Hi Leo
On Wed, Nov 22, 2017 at 10:15:10PM +0100, Leo wrote:
> Hi Bastian
> Thank you for your prompt reply. I would like to talk about it below.
> > The -E option is common to all current operation systems as it is
> > specified by POSIX. As this script not only runs on Linux, using GNU
> > specific options is not the way to go. I also found no supported Linux
> > distribution without support for it.
> I tell SuSE support. Their sed under SLES11 or SLES12 knows no option -E.
Somehow I doubt that. SLES12 comes with sed 4.2.2, which is years after
the option was introduced. SLES11 may be a different beast.
> >I would be more concerned about a stage2 that runs for over an hour.
> The problem is that e.g. my existing proxy has a timeout of 120 and rsync
> 3600 seconds. The program rsync calculates its keep alive transmissions
> according to the following calculation: timeout / 2. This results in a
> timeout of 3600 seconds keep alive packets every 1800 seconds. However, the
> proxy ends the connection after 120 seconds of inactivity. So while ftpsync
> checks in Stage2 what files it can delete (no traffic),the proxy's timeout
> fails and terminates the socks connection. If the rsync wants to access the
> remote station again, it first notices that it is offline and ends with an
> error. But if I synchronize the timeouts of the proxy and rsync, the keep
> alive packets will be sent on time and the connection will remain
> open.Another note, the proxy is not under my administration and I can not
> change that.
I'm reluctant to add options just so there are options. If a change in
this option does not have side-effects, we can change it for everyone.
But this I have to asses further.
Bastian
--
All your people must learn before you can reach for the stars.
-- Kirk, "The Gamesters of Triskelion", stardate 3259.2
