[DRBD-user] [Scst-devel] ]Re: vSphere MPIO with scst/drbd in dual primary mode. WAS: Re: R: Re: virt_dev->usn randomly generated!?

Chris Weiss cweiss at gmail.com
Wed Mar 10 18:56:59 CET 2010

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.

On Wed, Mar 10, 2010 at 8:23 AM, Pasi Kärkkäinen <pasik at iki.fi> wrote:
> If you have a RAID controller with battery backed cache memory,
> and the server shuts down, crashes, or you need to restart it for any reason,
> the cache contents will still be there after the reboot and all the data
> can be flushed to the disks from the battery backed memory.
> If you use server normal memory as a cache you can't really do the same, can you?

I'm with you on this one, not even to mention all the external
connections that can fail, or pulled.

Now, if you had that UPS battery on the mainboard, and software to
deal with the disks going away for a while, you'd have "the same
thing".  but you don't.

or if you had system ram on a battery and an OS that could detect that
the ram was left active before it starts overwriting it and find the
cache and commit it, that would work too.  I'm pretty sure no one has
a general purpose system that does this.

now, if scst and drbd could team up and make a way to also sync the
cache states and contents between nodes, and even hold IO till
relevant parts sync, or redirect the IO to the newer side... well,
there'd be some big storage vendors a bit scared.  and yes I know, far
easier said than done.  one can dream though, right?

More information about the drbd-user mailing list