Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
fix(admin): refresh cluster metadata while adding storage node
It fixes a transient error caused by fetching cluster metadata too lately while handling AddStorageNode. This PR lets the AddStorageNode RPC handler fetch the cluster metadata forcibly and wait for them to be updated. Although it isn't a critical error and can be resolved through a simple retry, it is resolved quickly by fetching the cluster metadata. Error example: ``` --- FAIL: TestVarlogAppend (5.38s) controller.go:62: Error Trace: /home/runner/work/varlog/varlog/tests/ee/controller/controller.go:62 /home/runner/work/varlog/varlog/tests/ee/cluster/local/cluster.go:238 /home/runner/work/varlog/varlog/tests/ee/cluster/local/cluster.go:63 /home/runner/work/varlog/varlog/tests/ee/ee_test.go:32 Error: Received unexpected error: exit status 255 Test: TestVarlogAppend Messages: /home/runner/work/varlog/varlog/bin/varlogctl storagenode add --storage-node-id 1 --storage-node-address 127.0.0.1:20001 --admin 127.0.0.1:9093: rpc error: code = Unavailable desc = add storage node: call again ```
- Loading branch information