Hello DRBD-users,
A detail I never cared about is if a syscall gets restarted or returns
EINTR after the arrival of a signal. What made me learn this detail is a
customer whose system newly created processes get a signal. Often, a
process open()s the resources it needs to work with first, and on
Hello DRBD-users,
A detail I never cared about is if a syscall gets restarted or returns
EINTR after the arrival of a signal. What made me learn this detail is a
customer whose system newly created processes get a signal. Often, a
process open()s the resources it needs to work with first, and i
Hi DRBD-users,
You might have noticed that this release comes about four weeks earlier
than our usual pace. The reason is that we discovered two regressions we
want to fix. Both regressions were introduced in the last cycle, into
drbd-9.1.20 and drbd-9.2.9.
One is that if the size of your backing
Hi DRBD-users,
You might have noticed that this release comes about four weeks earlier
than our usual pace. The reason is that we discovered two regressions we
want to fix. Both regressions were introduced in the last cycle, into
drbd-9.1.20 and drbd-9.2.9.
One is that if the size of your backi
Hello DRBD-users,
In the week of the release candidate, we saw one of the CI tests
delivering more faults than previously. It turned out that the changed
timing of the RC code triggered an independent bug more often. Fixed
that bug. With that, the stability tests (running all CI tests in 50
iter
Hello DRBD-users,
In this development cycle, we had two bug reports worth mentioning. One
was about a perplexing kernel crash stack trace. Fortunately, the
customer, who runs a big fleet of DRBD nodes, could reproduce it and
produce a kernel crash dump of this crash.
Joel and I took it as a cha
Hi Aleksandr,
Fixed with commits
https://github.com/LINBIT/drbd/commit/857db82c989b36993ff7a3df3944c9862db1408d
and
https://github.com/LINBIT/drbd/commit/343e077e9664b203e5ebf8146dacc5c869b80e30
Next DRBD release follows at the end of the month.
On Fri, Apr 12, 2024 at 10:39 AM Joel Colledge wr
Hello Tim,
We were able to write a reproducer test case and fix this regression
with this commit:
https://github.com/LINBIT/drbd/commit/be9a404134acc3d167e8a7e60adce4f1910a4893
This commit will go into the drbd-9.1.20 and drbd-9.2.9 releases.
best regards,
Philipp
On Fri, Mar 22, 2024 at 1:49
Hi DRBD-users,
As so often, we had to fix bugs in the area not-so-common corner cases.
We started with a long hunt for a bug that caused a diskless primary
node that seldom fails to re-issue pending read requests to another node
with a backing disk when it loses connection to the node it first s
Hello,
As always, this release brings several fixes. For many of those fixes,
we created new test cases or extended existing test cases to verify the
bug's existence and the patch's effectiveness.
What we learned was that we needed more testing in the area of the
checksum-based resync. We have
Hello,
Here comes the DRBD December release. drbd-9.1.18 and drbd-9.2.7.
The updates that are worth mentioning are...
The RDMA transport finally reaches general usability. As usual, I have
plans to improve it further, it never stops.
Regarding quorum, for a cluster containing two nodes with back
Hello,
Here comes the DRBD December release. drbd-9.1.18-rc.1 and drbd-9.2.7-rc.1.
We have two updates that are worth mentioning.
One, the RDMA transport finally reaches general usability. As usual, I
have plans to improve it further, it never stops.
The other improvement is in the area of the
Hi Michał,
Can you pack all the *.res files in /var/lib/linstor.d into a tar.gz
and make that available?
I want to have a look.
best regards,
Phil
___
Star us on GITHUB: https://github.com/LINBIT
drbd-user mailing list
drbd-user@lists.linbit.com
https:
Happy Halloween our spooky friends
Our latest release shows we've made some mends
To make sure your data processing has little delay
We have brought back our RDMA from decay
Our new DRBD encryption is meant to be scary
It will definitely make any unauthorized parties be wary
Our last update is
Here are drbd-9.1.17-rc.2 and drbd-9.2.6-rc.2. Only a few changes on the
9.1 branch. After we had this UBSAN warning in rc.1 I want to give it
one more RC week.
