Well, I shouldn't loose data. Placement groups become inactive until peering is over, that issue I get. But I've never seen a Ceph cluster letting someone write data into the void. What I was talking about, was a pool with size 4 and min_size 2 using the crush rules I mentioned. I don't see, how it's possible to tell, if a crush rule is safe to use without knowing the size (s) of the actual pool(s) using it. Maybe I'm better of testing. On May 9, 2022 11:45:38 AM GMT+02:00, Eneko Lacunza <elacunza@xxxxxxxxx> wrote: >Hi, > >El 9/5/22 a las 11:05, Maximilian Hill escribió: >> sure, let's say min_size is 2 (crush rule and pool). >> >> This would also allow me to read and write, when a site is down and would be able to recover, when the site comes back. Is there any more to stretch mode, or is this essentially it? >> >> I'm just wondering, if the magic is happening in stretch mode itself, or the other actions to take according to the documentation. > >What you are missing from stretch mode is that your CRUSH rule wouldn't guarantee at least one copy in surviving room (min_size=2 can be achieved with 2 copies in lost room). You may lose data until room is recovered (no copies present), or you may have blocked I/O until a second copy is peered in surviving room. You would need size=4/min=3 for having at least one guaranteed copy in each room; but I/O would block until min size is adjusted when room is lost. > >Also surviving room will have blocked I/O is an OSD crashes/reboots etc. until you adjust min_size. > >Cheers > >> >> On May 9, 2022 9:37:35 AM GMT+02:00, Eneko Lacunza<elacunza@xxxxxxxxx> wrote: >>> Hi Maximilian, >>> >>> El 7/5/22 a las 19:17, Maximilian Hill escribió: >>>> >>>> This would mean, if I already had a crush rule, like the following, I wouldn't really need to enable stretch mode: >>>> >>>> type replicated >>>> min_size 4 >>>> max_size 4 >>>> step take room0 ssd >>>> step chooseleaf firstn 2 type host >>>> step emit >>>> step take room1 ssd >>>> step chooseleaf firstn -2 type host >>>> step emit >>>> >>>> Am I right about that? >>>> >>> With your configuration, if one DC goes down all I/O will stop (min_size=4!) >>> Further, I/O will stop even with one OSD down, for the same reason. >>> >>> Enabling stretch mode will keep I/O going without intervention. > > > EnekoLacunza > >Director Técnico | Zuzendari teknikoa > >Binovo IT Human Project > > 943 569 206 <tel:943 569 206> > > elacunza@xxxxxxxxx <mailto:elacunza@xxxxxxxxx> > > binovo.es <//binovo.es> > > Astigarragako Bidea, 2 - 2 izda. Oficina 10-11, 20180 Oiartzun > > >youtube <https://www.youtube.com/user/CANALBINOVO/> > linkedin <https://www.linkedin.com/company/37269706/> _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |