Brian was disturbed when a routine, automatically-scripted invocation  
of darcs stopped with

darcs failed:  Not a repository:
source@allmydata.org:darcs/tahoe/trunk ((scp) failed to fetch:
source@allmydata.org:darcs/tahoe/trunk/_darcs/inventory

(We don't use darcs-1-format repositories which have an "inventory"  
file.)

After calling me for help and investigating a bit, we tried again and  
this time it worked.  Our best guess is that a transient network  
failure is responsible.

Zooko

On Mon, Mar 09, 2009 at 09:51:19AM +0000, Zooko wrote:
> After calling me for help and investigating a bit, we tried again
> and this time it worked.  Our best guess is that a transient network
> failure is responsible.

Unless a bug can be reproduced reliably, it's pretty much impossible
to track it down.  How do you feel about either closing this ticket,
or flagging it as "more info needed"?

Note that this may just be another instance of issue1036 (possible confounding
factors: hashed-repo, SSH).

I think we need somebody to work out a means of reliably recreating the sort of
network failure in mid-darcs that would lead to this.

I claim that this is the same issue as issue1036