Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[v23.2.x] Backport of #13964 #14739 #14803 #15049 #15032 #15139

Merged
merged 7 commits into from
Nov 28, 2023

Conversation

mmaslankaprv
Copy link
Member

@mmaslankaprv mmaslankaprv commented Nov 27, 2023

Backport of PR #13964
Backport of PR #14739
Backport of PR #14803
Backport of PR #15049
Backport of PR #15032

Fixes: #13992
Fixes: #15080
Fixes: #14829
Fixes: #14820

When member is added to the cluster `cluster::configuration_manager`
should be immediately aware of it instead of waiting for the first
status update.

Wired up cluster member added notification to update cluster
configuration reconciliation status with initial state for joining
member.

Fixes: redpanda-data#13497
Fixes: redpanda-data#13503

Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit 57c8d7a)
Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit cb4e607)
Recovery and replicate stms are not synchronized. It may be the case
when both of stms are active at the same time that the same batch is
delivered to the follower twice. In general this batch duplication is
harmless as Raft is not vulnerable for messages redelivery but it may
cause unnecessary truncation and latency increase.

Added a check validating expected log end offset right before sending
recovery append entries request. This will prevent sending the same set
of batches twice to the follower.

Fixes: redpanda-data#14413

Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit 7bb54d3)
Some of the operators were incorrectly declared while all of them were
lacking definition

Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit f65999a)
Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit b8f63a7)
When `rm_stm` is stopped it should release all used resources including
metrics name. This will allow partition to be recreated if required.

Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit 4a82177)
Previously validation was failing when `_kafka_start_offset` was
overridden with the same offset. Since re-setting the offset to the same
number doesn't change anything we may relax the validation here.

Signed-off-by: Michal Maslanka <michal@redpanda.com>
(cherry picked from commit b1c1d01)
@mmaslankaprv mmaslankaprv changed the title [v23.2.x] Backporting [v23.2.x] Backport of #13964 #14739 #14803 #15049 #15032 Nov 28, 2023
@mmaslankaprv mmaslankaprv marked this pull request as ready for review November 28, 2023 06:31
@mmaslankaprv mmaslankaprv merged commit c01a044 into redpanda-data:v23.2.x Nov 28, 2023
26 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants