Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On 06.06.2013 18:30, Lutz Vieweg wrote: >>> Should I be worried? >> >> It depends on how the layers above react on this situation. If they try >> again with smaller IOs, then it's okay. Otherwise, there can be a major >> issue. Kernel code has to be read to verify. > > I could not find the right place to look at in drivers/md/dm-crypt.c, > do you have a suggestion? The queue limit stacking is in drivers/md/dm-table.c function "dm_calculate_queue_limits()". It uses "blk_set_stacking_limits()" in kernel 3.7.6. So it gets correct limits once upon getting a new dm table. I also had a look at the latest DRBD 8.4 source. They still mess around with the blk queue limits. Set it larger, set it smaller dynamically while in use depending on the limits on the other node. That's really bad for stacking. So, yes, DRBD caused that issue. No other driver does it like this. A DRBD update can have caused that issue. Cheers, Sebastian