The 9.2 branch saw way more changes than what is healthy for a release
candidate week.
I want to maintain only one TCP transport imple
Hello,
This time, it took a bit longer, around ten weeks. For me, this was a
fascinating development cycle.
We got important fixes to the RDMA transport, which now works with the
more recent Mellanox cards and drivers. However, there is still room for
improvement in the performance of DRBD's RD
Hello DRBD users,
Here are the final drbd-9.1.16 and drbd-9.2.5 releases. No regression
was found during the release candidate week.
This is a boring release because we fixed no exceptional bugs. All the
addressed bugs need very particular conditions to trigger. Nearly all of
them have consequenc
Hello DRBD users,
Following the usual about eight-week release cadence, here are
drbd-9.1.16-rc.1 and drbd-9.2.5-rc.1 comming.
This is a boring release because we fixed no exceptional bugs. All the
addressed bugs need very particular conditions to trigger. Nearly all of
them have consequences tha
Hello DRBD users,
I am glad to announce yet another release, drbd-9.1.15 and drbd-9.2.4.
Our 'eat your own dog food' policy led to another bug we were able to
fix shortly before the release (during the rc.1 phase). It reproduces on
drbd-9.2. On drbd-9.1 it reproduces only with a non-default CPU m
Hi DRBD users,
I am glad to announce yet another release cycle, drbd-9.1.15-rc.1 and
drbd-9.2.4-rc.1.
I know that users of recent kernels are eagerly waiting for this
release. Even if a previous DRBD release was compile-time compatible
with Linux-6.2 and later, the resulting module crashed your
s
Hello DRBD-user,
Today I am announcing the drbd-9.1.14 and drbd-9.2.3 releases. In this
eight-week development cycle, we fixed interesting issues in different
places.
An area of attention was the "resync-after" functionality. This
feature allows you to serialize resyncs of resources. That is
int
Hello DRBD-user,
Today I am announcing the drbd-9.1.14 and drbd-9.2.3 releases. In this
eight-week development cycle, we fixed interesting issues in different
places.
An area of attention was the "resync-after" functionality. This
feature allows you to serialize resyncs of resources. That is
i
Hello DRBD-user,
Today I am announcing the release candidate 1 of the next releases of
drbd-9.1 and drbd-9.2. In this eight-week development cycle, we fixed
interesting issues in different places.
An area of attention was the "resync-after" functionality. This
feature allows you to serialize resy
Hi Markus,
Thanks for sending this bug report, including instructions on
reproducing it. At first, I ignored your report because I could not
reproduce the issue. Thanks to your persistence, I realized that this
issue only reproduces on the versions you reported. So it is something
that is alread
Hi,
This is the final release of drbd-9.1.13 and drbd-9.2.2. In the number
of changed lines, the work on improving quorum is the most significant
change in this release. Here is a minor example that highlights the
improvement.
Consider nodes A, B, and C with ascending node IDs.
C |
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
Hello,
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
Hello,
I am proud and glad to announce drbd-9.2.0. It is a new feature release,
bringing:
1) Lower latency for mirrored write requests
Every mirrored write request gets completed one scheduler context switch
earlier. This is a result of better integration with the networking stack.
2) Lower cont
Hello,
Finally, the RDMA transport got the long necessary attention. Now it tests
successfully on RHEL9 clones. And, BTW, we enabled also all of the regular
tests, that test network interruption via iptables also for the RDMA
transport. (iptables can only fetch RoCE packets with rdma_rxe, the
soft
Hello,
The fix for the request processing that was quickly released with drbd-9.1.10,
created another problem in the request processing code. This other problem is
only reproducibly by us when we artificially delay a CPU thread by
instrumenting the code.
