Option --retries

[Ole Tange]

I have experienced that a job on a compute host (--sshlogin) sometimes
fail randomly. It can be due to network error or missing libraries for
the command or out-of-memory or some other non-predictable error. In
other words: the command may succeed on another host or on the same
host but at another time.

So I have been thinking of implementing --retries:

--retries n
If a job fails, retry it on another computer. Do this n times. If
there are fewer than n computers in --sshlogin GNU parallel will
re-use the computers. This is useful if some jobs fail for no apparent
reason (such as network failure).

First: Does it make sense to have this option?

Second: Do you understand the description of the option? If not: Can
you rephrase it?


/Ole - who is going to speak at FSCONS 2010-11-07T11:30
http://www.fscons.org/program