. Two more drives have failed since then and will also not come back after a power cycle, for a total of 3 failed drives.
It is curious that the 64 GB version has not given any problems.
I am not able to reproduce the write problems using ubuntu 9.04, but I MUST use SLES 10 SP2.
what do I do???
I can provide any additional information required to troubleshoot this. Thanks.
Comment