Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 14 Sep 2001 23:33:50 -0600
From:      "Stephen Hurd" <deuce@lordlegacy.org>
To:        "Dan" <dphoenix@bravenet.com>, <freebsd-hackers@freebsd.org>
Subject:   RE: PIO mode
Message-ID:  <NFBBJPHLGLNJEEECOCHAAEHFCDAA.deuce@lordlegacy.org>
In-Reply-To: <20010914191522.X61225-100000@gandalf.bravenet.com>

next in thread | previous in thread | raw e-mail | index | archive | help
> ad0s1a: UDMA ICRC error reading fsbn 191 of 64-79 (ad0s1 bn 191; cn 0 tn 3
> sn 2) retrying
> ad0s1a: UDMA ICRC error reading fsbn 191 of 64-79 (ad0s1 bn 191; cn 0 tn 3
> sn 2) retrying
> ad0s1a: UDMA ICRC error reading fsbn 191 of 64-79 (ad0s1 bn 191; cn 0 tn 3
> sn 2) retrying
> ad0s1a: UDMA ICRC error reading fsbn 191 of 64-79 (ad0s1 bn 191; cn 0 tn 3
> sn 2) falling back to PIO mode
>
> we currently do replication on disks for mirrors of webservers.
> I am wondering what this PIO mode is and if this message should be any
> concern to me at all.

Basically, the ata driver is having a problem reading (WHICH MAY BE NORMAL...
read on) and is falling back to Programmed Input Output mode... which is how
ALL IDE devices were accessed before DMA mode came along... basically it's
slower, uses more CPU, and is generally not as good.  However, some Hard
Drives/Controller combinations just plain won't run in DMA mode.  This MAY not
be a serious problem... basically if all was fine, but this suddenly started
happening, something is failing...

Actually, the ICRC error IS a bit frigtening... I always got read timeouts
when it was a bad HD/controller combo... I'd like to revise my opinion to say
"Get your data off now while you still can!"  But maybe dig through the ata
source... there's probobly helpful comments in that bit.



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?NFBBJPHLGLNJEEECOCHAAEHFCDAA.deuce>