misunderstanding in: "STR #3181: polled printers are "not published""

Timo Strunk Timo.Strunk at gmx.de
Fri Aug 7 02:31:05 PDT 2009


Hi everybody,

I'm not entirely sure, but I guess there was a misunderstanding before closing  bug STR #3181.

A similar "bug" was discussed also here:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=520396

3181 says that it should be impossible to re-publish a printer published from a different server.

So, a configuration like:
"Printer1 on Server1, published
Printer1 -> Server1 -> Server2 published
should be forbidden, which means that there cannot be a Server3 listing Printer1 through Server2. Server3 has to access Printer1 through Server1."
This makes sense; hence the bug should be closed.

The bug described on the debian bug report however wants the following configuration:
Printer1 on Server1, published
Printer1 -> on Server1 listed on Server2 unpublished.
Client1 (ServerName Server2) has access to all printers on Server2 and remote printers on Server1.
This should be possible (and was possible before, too), otherwise there would be no other choice to access printers from two different printservers on a client apart from setting up a complete own cups server with two Poll addresses on each client machine, which should access both printers.

Currently the behaviour is like this:
Printers are listed on Server2, local ones published, remote ones unpublished.
Clients with ServerName Server2 list all Printers on Server2 (published and unpublished). Printing on local printers on Server2 succeeds. Printing on remote printers on Server2 fails with client-error-not-authorized.

I hope I didn't waste everybody's time here, because the misunderstanding is on my side.

Thanks for reading,
Timo





More information about the cups mailing list