This bug in the request processing code can
Hello,
It happened. There is a regression in drbd-9.1.9 that can cause kernel
crashes. We are releasing drbd-9.1.10 immediately. With just this regression
fixed.
It is one of these cases, where a change interferes with the existing code
in a way that we were not able to see during development and
Hi,
we think that we fixed all known issues in the drbd-9.2 branch. (With the
exception of the RDMA transport, which still has known issues)
This time it comes with a new feature: Support for network namespaces
That means that a drbd path (the part of a drbd connection that has IP
addresses), wi
Hello,
during the summer season, we learned that in the progress of following
upstream changes and introducing compatibility code (tests and coccinelle
spatch files) we broke DRBD on partitions on all kernels older than Linux
5.10.
Yes, partitions are handled differently than logic volumes in som
Hello,
during the summer season, we learned that in the progress of following
upstream changes and introducing compatibility code (tests and coccinelle
spatch files) we broke DRBD on partitions on all kernels older than Linux
5.10.
Yes, partitions are handled differently than logic volumes in som
Hello DRBD users,
Here is the final drbd-9.1.8 release. Another rc in the 9.2 series will
follow next week.
With this release we restored the protocol compatibility with drbd-8.4. The
reason behind this is that we want to discontinue drbd-9.0. Since upstream
Linux still contains drbd-8.4 we want t
Hello DRBD users,
Announcing the parallel release of drbd-9.1.8-rc.1 and drbd-9.2.0-rc.6. In both
you will find that we restored the protocol compatibility with drbd-8.4. The
reason behind this is that we want to discontinue drbd-9.0. Since upstream
Linux still contains drbd-8.4 we want to have ou
Hello,
Since rc.2 we fixed the issue on Linux 5.15. Other than that I am
repeating the text I wrote for rc.2: We were working on smaller
issues with unfreezing IO-requests after quorum loss and regaining
quorum. At this point, one thing gave the other. We discovered how difficult
it is to graceful
Hello,
You all noticed the unusual long rc.1 phase. We were working on smaller
issues with unfreezing IO-requests after quorum loss and regaining
quorum. At this point, one thing gave the other. We discovered how difficult
it is to gracefully recover a primary node that lost quorum. It was like a
Hello,
I am starting the drbd-9.1.7 release cycle early. This time, the most
relevant change is fixing IO request unfreezing on diskless nodes. It
turned out this was broken, now there is a test that examines all these
cases. A number of bugs got fixed in this area.
These are important fixes for u
Hello,
In this cycle, we learned that DRBD was not dealing correctly with internal
meta-data on backing devices larger than 128TB. Why did we not discover this
earlier? (we pretend that DRBD can hand up to 1PB per volume/device) Well,
we tested with device-mapper-based thinly provisioned LVs. It t
Hi,
In this cycle, we learned that DRBD was not dealing correctly with internal
meta-data on backing devices larger than 128TB. Why did we not discover this
earlier? (we pretend that DRBD can hand up to 1PB per volume/device) Well,
we tested with device-mapper-based thinly provisioned LVs. It turn
Hello,
and here is the 3rd release candidate for drbd-9.2.0
It turned out that in the new coordination between resync and
application IOs a corner case was not properly considered. Fixed and
tested this, now we are turning back to the community and asking you
if you find a way to break it.
We wi
Hi,
drbd-9.2.0 gets a second release candidate.
Accidentally, rc.1 broke wire protocol compatibility with all earlier DRBD
releases. That is fixed in rc.2 of course. The other change relates to the
DKMS build process on kernels that have CONFIG_INFINIBAND not set.
We give it at least another wee
Hi,
Here are the final releases of drbd-9.0.32-1 and drbd-9.1.5.
