imbajin commented on code in PR #2952: URL: https://github.com/apache/hugegraph/pull/2952#discussion_r2904660900
########## hugegraph-server/hugegraph-dist/src/assembly/static/bin/wait-storage.sh: ########## Review Comment: There’s one more thing that might be worth looking at: in `wait-storage.sh`, the PD REST address is derived from `pd.peers`, but in practice the script only checks the first peer. This works fine in the current compose example, but it still feels a bit fragile in a multi-PD setup. For example, during a fairly common rolling restart, a single-node upgrade, or when one PD node simply comes up more slowly than the others, if the first PD in the list is temporarily unavailable, the script will keep treating PD as not ready even if the other two PD nodes are already up and working normally. So I think a more reliable approach here would be either: - explicitly passing a PD REST endpoint, or - trying the peers from `pd.peers` in order instead of checking only the first one This may not cause an immediate problem in the default compose setup, but from a multi-PD robustness perspective, I think it’s still worth noting. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
