secfree opened a new pull request #3790: URL: https://github.com/apache/hadoop/pull/3790
### Description of PR 1. Support refreshing topology of clients/datanodes in NameNode 2. Usage: `hdfs dfsadmin -refreshTopology <ipAddr>` 3. Based on the patch `HDFS-11242.002.patch` at https://issues.apache.org/jira/browse/HDFS-11242 with the following changes - Rebase to `trunk` - Change from refreshing all nodes to a specific node, as the former holds the lock for a long time ### How was this patch tested? 1. Tested with UT 2. Tested in a real HDFS cluster ``` $ hdfs dfsadmin -fs hdfs://test -refreshTopology 1.2.3.4 Refresh topology successful at active-namenode/1.1.1.1:8020 for 1.2.3.4 Refresh topology successful at standby-namenode/1.1.1.2:8020 for 1.2.3.4 Refresh topology successful at observer-namenode/1.1.1.3:8020 for 1.2.3.4 ``` ### Note 1. It is also ok to pass in a hostname instead of an IP address, as keys of the cache may be hostnames sometimes. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
