New "ahcisata0 port 1: device present" messages with NetBSD 9

[Simon Burge <simonb%NetBSD.org@localhost>]

Hi Jaromir, folks,

I've just upgraded a HP N40L microserver from NetBSD 8.1 to NetBSD 9.0_RC1
and am seeing lots of

    ahcisata0 port 1: device present, speed: 3.0Gb/s

for the two attached SSDs, seemingly triggered by disk activity.  These
SSDs are running as a RAIDframe mirror.  Unpacking the 9.0_RC1 sets with
the new kernel triggered around 50 of these, as well as a few "soft
error (corrected)" messages like these:

    wd0a: error writing fsbn 15440256 of 15440256-15440271 (wd0 bn 15452544; cn 15329 tn 14 sn 30), xfer 330, retry 0
    wd0: (aborted command, interface CRC error)
    wd0a: requeue writing fsbn 15466952 of 15466952-15466959 (wd0 bn 15479240; cn 15356 tn 6 sn 14), xfer d0, retry 0
    wd0a: requeue writing fsbn 15440080 of 15440080-15440091 (wd0 bn 15452368; cn 15329 tn 11 sn 43), xfer 200, retry 0
    wd0a: requeue writing fsbn 15440192 of 15440192-15440203 (wd0 bn 15452480; cn 15329 tn 13 sn 29), xfer 460, retry 0
    wd0a: requeue writing fsbn 15404472 of 15404472-15404479 (wd0 bn 15416760; cn 15294 tn 6 sn 30), xfer 590, retry 0
    wd0a: requeue writing fsbn 15404384 of 15404384-15404407 (wd0 bn 15416672; cn 15294 tn 5 sn 5), xfer 6c0, retry 0
    wd0: soft error (corrected) xfer 460
    wd0: soft error (corrected) xfer 200
    wd0: soft error (corrected) xfer d0
    wd0: soft error (corrected) xfer 6c0
    wd0: soft error (corrected) xfer 590
    wd0: soft error (corrected) xfer 330

This machine has been running NetBSD 6 and then NetBSD 8.1 for a number
of years without any sata errors of any kind.

First, is this something I should be worried about from a data loss
point of view?

Second, can this be fixed assuming it's a software issue? :)


Also see https://gnats.netbsd.org/54855 for a report in Gnats of the
same problem.

Cheers,
Simon.