Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
MB-54959 - adjust log level for watcher on valid closes
bg: we create metadata provider object on all indexer nodes periodically this starts watcher for each node. upon cleanup, peer pipe logs an error that connection is terminated but this is expected behaviour. this error log is creating alerts in customer log monitors exp: if we are killing the peer pipe connection on a valid condition don't log it as error stubs: in my experiments, when we kill a server (with -9/-15 flag) the connection is still gracefully terminated and go returns io.EOF because of this we cannot diff between indexer crash and conn close at gometa level. metadata provider will still be aware if conn was expected to die or closed abruptly assumptions: - Change-Id: I015ac0e036a1ad07ffacb20d3ef760149047d45d
- Loading branch information