On 09/07/2011 05:09 AM, Carsten Otto wrote:
On Wed, Sep 07, 2011 at 11:07:43AM +0200, Markus Knauer wrote:
Without being a webmaster there... it looks like the site.zip file contains
the nightly builds of the EGit team and is replaced by a new one after every
build. The last replace happened at about 4am server time (~8am UTC) and I
guess that you started your rsync *before* the file has been replaced.
During the sync process it got replaced by a new one and therefore you see
the stale NFS handle.
I got that error multiple times, over the course of several days.
Like Markus has said, that project runs a build frequently, and that
site.zip file changes every time.
I believe your rsync job and the project's build job are
coincidentally at the same time, hence the reason you get the stale
file handle. You can use incremental rsync to reduce the errors, or
simply --ignore-errors like everyone else does...
Apologies for the delay in replying -- Monday was a holiday for most
of North America, and it usually takes us about a day to recover
from a holiday ;)
Denis
--
|