On 07/22/10 08:33, Will Fiveash wrote:
Let me put it another way; when I originally ran onu I followed the example from onu(1):# onu -d /path/to/my/ws/packages/`uname -p`/nightly -t new- nightly-be This failed badly as I noted in the start of this thread. If this was because the pkg publishers were: # pkg publisher PUBLISHER TYPE STATUS URI on-nightly (preferred) origin online http://alton.central:13005/ on-extra origin online http://alton.central:13006/ opensolaris.org (non-sticky) origin online http://ipkg.sfbay/dev/ opensolaris.org (non-sticky) mirror online http://aus-install.central:180/ extra origin online http://ipkg.sfbay/extra/ extra mirror online http://aus-install.central:180/extra/
Yes I think onu -d would fail from there if depotd were not running for the first two origins above.
then why didn't onu either: - Replace the publishers with ones that made sense given the command args.
The above could be a valid configuration. It's not the usual/expected for regular onu use, but it is quite legal to have multiple distinct origins for the same publisher. ie you have an origin for publisher on-nightly above - how does onu know whether you want to replace that with the one it normally installs, or configure it to search before or after yours.
or: - Warn me that my config was broken. ?
I think that it should at least warn, however. I suspect that 9 times out of 10 the above configuration will not be a delicately arranged hierarchy of origins but simply something waiting to bite you. Gavin _______________________________________________ on-discuss mailing list on-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/on-discuss
