Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Thank you everyone ! I'll do some tests monday and tell you if I could fix that stuff. Regards, Olivier XO Project http://xen-orchestra.com On Fri, Mar 12, 2010 at 11:16 PM, Lars Ellenberg <lars.ellenberg at linbit.com> wrote: > On Fri, Mar 12, 2010 at 10:38:27AM +0100, Olivier LAMBERT wrote: >> Ok, so after reading that : >> http://lists.linbit.com/pipermail/drbd-user/2008-June/009636.html >> and >> http://archives.free.net.ph/message/20090319.001657.5a0095e9.en.html >> >> Where you talking about cache consistency, you think the problem might >> be on the Dell PowerVault MD1000 SAS ? > > No. > Above DRBD layer. > Not below. > > I'm still guessing here. > > But here is what I _know_. > 1) > DRBD replicates whatever is written to it. > This is proven in the real world. > LINBIT would not exist if it did not. > So that cannot be the problem. > > 2) > You observe that data written on some iscsi client > to a target on top of DRBD is not (immediately) visible > from some other iscsi client. > This I have to believe you. > > Which leaves me at: > either the data in question does not reach DRBD, > or someone who is supposed to read from DRBD does not, > because some layer in between servers "cached" data > which happens to be stale content. > > Does that make sense? > >> Sorry, but sometimes only "cache consistency" is too vague for me... > > We also do consulting, workshops, > project guidance and monitoring, > and in general support ;) > > -- > : Lars Ellenberg > : LINBIT | Your Way to High Availability > : DRBD/HA support and consulting http://www.linbit.com > > DRBD® and LINBIT® are registered trademarks of LINBIT, Austria. > __ > please don't Cc me, but send to list -- I'm subscribed > _______________________________________________ > drbd-user mailing list > drbd-user at lists.linbit.com > http://lists.linbit.com/mailman/listinfo/drbd-user >