Re: gvinum again?
Hi, Am Montag, 15. November 2004 19:41 schrieb Paul Mather: > On Mon, 2004-11-15 at 19:33 +0100, Matthias Schuendehuette wrote: > > The extension of filesystems on concat plexes works (simply) as > > advertised... :-) At least *I* had no problems so far. > > That's encouraging to hear! Can you shrink volumes and filesystems > like you can on, say, AdvFS under Tru64? That would be really nice. No, look at the growfs(8) man-page... But, honestly, my users are only *extending* their space desires... BTW my operating system too :-) So, I never had the need to shrink a filesystem. Only that of Windows if I want to install FreeBSD additionally :-) -- Ciao/BSD - Matthias Matthias Schuendehuette , Berlin (Germany) PGP-Key at and ID: 0xDDFB0A5F ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
On Mon, Nov 15, 2004 at 07:33:36PM +0100, Matthias Schuendehuette wrote: > Am Montag, 15. November 2004 18:25 schrieb Paul Mather: > > [...] > > I don't know if growfs is 100% robust enough yet to provide the other > > important ingredient to a true LVM storage management system a la the > > logical volume manager on AIX or AdvFS on Tru64, say. > > Yes, it is. I use (g)vinum primarily as a LVM with concat plexes on > ProLiant Servers with SmartRAID-Controllers, so mirroring and/or RAID5 > is done in hardware. > > The extension of filesystems on concat plexes works (simply) as > advertised... :-) At least *I* had no problems so far. I can also claim lots of positive experiences with this, though mostly on 4.X systems (I've only recnetly started to use 5.X systems on more than a test machine). In my case, I growfs mirrored volumes. I've also deleted unused volumes (mirrored), moved their subdisks over to another volume to increase available free space and then run growfs to take advantage of the new space. It's been problem free for me, and I found myself taking advantage of volume management to solve disk management issues (like user home directories). -T -- "Robert Metcalf [the inventor of Ethernet] says that if something comes along to replace Ethernet, it will be called ``Ethernet'', so therefore Ethernet will never die. Unix has already undergone several such transformations."-- Ken Thompson ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
On Mon, 2004-11-15 at 19:33 +0100, Matthias Schuendehuette wrote: > Hi, > > Am Montag, 15. November 2004 18:25 schrieb Paul Mather: > > [...] > > I don't know if growfs is 100% robust enough yet to provide the other > > important ingredient to a true LVM storage management system a la the > > logical volume manager on AIX or AdvFS on Tru64, say. > > Yes, it is. I use (g)vinum primarily as a LVM with concat plexes on > ProLiant Servers with SmartRAID-Controllers, so mirroring and/or RAID5 > is done in hardware. > > The extension of filesystems on concat plexes works (simply) as > advertised... :-) At least *I* had no problems so far. That's encouraging to hear! Can you shrink volumes and filesystems like you can on, say, AdvFS under Tru64? That would be really nice. Cheers, Paul. -- e-mail: [EMAIL PROTECTED] "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
Hi, Am Montag, 15. November 2004 18:25 schrieb Paul Mather: > [...] > I don't know if growfs is 100% robust enough yet to provide the other > important ingredient to a true LVM storage management system a la the > logical volume manager on AIX or AdvFS on Tru64, say. Yes, it is. I use (g)vinum primarily as a LVM with concat plexes on ProLiant Servers with SmartRAID-Controllers, so mirroring and/or RAID5 is done in hardware. The extension of filesystems on concat plexes works (simply) as advertised... :-) At least *I* had no problems so far. -- Ciao/BSD - Matthias Matthias Schuendehuette , Berlin (Germany) PGP-Key at and ID: 0xDDFB0A5F ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
On Mon, 2004-11-15 at 02:31 -0500, Zaphod Beeblebrox wrote: > On Mon, 15 Nov 2004 01:26:54 -0500, Paul Mather <[EMAIL PROTECTED]> wrote: > > > Your observation is correct: it doesn't (yet) load balance across plexes > > of mirrored volumes; geom_mirror (gmirror) does, though (and offers > > various load-balancing options). > > Is this something planned for the near future? I'm migrating legacy > units here, so I don't really have a choice. I don't know about the *near* future, but I believe it is on Lukas' ([EMAIL PROTECTED]) list of things to do. You'd probably have to ask him how high it is on the list, though. > ... also ... I suspect, whatever form it eventually takes, that volume > management and geom and/or vinum is essential. I my case I had a legacy root-on-vinum all-mirrored setup, and so was interested in mirroring more than LVM. (Like many, I'd used vinum just as a way of accomplishing a software RAID 1 across two drives to provide extra tolerance to drive failures.) So, migrating from geom_vinum to geom_mirror was not such a burden (or a hurdle). I don't know if growfs is 100% robust enough yet to provide the other important ingredient to a true LVM storage management system a la the logical volume manager on AIX or AdvFS on Tru64, say. Cheers, Paul. -- e-mail: [EMAIL PROTECTED] "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
On Mon, 15 Nov 2004 01:26:54 -0500, Paul Mather <[EMAIL PROTECTED]> wrote: > Your observation is correct: it doesn't (yet) load balance across plexes > of mirrored volumes; geom_mirror (gmirror) does, though (and offers > various load-balancing options). Is this something planned for the near future? I'm migrating legacy units here, so I don't really have a choice. ... also ... I suspect, whatever form it eventually takes, that volume management and geom and/or vinum is essential. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: gvinum again?
On Mon, 2004-11-15 at 00:36 -0500, Zaphod Beeblebrox wrote: > I'm looking at gstat while fsck'ing a gvinum partition. The trouble > I'm seeing is that I don't see activity on the second disk. Now... > I'm using fsck -n ... just checking things ... so there's no writes, > but does gvinum not (yet) load balance on reads? Your observation is correct: it doesn't (yet) load balance across plexes of mirrored volumes; geom_mirror (gmirror) does, though (and offers various load-balancing options). Cheers, Paul. -- e-mail: [EMAIL PROTECTED] "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
gvinum again?
I'm looking at gstat while fsck'ing a gvinum partition. The trouble I'm seeing is that I don't see activity on the second disk. Now... I'm using fsck -n ... just checking things ... so there's no writes, but does gvinum not (yet) load balance on reads? ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
