Cherry-picks for 2.10.23-RC.3 #6129
Draft
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…nment If we are reassigning a stream assignment to a different placement, i.e. due to insufficient resources or another placement error, then don't process any further stream assignment results until the new assignment is processed. Otherwise we might generate multiple new assignments unnecessarily which could delete/recreate/update the stream multiple times, potentially on different peer sets. This should also de-flake `TestJetStreamSuperClusterConcurrentOverflow`. Signed-off-by: Neil Twigg <neil@nats.io>
Signed-off-by: Maurice van Veen <github@mauricevanveen.com>
In complex setup, a message produced from a cluster that had queue interest from leafnodes (either hub or spoke) would sometimes not deliver a message if the interest was a leafnode that had the interest on behalf of a gateway. In the setup described in the issue this PR fixes, "Cluster B" may have picked "Cluster C", but that cluster does not have local queue interest, only the leafnode interest from "Cluster B", and would pick a LEAF connection to this cluster, but then suppress the message since it came from "B" so "C" cannot send it back there. But picking a queue sub for "B" in "C" would then prevent the message to be delivered to the gateway "D". Resolves #6125 Signed-off-by: Ivan Kozlovic <ivan@synadia.com>
…g of the hdr for each additional addition. Signed-off-by: Derek Collison <derek@nats.io>
Previously this was always firing at the exact interval. Then #6041 changed this so that the first firing was anytime between 1s and the interval to spread out different stores. This tweaks it once more to anytime between interval/2 and interval. It will still spread things out a bit but hit the benchmarks a bit less badly. Also go back to `math/rand` instead of `math/rand/v2` due to availability in Go versions that we still care about for 2.10.x. Signed-off-by: Neil Twigg <neil@nats.io>
Signed-off-by: Neil Twigg <neil@nats.io>
This reduces allocations in the NRG `sendRPC` and `sendReply` functions by promoting buffer reuse. Signed-off-by: Neil Twigg <neil@nats.io> --------- Signed-off-by: Neil Twigg <neil@nats.io>
Signed-off-by: Maurice van Veen <github@mauricevanveen.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Includes the following:
mset.noInterestrequires write lock #6122n.Appliedfor consumer when shutting down #6133Signed-off-by: Neil Twigg neil@nats.io