Kamil Braun

A potential alternative solution to introducing barriers etc., would be to bring back "wait for gossip to settle" at the end of node startup procedure, before it says "initialization completed"....

@mykaul in that case please open in scylla-dtest. We won't fix it on Scylla side for 5.4 (no point in doing that), but we can maybe adjust the test to...

My comments on the old issue: - https://github.com/scylladb/scylladb/issues/17545#issuecomment-2109869522 - https://github.com/scylladb/scylladb/issues/17545#issuecomment-2109886465 Summarizing, we should compare if it's a regression or not, it could indeed be as @aleksbykov suggested that the instance...

Marking as release blocker for now, before checking the above, but there's a chance we will be able to take it off and/or close the issue quickly.

From the other issue: > > Also, how many shards is this using? Is it `--smp 2` (IIRC the default with dtest)? Is it running with `--overprovisioned`? > yes: >...

> Job https://jenkins.scylladb.com/job/scylla-staging/job/abykov/job/Dtest/job/master-dtest-with-raft/272/ also failed. also around 71 node. not all logs are collected yet. But also there is coredumps. Ok, so there's a chance there's no regression. Let's confirm...

> Increased and job has been rerun. 68 nodes were provisioned successfully Are you saying it failed at node 69?

> Issue reproduced with 5.4. Looks very similar abort: So it's most likely as you said -- the instance is too weak to deal with that many nodes. Let's see...

@aleksbykov we should compare to 5.4 for that larger instance as well.

> Moving to 6.1, just to remove it from 6.0 blockers list for the time being. Can you explain to me this decision? @mykaul if it's not a blocker for...