Wait until caps renegotiation completes on node upgrade - #33
Open
hmmr wants to merge 3 commits into
Open
Conversation
added 3 commits
July 24, 2026 02:02
use rt:get_backends/0 instead, which was written for this purpose
hmmr
changed the title
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When a node with reduced capabilities joins the cluster, we should take care accordingly to reduce the set of supported capabilities in the cluster as well. This was done, in 2016, in a commit that was reverted soon thereafter, because it introduced "race conditions when nodes restart and gossip capabilities".
These "racing conditions", apparently, refer to a temporary state when nodes report inadequate capabilities, as when, for example, triggered here.
A mirror condition exists when the last node with reduced capabilities relative to the rest of the cluster, is upgraded. The result, after capability gossip and renegotiation, is the full set of capabilities restored cluster-wide.
This PR adds a wait, in the only test that regressed following un-reversion of the "filter caps on downgrade" commit (verify_dt_upgrade), until, when an upgraded node is starting, the capabilities are renegotiated.
Related PR in riak_core is OpenRiak/riak_core#63.
(Also included in this PR is an unrelated change that uses direct
rt:get_backends/0instead of trying to extract backend info from riak_test metadata, where such info is no longer found.)