Hide Forgot
This is a feature of 'distribute' than an error. We don't allow rmdir() to succeed when a node is down.
With a two dht setup and testing with master commit-id "73eca3be5c5ccc71bbad934338c1ef58ed37c483" and also with "f3c0281d9b4654888a5525b443f1e6090d874725" both fails. # ls -l total 40 drwxr-xr-x 2 root root 4096 Jul 28 23:57 b drwxr-xr-x 2 root root 4096 Jul 28 23:57 c drwxr-xr-x 2 root root 4096 Jul 29 00:01 cd drwxr-xr-x 2 root root 4096 Jul 28 23:57 e drwxr-xr-x 2 root root 4096 Jul 28 23:55 i #killed one process. # rmdir b rmdir: b: Transport endpoint is not connected # rmdir c rmdir: c: Transport endpoint is not connected # ls -l total 40 drwxr-xr-x 2 root root 4096 Jul 28 23:57 b drwxr-xr-x 2 root root 4096 Jul 28 23:57 c drwxr-xr-x 2 root root 4096 Jul 29 00:01 cd drwxr-xr-x 2 root root 4096 Jul 28 23:57 e drwxr-xr-x 2 root root 4096 Jul 28 23:55 i