Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
grrr, sent from the wrong "FROM address". List admins, please ignore the previous message just like this. > On Mon, 27 Aug 2007 08:51:09 +0300, "Michael Tewner" <tewner at gmail.com> > wrote: >> >> On 8/27/07, Lars Ellenberg <lars.ellenberg at linbit.com> wrote: >>> On Sun, Aug 26, 2007 at 06:00:16PM -0700, kbyrd-drbd wrote: >>> > >>> > I haven't used drbd yet, I'm getting ready to test and deploy it. >>> > I've searched the archives I've seen various posts about md and >>> > lvm2 and I know some of this has been covered. But, I'm confused >>> > about the current state of things with 0.8 >>> > >>> > I'd like something that feels like clustered RAID1+0 (that's >>> > striping on top of drbd). My ideal plan: lvm2 striping on top of >>> > four drbd pairs active/active pairs. I'd run GFS on top of this. >>> > Do I need cLVM instead of LVM2 for this. Does LVM striping even >>> > work with drbd? >>> >>> don't. >>> >>> DRBD does not (yet) support consistency bundling of several drbd. >>> so whenever you have a connection loss, your four devices will >>> disconnect in a slightly different order. >>> consistency of your stripe set cannot be guaranteed. >>> >>> I also think you'd get better performance out of drbd on top of raid. >>> > >> Doesn't that setup seem somewhat obsessive? >> >> > > Was that directed at me or Lars? If me, what about it? I'm new to > all this and I'm happy to hear my reasoning isn't sound and I can > simplify things. With this, I can lose more than one drive as long > as no single pair goes out. I get more performance locally because > I'm using more spindles. If drbd needs to resync after I replace a > drive, that drbd instance only has to resync one drive's worth of > data. Using LVM or md to achive the RAID0-ness of it doesn't matter > to me, I thought maybe just MD was problematic because stuff was > changing underneath it. > > What's a common active/active setup with two nodes where you've > glued together multiple drives in each node?