For 9.0 we got a dividend from the WinDRBD effort. There was an
access-after-free bug hiding in the code, that is no issue when DRBD is
running in the Linux kernel since the Linux kernel does not un-map the
addresses after freeing o
Hello,
This Thursday, December 9th, I will present the planned, tentative
technical ROADMAP for DRBD, LINSTOR and the other LINBIT software
components for 2022. It will be an "open" zoom call, with input, discussion
and feedback welcome!
https://linbit.com/community-meeting/
best regards,
Phil
Hello,
Nikolaus/Saint Nicholas brings you the first triple release of DRBD.
For 9.0 we got a dividend from the WinDRBD effort. There was an
access-after-free bug hiding in the code, that is no issue when DRBD is
running in the Linux kernel since the Linux kernel does not un-map the
addresses afte
Hello,
no show-stopper came up last week, so here is the final release.
Many of the fixes are fixes to specific corner cases.
A noticeable change is that the "--discard-my-data" flag (it is an
option to the connect command) was changed. So far it was only used to
express how a split-brain situat
Hello,
Here we are in the process of flushing out the fixes the accumulated
during August and September. Most are quite specific corner cases.
A noticeable change is that the "--discard-my-data" flag (it is an
option to the connect command) was changed. So far it was only used to
express how a sp
Hi,
Here is the final of drbd-9.0.30 & drbd-9.1.3. No changes that are
worth mentioning since the rc1.
In general, this is a "small" release. I am happy to see that the
issues in the 9.1 branch get shaken out as it is more tested and used.
For autumn, we will have two changes in the works for th
Hello,
DRBD enters the next one-week release-candidate phase.
This release contains a small number on fixes on the 9.0 branch. It feels
like
we are reaching the bottom of the bug-pool.
The 9.1 branch has a few specific fixes, also showing that it gets picked up
by more users.
This is a release
Hello,
I have the honor to announce new DRBD releases.
We got a report about data corruption on DRBD volumes if the backing device is
a degraded Linux software raid 5 (or raid 6). Affected are kernels >=4.3. I.e.
the distros: RHEL8/CentOS8/AlmaLinux8/RockyLinux8, Xenial and newer etc.
Not affecte
Hi,
here is the next release candidate for both of our branches. I promise to write
a bit more text for the final release, which will happen in one week if no show
stoppers are found.
This is a release candidate, please help testing it.
9.0.29-0rc1 (api:genl2/proto:86-120/transport:14)
Hi,
Debian bullseyes (Debian 11.0) entered 'hard freeze' two weeks ago and we
have customers that want to use DRBD and LINSTOR on that
platform. Unfortunately, I made a stupid mistake a few weeks back and that
broke the kernel compat for kernel 5.10 (which comes with bullseye). So, here
is a relea
Hi,
drbd-9.1.0 is a drop-in replacement for drbd-9.0.x. Same config. Same
userspace utilities. Same wire protocol.
Compared to drbd-9.0.z it contains a changeset that reduces
contention on split-locks in the IO-submit code path of DRBD. That is
the code path that is most critical for DRBD's IO per
Hi,
during the last week before the release, we let another few fixes in,
which again, are really hard to trigger.
Two minor features are coming with this release:
1 Resource renaming
2 ability to disable the upper bandwidth boundary for the resync controller
9.0.28-1 (api:genl2/proto:86-119/t
Hi,
It is a drop-in replacement for drbd-9.0.x. Same config. Same
userspace utilities. Same wire protocol.
This is a release candidate, please help testing it.
9.1.0-rc.2 (api:genl2/proto:110-119/transport:16)
* merged fixes from 9.0.28-rc2
- fix a regression introduces with 9.0.25;
Hi,
during rc1 week two bug fixes got merged. Both pretty exotic bugs (in
the sense of affecting very few users). Just yesterday new issues were
reported, we want to examine them before we make this a final release.
This is a release candidate, please help testing it.
