From: Roland Kammerer <roland.kammerer@linbit.com>
To: drbd-announce@lists.linux.dev
Subject: drbd-utils v9.36.0-rc.2
Date: Wed, 11 Nov 2026 15:53:21 +0100 [thread overview]
Message-ID: <avSBYSLIYXTqGR3z@arm64> (raw)
[-- Attachment #1: Type: text/plain, Size: 2637 bytes --]
Dear DRBD users,
this is the second RC of the upcoming version 9.36.0. Besides the
changes already in rc.1, it fixes a drbdmeta apply-al bug present
since 9.33.0, where the last activity log extent of a device was not
resynced after a crash of the Primary, and keeps the bitmap provenance
feature of recent kernel modules, which apply-al used to clear. The
cbt commands now also work over the legacy "drbd" netlink family, and
cbt-dump learned to report never written extents and to take the zero
extents of a dm-thin volume from its pool. drbdsetup attach gained
--discard-my-data to resolve a split brain detected on attach.
Please test, if we don't find any issues the final release will be about
a week from now. The PPA build will follow, looks like it currently has
problems out of our control.
Regards, rck
GIT: https://github.com/LINBIT/drbd-utils/commit/fb66d2599ae16b89cfe2fd51d5d7ec28918de487
TGZ: https://pkg.linbit.com//downloads/drbd/utils/drbd-utils-9.36.0-rc.2.tar.gz
PPA: https://launchpad.net/~linbit/+archive/ubuntu/linbit-drbd9-stack
Changelog:
9.36.0-rc.2
-----------
* drbdadm: raise the default cmd-timeout-short from 5 to 10 seconds;
keep the medium and long defaults that were in effect (600, none)
* drbdadm: show changed cmd-timeout-* settings in the v9 dump
* drbdadm: fix adjust failing silently when a wait-connect lasted
longer than the remaining cmd-timeout of the preceding connect
* drbdmeta: keep the bitmap provenance feature of DRBD 9.2.20 and 9.3.4
when writing the superblock; apply-al after a crash cleared it, and
DRBD then counted a crashed Secondary's activity log bits as set for
blocks its peers lack
* drbdmeta: forget-peer makes the freed bitmap slot a day0 slot, a
copy of one or all bits set, as the kernel does
* drbdmeta: cbt-dump --mode=dirty|zero also reports the extents never
written since create-md, from the day0 bitmap; "format": 1; the
extents are start and length and cover the volume, as in ovirt-imageio
* drbdmeta: cbt-dump on a dm-thin volume takes the zero extents from
the mappings of the thin pool, which, unlike the day0 bitmap, know
about discards; with external meta data, --data-device (drbdadm:
--data-snapshot) names the snapshot of the backing device
* drbdmeta: fix apply-al skipping the last activity log extent of the
device as beyond the end of the bitmap, so that extent was not
resynced after a crash of the Primary; since 9.33.0
* drbdmeta: apply-al reports the amount it marked out of sync
correctly; it reported 0 KB with up to 128 KB per bitmap bit, and
counted bits past the end of the device
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
reply other threads:[~2026-11-11 14:53 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=avSBYSLIYXTqGR3z@arm64 \
--to=roland.kammerer@linbit.com \
--cc=drbd-announce@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox