On Mon, Nov 14, 2022 at 6:30 AM Philipp Reisner <philipp.reis...@linbit.com> wrote:
> Hello, > > During the last week no showstopper came up. Here are the final releases. > > The most outstanding improvement is supporting thinly provisioned backing > devices with discard granularities larger than 128MiB. Here I mean with > supporting, that during a resync we make sure not to cause unnecessary > allocations in the backing block device, and merging the discard requests > for good resync performance. > Previously DRBD was only dealing with discard granularities of up to 1MiB. > > The LVM tools create big chunksizes in thinpools for large pools. The > chunksize is then what determines the discard granularity. ZFS on the other > hand always stays with 8k, and it consumes too much space for its metadata > on large pools. The middle ground would be optimal. We will write an > article > on these chunksize considerations. Also mentioning how to set these > settings > with LVM-thin and ZFS. > > On top of that, we have a bunch of fixes and lots of progress with upstream > kernel compatibility. > > 9.2.1 (api:genl2/proto:110-121/transport:18) > -------- > * fix spurious PingAck timeout, a regression of ack-processing in bottom > half (introduced with 9.2.0) > * support merging of discards during resync even if the discard granularity > of the backing device is larger than 128MiB > * merged changes from the drbd-9.1 branch, including > - fix a race that could result in connection attempts getting aborted > with the message "sock_recvmsg returned -11" > - rate limit messages in case the peer can not write the backing storage > and it does not finish the necessary state transitions > - reduced the receive timeout during connecting to the intended 5 seconds > (ten times ping-ack timeout) > - losing the connection at a specific point in time during establishing > a connection could cause a transition to StandAlone; fixed that, so > that it keeps trying to connect > - fix a race that could lead to a fence-peer handler being called > unexpectedly when the fencing policy is changed at the moment before > promoting > > 9.1.12 (api:genl2/proto:110-121/transport:17) > -------- > * fix a race that could result in connection attempts getting aborted > with the message "sock_recvmsg returned -11" > * rate limit messages in case the peer can not write the backing storage > and it does not finish the necessary state transitions > * reduced the receive timeout during connecting to the intended 5 seconds > (ten times ping-ack timeout) > * losing the connection at a specific point in time during establishing > a connection could cause a transition to StandAlone; fixed that, so > that it keeps trying to connect > * fix a race that could lead to a fence-peer handler being called > unexpectedly when the fencing policy is changed at the moment before > promoting > > https://pkg.linbit.com//downloads/drbd/9/drbd-9.2.1.tar.gz > https://pkg.linbit.com//downloads/drbd/9/drbd-9.1.12.tar.gz > > > https://github.com/LINBIT/drbd/commit/86ec2326fef3aede9f4d46f52bfd35aac4d5eb7e > > https://github.com/LINBIT/drbd/commit/bfc1691453a36de688e013239ce5c586263933fa > I've been maintaining the kmod-drbd9x package for ELRepo. Version 9.1.12 builds cleanly on RHEL 9.1 beta (kernel-5.14.0-160.el9). It came out just in time for the GA release of RHEL 9.1. Thank you, Akemi
_______________________________________________ Star us on GITHUB: https://github.com/LINBIT drbd-user mailing list drbd-user@lists.linbit.com https://lists.linbit.com/mailman/listinfo/drbd-user