9.0.28-0rc2 (api:genl2/prot
Hi,
In May 2018, I started to work on a changeset that should reduce
contention on split-locks in the IO-submit code path of DRBD. That is
the code path that is most critical for DRBD's IO performance. It
needed a rework to reflect CPUs that have an increasing number of CPU
cores and to address NV
Hi community,
I called the 9.0.26 release a 'big release'. Compared to that this will be
a small release. Though it contains a number of fixes, they are very hard to
trigger.
Two minor features are coming with this release:
1 Resource renaming
2 ability to disable the upper bandwidth boundary f
Dear DRBD users,
This is a big release. The release candidate phase lasted more than a
month. Bug reports and requests were coming in concurrently from different
customers/users working on different use-cases and scenarios.
One example: the XCP-ng driver developers need to switch all nodes quick
Hi,
What a week. So many changes came in. I will write some more text for the final
9.0.26. We hope to have it ready in a week from now.
This is a release candidate, please help testing it.
9.0.26-0rc4 (api:genl2/proto:86-118/transport:14)
* fix a source of possible data corruption; re
Hi,
It is time for a rc3, rc2 is already nearly 3 weeks old!
We were very busy ironing out details with the state engine for strate
transitions when nodes establish a connection. Well, two partitions
join. It looks really good now. A new test tortures it in a way we
never tested it before. I am c
Hi,
we had a very busy week. A number of merge-requests got ready. A bug
fix from an external contributor cam in this week. The only issue that
was reported from the -rc1 release was a minor glitch in the compat
layer. That was fixed as well.
I hope it calms down so that we can tag the final next
Hi,
with this release, DRBD becomes compatible with Linux 5.8 and 5.9. Thanks to
the coccinelle based compatibility framework it stays compatible all the
way back to the kernel 2.6.32.
In upstream quite a few things changed for block device drivers, following
all that and staying compatible with t
Hi,
Here is the final release! It has been busy hours before the release as
some last minute fixes came in.
It is worth noting that performance increases with this release. The IO
submission overhead was reduced in general and even more if there are
multiple or many (busy) DRBD devices on a serve
Hi,
The main reason to do a release is that with 9.0.25 the overhead submitting
IO got smaller. On some specific hardware the fio results go from IOPS=229k,
BW=896MiB/s to IOPS=370k, BW=1445MiB/s (single DRBD device).
Especially with multiple or many DRBD devices on a single server, the
improvemen
Hi,
The second DRBD release in June 2020. The reason is _not_
a regression in 9.0.23 but separating the bug-fix only .23
release form the bigger code change that the .24 release
introduces.
So far with all DRBD-9 releases, if you have two nodes primary
and they try to reconnect the abort establis
Hi,
only two weeks after the 9.0.23 release I am already starting the
next release cycle. The main reason is a change that was sitting
in a branch that I wanted to postpone after the 9.0.23
release. It is about making the detection of split-brain
situations more reliable.
So far with all DRBD-9 r
Hi,
it is out, finally! Upgrade from all 9.0.x versions to 9.0.23-1
recommended. A number of new tests were created while working on this
release.
9.0.23-1 (api:genl2/proto:86-116/transport:14)
* fix a deadlock (regression introduced in 9.0.22) that can happen when
new current UUID i
Hi,
The week after rc2 brought more reports and last-minute than expected. I
better give it another week to make sure it is of good quality when I
declare it as
final.
Here the changes relative to rc2:
9.0.23-0rc3 (api:genl2/proto:86-116/transport:14)
* fix a deadlock (regression introd
Hi,
Everything takes longer these days. Home-office, kids jumping around,
excuses, excuses...
I intend to make the final release very soon...
Here is the updated change-log:
9.0.23-0rc2 (api:genl2/proto:86-116/transport:14)
* fix a deadlock (regression introduced in 9.0.22) that can be
Hi,
A few fixes are waiting...
The regression is, of course, painful, and causes a lot of thinking
here about how we can get better at testing and verifying our changes.
