hdfs-deprecated
hdfs-deprecated copied to clipboard
[DEPRECATED] This project is deprecated. It will be archived on December 1, 2017.
Handle failure of restarting both NN's on new nodes in the case both machines go down. This will need to be done by restoring from a backup image.
Handle case of a journal node relaunching on a new machine. We will need to reinitialize the journal nodes or we can restore from a snapshot backup.
- Framework reregisters after all slaves have reregistered - Framework reregisters before some/all slaves reregister (needs reconciliation retries) - Framework reregisters, but some slaves never reregister (failed) - Mesos master...
I noticed this in the log. The framework stopped attempted to bootstrap the 2nd namenode. It looks like if there is a task lost then running it doesn't update livestate...
Can launch all JNs, or all NNs, or all DNs together if the master offers resources on multiple slaves. Otherwise, HDFS must wait for the next offer cycle before launching...
Use disk resources whenever possible by making the appropriate resource requests and using the appropriate offers.
After #64, if the archive containing the HDFS Binary is not on the server, `http://127.0.0.1:8765/hdfs-mesos-0.0.2.tgz` will return the `hdfs-site.xml` file. Is there a way to check the existence of the...
Currently the framework uses the hadoop/hdfs 2.4.0 dependencies in the pom, but the `build-hdfs` packaging script packages up hadoop-2.3.0-cdh5.1.0 from cloudera. 1. The pom dependency and packaged binary versions should...
Document the procedure for migrating from HDFS running separately to HDFS on Mesos without data loss or other issues. Testing required.