Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 10 Aug 2000 09:56:40 +0930
From:      Greg Lehey <grog@lemis.com>
To:        "Kevin J. Menard, Jr." <kmenard@cisco.com>
Cc:        freebsd-questions@FreeBSD.ORG
Subject:   Re: vinum: faulty plex -- empty subdisks
Message-ID:  <20000810095640.J47132@wantadilla.lemis.com>
In-Reply-To: <001001c0021f$6ddc26c0$5acf45ab@cisco.com>; from kmenard@cisco.com on Wed, Aug 09, 2000 at 09:32:37AM -0700
References:  <001001c0021f$6ddc26c0$5acf45ab@cisco.com>

next in thread | previous in thread | raw e-mail | index | archive | help
[Format recovered--see http://www.lemis.com/email/email-format.html]

On Wednesday,  9 August 2000 at  9:32:37 -0700, Kevin J. Menard, Jr. wrote:
> X-Mailer: Microsoft Outlook Express 5.00.2314.1300
> X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300

Please don't write "line per paragraph" text.  It's really painful to
read, and it takes time to recover.

> I've been asking several vinum questions of this list as of late.
> I'd really like to thank everyone that has helped me out thus far.
>
> I guess first I should explain my HDD situation.  I have six (6)
> 10gig IDE drives.  They are all ATA66 devices, and come up as
> /dev/ad0, /dev/ad2, /dev/ad4, /dev/ad6, /dev/ad8, and /dev/ad10.  I
> would like to put them in two striped plexes, the second mirroring
> the first.  I put a single slice on each one.  The "boot drive"
> (/dev/ad0) now has (after yet another install ;-) / = 150M, swap =
> 150M, and then /vinum as 9470M.  I partitioned /dev/ad6 in a similar
> manner.  For the remaining drives, I just put on one partition of
> 9770M.  (As I progressed through the drives, I named the mount points
> /vinumx; x++; )
>
> As these partitions would be used just as part of a vinum volume, I
> figured there would be no need to actually mount them, and thus
> rmdir'ed their mount point dirs from /.

I'm not sure what you mean here.

> In all the drives, partition e ended up being the partition I wanted
> to use in the vinum volume.  So I disklabel -e 'd all of them and
> changed their partition types to that of "vinum".
>
> Recall that I wanted to stripe the two plexes.  If I interpreted it
> correctly (and this would sound rather logical to me), all the
> subdisks in the plexes had to be of equal size.  Thus, when I defined
> the plex, I had to set the sd length on all the drives to 9470M, the
> smallest drive size.
>
> Now, when I go to create the vinum volume, something odd happens.
> The volume comes up.  All the drives are up.  The first plex is up.
> But the second plex is "faulty".  Then the three drives in that plex
> are "empty".

This is a feature, not a bug.

> I read the online man page for vinum(4) at lemis.com that discusses
> all of the drive states, but alas, there is some type of typo there,
> and a word is omitted.

As in the rest of your text, an example would help.  At the moment I
don't even know what page to look at.

> I did not even dare to guess what that word may be, as I would
> probably just dig my hole deeper ;-)
>
> Does anyone know why a disk would be "empty".

Yes.

> If so, how can that be fixed?

Start it.

> And would that cause the plex to be "faulty"?

Yes.

From vinum(8):

GOTCHAS
     The following points are not bugs, and they have good reasons for exist-
     ing, but they have shown to cause confusion.  Each is discussed in the
     appropriate section above.

     2.   When you create a volume with multiple plexes, vinum does not auto-
          matically initialize the plexes.  This means that the contents are
          not known, but they are certainly not consistent.  As a result, by
          default vinum sets the state of all newly-created plexes except the
          first to stale. In order to synchronize them with the first plex,
          you must start their subdisks, which causes vinum to copy the data
          from a plex which is in the up state.  Depending on the size of the
          subdisks involved, this can take a long time.

          In practice, people aren't too interested in what was in the plex
          when it was created, and other volume managers cheat by setting them
          up anyway.  vinum provides two ways to ensure that newly created
          plexes are up:

          o   Create the plexes and then synchronize them with vinum start.

          o   Create the volume (not the plex) with the keyword setupstate,
              which tells vinum to ignore any possible inconsistency and set
              the plexes to be up.

     volume name [options]

           Define a volume with name name.

           Options are:

	   (....)

           setupstate

                               When creating a multi-plex volume, assume that
                               the contents of all the plexes are consistent.
                               This is normally not the case, and correctly
                               you should use the init command to first bring
                               them to a consistent state.  In the case of
                               striped and concatenated plexes, however, it
                               does not normally cause problems to leave them
                               inconsistent: when using a volume for a file
                               system or a swap partition, the previous con-
                               tents of the disks are not of interest, so they
                               may be ignored.  If you want to take this risk,
                               use this keyword.  It will only apply to the
                               plexes defined immediately after the volume in
                               the configuration file.  If you add plexes to a
                               volume at a later time, you must integrate
                               them.

                               Note that you must use the init command with
                               RAID-5 plexes: otherwise extreme data corrup-
                               tion will result if one subdisk fails.

Greg
--
When replying to this message, please copy the original recipients.
For more information, see http://www.lemis.com/questions.html
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers


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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000810095640.J47132>