Date: Sat, 19 Nov 2005 13:38:31 -0700 (MST) From: "M. Warner Losh" <imp@bsdimp.com> To: jdp@polstra.com Cc: jdp@FreeBSD.org, cvs-src@FreeBSD.org, src-committers@FreeBSD.org, jeremie@le-hen.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/cam/scsi scsi_cd.c scsi_da.c src/sys/geom geom_disk.c geom_disk.h geom_subr.c Message-ID: <20051119.133831.81090859.imp@bsdimp.com> In-Reply-To: <XFMail.20051119104237.jdp@polstra.com> References: <20051119182853.GX5197@obiwan.tataz.chchile.org> <XFMail.20051119104237.jdp@polstra.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <XFMail.20051119104237.jdp@polstra.com>
John Polstra <jdp@polstra.com> writes:
: On 19-Nov-2005 Jeremie Le Hen wrote:
: >> This fix adds a new disk_gone() function which is called by CAM when a
: >> drive goes away. It orphans all of the providers associated with the
: >> drive, setting an error condition of ENXIO in each one. In addition,
: >> we prevent a re-taste on last close for writing if an error condition
: >> has been set in the provider.
: >
: > This commit stirred my memory up about panicking when hot-unplugging
: > an USB key while it is still mounted. I have not been able to track
: > HEAD source changes attentively for some time recently but it
: > appears from the above description that this awkard behaviour has been
: > replaced by a smarter one (IOW, return ENXIO on all pending filesystem
: > actions).
: >
: > Thanks in advance for explanations.
:
: This commit may or may not fix those panics -- I don't really know.
: There is a lot that can go wrong if you remove a mounted filesystem
: from the system.
Meaning no offense to John, I somewhat doubt that his fix will help
Jeremie's problems. The problems with mounted file systems and their
disks going away is pervasive in the filesystem (and other) layers and
will require a bunch of work to fix.
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051119.133831.81090859.imp>