Details
-
Improvement
-
Status: Closed
-
Major
-
Resolution: Fixed
-
1.1.0-Ducc
-
None
Description
Broken nodes can cause ssh hangs, e.g. if NFS is broken. Check_ducc and start_ducc should handle this.
Proposed design: it's nearly impossible to stop the ssh without a ctl-c -like interrupt if the remote work is stuck in NFS. check_ducc and start_ducc will catch the ctl-c, terminate any stuck ssh's, and report on the nodes that didn't return correctly.