Date: Fri, 11 Sep 2009 12:21:39 +0200 From: Ivan Voras <ivoras@freebsd.org> To: freebsd-stable@freebsd.org Subject: Re: 8.0-B4 gstripe / GEOM_PART_* upgrade woes Message-ID: <h8d8b5$b3q$1@ger.gmane.org> In-Reply-To: <4AAA1296.2080705@sasktel.net> References: <4AAA1296.2080705@sasktel.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Stephen Hurd wrote:
> I've upgraded from 7.2-RELEASE_p2 to 8.0-BETA4 and using GEOM_PART_*
> with my sliced gstripe array causes the /dev/stripe/raid0a to disappear
> and the reset of the /dev/stripe/raid0[a-z] file systems to be unmountable.
>
> My gvinum array is still working fine and, after chasing the ad* slices,
> they can be mounted as well. It's just the gstripe slices which are
> corrupt/missing.
Ouch.
> GEOM_STRIPE: Device raid0 created (id=40432321).
> GEOM_STRIPE: Disk da0s2 attached to raid0.
> GEOM_STRIPE: Disk da1s2 attached to raid0.
> GEOM_STRIPE: Disk da2s2 attached to raid0.
> GEOM_STRIPE: Disk da3s2 attached to raid0.
> GEOM_STRIPE: Device raid0 activated.
> Trying to mount root from ufs:/dev/da0s1a
> <<111188>C>aCna'tn 'stta ts t/atd e/vd/aedv0s/1sgt:ri pNeo/ rsuacidh0
> af:i lNeo sour directory
> c
> =========== END OF dmesg ===========
>
> =========== gstripe list ===========
> Geom name: raid0
> State: UP
> Status: Total=4, Online=4
> Type: AUTOMATIC
> Stripesize: 262144
> ID: 40432321
> Providers:
> 1. Name: stripe/raid0
> Mediasize: 42934992896 (40G)
> Sectorsize: 512
> Mode: r0w0e0
> Consumers:
> 1. Name: da0s2
> Mediasize: 10733990400 (10G)
> Sectorsize: 512
> Mode: r0w0e0
> Number: 0
> 2. Name: da1s2
> Mediasize: 10733990400 (10G)
> Sectorsize: 512
> Mode: r0w0e0
> Number: 1
> 3. Name: da2s2
> Mediasize: 10733990400 (10G)
> Sectorsize: 512
> Mode: r0w0e0
> Number: 2
> 4. Name: da3s2
> Mediasize: 10733990400 (10G)
> Sectorsize: 512
> Mode: r0w0e0
> Number: 3
> =========== END OF gstripe list ===========
>
> =========== gstripe status ===========
> Name Status Components
> stripe/raid0 UP da0s2
> da1s2
> da2s2
> da3s2
> =========== END OF gstripe status ===========
> When I build without GEOM_PART_* and use GEOM_BSD and GEOM_MBR, it works
> fine.
An interesting problem. I presume that in either case (gpart or
GEOM_BSD/MBR) the output of "gstripe status" is the same? Only the
interpretation of the partition tables is problematic?
What is the expected ("good") structure of the partitions/file systems?
Do you have a single MBR partition and inside it multiple BSD
partitions? What are their partition types?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?h8d8b5$b3q$1>