Asias He

gossip: Make boot process faster and safer

81

gossip: Make boot process faster and safer During boot up, we need to wait for gossip in multiple places: 1) Wait for gossip to settle to enable features. See commit...

repair: Yield in repair_service::do_decommission_removenode_with_repair

3

When walking through the ranges, we should yield to prevent stalls. We do similar yield in other node operations. Fix a stall in 5.1.dev.20220724.f46b207472a3 with build-id d947aaccafa94647f71c1c79326eb88840c5b6d2 ``` !INFO |...

17ms stalls in reclaim_from_evictable

12

Run from: https://groups.google.com/a/scylladb.com/g/scylla-perf-results/c/7Y2EFr21U-4 Scylla version 5.1.dev-0.20220711.7e2d2cf1c18d with build-id 8d31af83d550feaccdcf3d243d522685440315a2 ``` perf-latency-nemesis-perf-v12-db-node-b6e3deb1-1/messages.log:2022-07-11T14:42:21+00:00 perf-latency-nemesis-perf-v12-db-node-b6e3deb1-1 !INFO | scylla[6623]: Reactor stalled for 17 ms on shard 4. Backtrace: 0x4b7b652 0x4b7a2b0 0x4b7b560 0x7ff414e6ca1f 0x198072c 0x15d99e5...

streaming: Add finished percentage metrics for node ops using streaming

5

We have added the finished percentage for repair based node operations. This patch adds the finished percentage for node ops using the old streaming. Example output: scylla_streaming_finished_percentage{ops="bootstrap",shard="0"} 1.000000 scylla_streaming_finished_percentage{ops="decommission",shard="0"} 1.000000...

storage_service: Prevent removed node to rejoin in handle_state_normal

9

- Start n1, n2, n3 (127.0.0.3) - Stop n3 - Change ip address of n3 to 127.0.0.33 and restart n3 - Decommission n3 - Start new node n4 The node...

storage_service: Reject decommission if nodes are down

6

- Start n1, n2, n3 - Apply network nemesis as below: + Block gossip traffic going from nodes 1 and 2 to node 3. + All the other rpc traffic...

storage_service: Reject to bootstrap new node when node has unknown gossip status

8

- Start a cluster with n1, n2, n3 - Full cluster shutdown n1, n2, n3 - Start n1, n2 and keep n3 as shutdown - Add n4 Node n4 will...

This forks from https://github.com/scylladb/scylla-monitoring/issues/1788 where we want to see the current progress of the streaming.

repair: Add ranges option support for tablet repair

6

The management tool, e.g., scylla manager, needs the ranges option to select which ranges to repair on a node to schedule repair jobs. This patch adds ranges option support. E.g.,...

Provide a way to query if a table is dropped for streaming and repair

6

Consider: - n1, n2, n3 in the cluster - n1 streams data to n2 / n1 runs repair with n2 - a table t1 is dropped during streaming / repair...