There is a plan.
The area of handling IO errors needed a bit of love and quickly adding
and removing nodes with LINSTOR is an a
Hi,
this release has only bug-fixes. Most of the fixed issues are hard
to reproduce. Some are easy to reproduce, you just need the idea
to try exactly that.
Of course, also the test-suite was extended during this cycle.
9.0.22-1 (api:genl2/proto:86-116/transport:14)
* introduce locking
Hi,
It is just a number of smaller bug-fixes. No single one that I want to point
out.
Please help testing this release candidate.
9.0.22-0rc1 (api:genl2/proto:86-116/transport:14)
* introduce locking to avoid connection retries when UUIDs or
relevant flags change during the exchange
Hi,
from top of my head ZFS is special in that sense that it opens the backing
device only for a short amount of time.
I mean it does the eqivalent of an open(2) call on the backend block device
and than very soon after that does the equivalent of a close(2).
All other linux file systems open(2
Hi,
With the drbd-9.0.20 release we changed the kernel compat layer. We tested
carefully on the platforms most of our customers and users have:
RHEL/CentOs 6,7,8 and Ubuntu LTS 16.04 & 18.08. No we realize that
Debian was our blind spot. 9.0.20 does not work on Debian's 4.9 kernel:
It causes DRBD'
Hi,
With the drbd-9.0.20 release we changed the kernel compat layer. We tested
carefully on the platforms most of our customers and users have:
RHEL/CentOs 6,7,8 and Ubuntu LTS 16.04 & 18.08. No we realize that
Debian was our blind spot. 9.0.20 does not work on Debian's 4.9 kernel:
It causes DRBD'
Hi,
With this release the big news is the new way we deal with compatibility
for older linux kernels. It used to be a mechanism based of #IFDEFs that
tainted the main drbd code more and more.
Now the DRBD code is clean Linux-upstream compatible code. In order that
you can compile it against older
Hi,
Rc1 was broken on recent RHELs. The problem was that some compat
tests were failing for wrong reason, compile-time disabling some features.
Well, and then the running RHEL kernel was sending IO requests with
exactly those feature flags down to DRBD. Not good. That resulted in
DRBD sending stuf
Hi,
a lot has happened since the last release. We rolled in a new way how
kernel compatibility is handled. It is based on spatch/coccinelle
now. While it was painful to switch we hope that the compat layer
is now easier to understand while we move forward.
Here is a blog-post that describes that i
Hi,
drbd-9.0.19 final is out!
This release mainly contains fixes all over the board. The only new
feature is that it allows to mark a remote as "write only". I.e.
DRBD will never try to ship read requests to it. That is of course
usefull for DR purposes.
As a side note, the changes in this relea
Hi,
As usual this release brings a number of bug fixes, that alone is
reason enough to recommend the upgrade to every user.
Exciting are the two new features:
Quorum tie breaker
So far enabling quorum only made sense if you have a three way
redundant data set. With the new quorum tie breaker
Hi,
As usual this upcoming release brings a number of bug fixes,
that alone will be reason enough to recommend the upgrade to
every user.
Exciting are the two new features:
Quorum tie breaker
So far enabling quorum only made sense if you have a three way
redundant data set. With the new quor
Hi,
here is the drbd-9.0.17 release. It has a number of fixes, evenly
spread over the different use-case groups.
The improved resync speed should benefit all users that have
backing devices and networks that allow resync-rates above
800MByte/sec.
BTW, really new is the idea of having a container
Hi,
drbd-9.0.17 should be ready in a week from now. Please help
testing the pre-release
drbd
9.0.17-0rc1 (api:genl2/proto:86-114/transport:14)
* Fix UUID handling of a diskless primary that has not peer with
usable data may not touch the current UUID
* Fix resync-after dependencies;
Hi,
Final. No changes relative to rc1.
