ControlMaster cwd is not /

raf ssh at raf.org
Fri Apr 9 13:30:36 AEST 2021


On Thu, Apr 08, 2021 at 12:16:42PM +0200, Bert Wesarg <bert.wesarg at googlemail.com> wrote:

> Dear all,
> 
> see
> 
>     https://bugzilla.mindrot.org/show_bug.cgi?id=1902
> 
> for one of the first reports and also a reason why it is currently not accepted.
> 
> Best,
> Bert

I must be missing something. I can't see where on that
page a reason is given, but I'll take your word for it
that there is one. :-) If it has something to do with
the comment about relative paths for known hosts files,
I don't see that as a problem. By default, the known
hosts file has a non-relative value, i.e.
~/.ssh/known_hosts. So, if that's the reason, it seems
to be a fairly weak one. In the probably rare case that
someone does manually override the default known hosts
file path, it could be documented in the manpage that
it should be set to an absolute path so as not to cause
problems with ControlMaster. That would be better than
leaving the ControlMaster problem in place for everyone
that uses ControlMaster.

cheers,
raf

> On Thu, Apr 1, 2021 at 12:48 AM Steffen Nurpmeso <steffen at sdaoden.eu> wrote:
> >
> > Hello!
> >
> > raf wrote in
> >  <20210331075157.sgfbkvofdx6dgiyi at raf.org>:
> >  |On Wed, Mar 31, 2021 at 12:33:43AM +0200, Steffen Nurpmeso <steffen at sdao\
> >  |den.eu> wrote:
> >  |> I just stumbled over the fact that the ControlMaster process does
> >  |> not do "cd /" but remains where the ssh client was invoked.  Is
> >  |> this intentional, or should i open a bug report on that?
> >  |>
> >  |> [snip]
> >  ...
> >  |I reported this too a little while ago and there was no
> >  |interest. But it is a pain when your cwd is a USB stick
> >  |that you want to umount. I've created a pull request
> >
> > Yeah, and encrypted volumes (in my case).
> >
> >  |that fixes it:
> >  |
> >  |  https://github.com/openssh/openssh-portable/pull/239
> >
> > That is great!  I had not looked at github.  (That pull i could
> > not have found by then?)
> >
> >  |But it's failing checks on 4 Ubuntu kitchensink/sk
> >  |versions.
> >  |
> >  |This page shows the error message:
> >  |
> >  |  https://github.com/openssh/openssh-portable/actions/runs/704178803
> >  |
> >  |  No files were found with the provided path:
> >  |  regress/*.log regress/valgrind-out/. No artifacts
> >  |  will be uploaded.
> >
> > I am not at github so cannot see the log, but since configure
> > failed also locally, a "autoreconf" seems to fix it.  After that:
> >
> >   make[1]: Leaving directory '/tmp/x/openssh.tar_bomb_git/regress'
> >   unit tests passed
> >   echo all tests passed
> >   all tests passed
> >
> > (8.5p1 plus your patch.)
> > It would be nice if that patch made it in.
> >
> > Thanks, and Ciao from Germany,
> >
> > --steffen
> > |
> > |Der Kragenbaer,                The moon bear,
> > |der holt sich munter           he cheerfully and one by one
> > |einen nach dem anderen runter  wa.ks himself off
> > |(By Robert Gernhardt)
> > _______________________________________________
> > openssh-unix-dev mailing list
> > openssh-unix-dev at mindrot.org
> > https://lists.mindrot.org/mailman/listinfo/openssh-unix-dev
> _______________________________________________
> openssh-unix-dev mailing list
> openssh-unix-dev at mindrot.org
> https://lists.mindrot.org/mailman/listinfo/openssh-unix-dev


More information about the openssh-unix-dev mailing list