Doh. Okay, I guess I was thinking that DRBD deals with logical blocks and it would not know anything about the internal physical structure of the SSD drives, and the physical structure of one SSD drive would not necessarily be the same as the structure of the one at the other end of the network, so why would discards be replicated? But that's dumb because it still has to replicate the logical block info.
But that still does not explain why discards across the network are so slow. I mean, it's been going for like an hour and it's still only about this far... Discarding device blocks: 13111296/322122752 -- Eric Robinson > -----Original Message----- > From: [email protected] [mailto:drbd-user- > [email protected]] On Behalf Of Robert Altnoeder > Sent: Monday, July 31, 2017 1:29 AM > To: [email protected] > Subject: Re: [DRBD-user] Discarding device blocks very slow > > On 07/31/2017 09:46 AM, Eric Robinson wrote: > > > > When I create a filesystem on an LVM logical volume on a drbd device, > > I see "Discarding device blocks." I wouldn't think that discards would > > be replicated, but it is going super slow. I mean SUPER slow. Discards > > normally go a 100 times faster. Are discards replicated across the > > network for some reason? > > > Obviously. If a block is discarded locally, the same change has to be applied > to > the replica, otherwise a read from the same block on both systems can yield > different results. > > > > -- > > > > Eric Robinson > > > -- > Robert Altnoeder > +43 1 817 82 92 0 > [email protected] > > LINBIT | Keeping The Digital World Running DRBD - Corosync - Pacemaker f / t > / in / g+ > > DRBD(r) and LINBIT(r) are registered trademarks of LINBIT, Austria. > _______________________________________________ > drbd-user mailing list > [email protected] > http://lists.linbit.com/mailman/listinfo/drbd-user _______________________________________________ drbd-user mailing list [email protected] http://lists.linbit.com/mailman/listinfo/drbd-user
