Prolonging a drive's life

[Mikhail T.]

One of the four drives in my system is frequently timing out of late,
although the operation succeeds on a second attempt:

Aug 14 13:51:59 aldan kernel: (ada4:ahcich5:0:0:0): FLUSHCACHE48.
ACB: ea 00 00 00 00 40 00 00 00 00 00 00
Aug 14 13:51:59 aldan kernel: (ada4:ahcich5:0:0:0): CAM status:
Command timeout
Aug 14 13:51:59 aldan kernel: (ada4:ahcich5:0:0:0): Retrying command
Aug 14 13:59:12 aldan kernel: (ada4:ahcich5:0:0:0): FLUSHCACHE48.
ACB: ea 00 00 00 00 40 00 00 00 00 00 00
Aug 14 13:59:12 aldan kernel: (ada4:ahcich5:0:0:0): CAM status:
Command timeout
Aug 14 13:59:12 aldan kernel: (ada4:ahcich5:0:0:0): Retrying command

While I'm getting a replacement, maybe, I can use camcontrol to somehow
lower the operating system's exceptions about it? For example, the
"camcontrol negotiate" returns the following about it:

Current parameters:
(pass5:ahcich5:0:0:0): SATA revision: 2.x
(pass5:ahcich5:0:0:0): ATA mode: UDMA6
(pass5:ahcich5:0:0:0): ATAPI packet length: 0
(pass5:ahcich5:0:0:0): PIO transaction length: 8192
(pass5:ahcich5:0:0:0): PMP presence: 0
(pass5:ahcich5:0:0:0): Number of tags: 32
(pass5:ahcich5:0:0:0): SATA capabilities: 00000030
(pass5:ahcich5:0:0:0): tagged queueing: enabled

Is there anything I can tweak for it to keep working even if at lower
speeds?

Also, years ago, some BIOSes had the feature, which would "verify" a
drive -- is there something similar I can trigger with camcontrol or
smartctl?

Thanks!

-mi


_______________________________________________
freebsd-...@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-hardware
To unsubscribe, send any mail to "freebsd-hardwa...@freebsd.org"