What really made us to release now, is fixing the regression introduced with
9.0.15. It was probably not triggered by many parties, because you can only
trigger it if you have requests in flight in excactly the moment a
timer comes by to check if the networ
Hi,
the list is shorter than with the last releases. I think this is good news.
What really made us to release now, is fixing the regression introduced with
9.0.15. It was probably not triggered by many parties, because you can only
trigger it if you have requests in flight in excactly the moment
Hi,
Today we publish the "addon-linstor" driver for OpenNebula that allows
OpenNebula to consume storage that is managed by LINSTOR. OpenNebula
is a Open Source cloud management system for IaaS clouds excels by its
ease of use.
It allows you to create multiple Datastores with different properti
Hi,
This is an upgrade ever drbd-9 user should follow. It has two important
fixes in the areas of
* handling IO errors reported by the backing device
handling of IO errors on the backend was completely broken since
drbd-9.0 including the recovery options like replacing a failed
disk. Whe
Hi,
Another DRBD release is coming up. Please help in testing the release
candidate.
This release contains one important fix for users that use DRBD9
in combination with LINSTOR or DRBDmanage and intend to do live
migrations of VMs.
What is in for everyone is the performance optimization, giving
Hi,
Yes, this is not a typo. DRBD gets ported to MS Windows. It is not
finished yet but you can definitely try it out. The .zip file contains
the usual .exe and a description fir with more information what to
expect.
It is supposed to work with Windows 7 (latest SP), Windows 10 and
Windows Server
Hi,
This is a strongly recommended update for all drbd-9.0.x users.
It contains serve fixes for for cases with multiple diskless
nodes. Without these fixes you can even see wrong data read back from
DRBD under complicated failure cases.
We slightly delayed the release to finish compatibility with
Hi,
This is the release candidate of an important drbd-9 update.
As soon as it is final it will be a strongly recommended update for
all uers. It contains serve fixes for for cases with multiple diskless
nodes. Without these fixes you can even see wrong data read back from
DRBD under complicated f
Hi,
Lately we had a to do more releases than we like. This time we became aware
of an issue that can cause that a drbd-resource can not be removed from
the DRBD kernel driver. We decided that this bug is so annoying that we need
to release an update immediately.
It is a race condition so, it does
Hi,
Relative to the rc1:
- one corner case with the quorum code fixed
- quorum-minimum-redundancy configure option in the drbd-utils
package
- Japanese translations for the man-pages!
still on track to tag the final before X-mas...
Oh, and for drbd-8.4, the only the compat code for upstrea
Hi,
In this development cycle we could look at a bug that was known
for some time but needed a few calm weeks to be understood and
fixed. So, no more resyncs that stuck near the end (~95%)
On the side of new features: The quorum feature got advanced a
lot and is not usable with pacemaker. It is v
Hi,
in short, this is a _must_ upgrade for every use that uses DRBD in
the SDS use-case. I.e. with diskless primaries. Two bugs where
identified that can cause DRBD to return wrong data to read
requests on the diskless primary node.
drbd
9.0.9-1 (api:genl2/proto:86-112/transport:14)
* f
Hi,
due to a number of regressions with the rc1 tag, and more "old" bugs,
we decided to do one more rc before this becomes a final release.
This release will be very important for users that have diskless,
primary nodes in their setups.
9.0.9rc2-1 (api:genl2/proto:86-112/transport:14)
--
Hi,
good news for the weekend, the next drbd-9 release will be ready
in one week.
Get your test equipment ready next week, and let us know if you
find anything suspicious.
This release will be very important for users that have diskless,
primary nodes in their setups.
drbd
9.0.9rc1-1 (api:
Hi,
finally it is there! With two rc's and surprising bug reports coming in late,
the release process has been a bit more challenging than usual. Maybe
the fact that I am doing this release while being in Beijing for the first
LinuxCon in China is also contributing it's share.
Due to the high num
1 - 100 of 247 matches
Mail list logo
