On Feb 11, 2008 9:47 PM, Sanjay Nadkarni <[EMAIL PROTECTED]> wrote: > Shawn Walker wrote: > > On Feb 11, 2008 8:47 PM, Shawn Walker <[EMAIL PROTECTED]> wrote: > > > >> On Feb 11, 2008 8:39 PM, Tom Mueller <[EMAIL PROTECTED]> wrote: > >> > >>> Do you have localhost pointing to 127.0.0.1 in your hosts file? > >>> > >> Ah, no, not explicitly anyway. > >> > > > > Actually, there is an explicit entry set for my hostname in 127.0.0.1. > > > > I've added one for localhost too though I wonder if it will stick > > since I'm using dhcp for my network interface. > > > > > >>> Do you have http_proxy set in your environment? (it shouldn't be set when > >>> running the tests). > >>> > > > > No; no proxies. > > > > > >>> Do you have some other process running on port 8000? > >>> > > > > No. > > > > > >>> Yes, the tests do start a depot server. I've found that certain types of > >>> failures can sometimes leave a depot server running and that making sure > >>> there are no depot servers running before starting the test will produce > >>> better results. > >>> > >>> I've not found that the tests require an ON tree. > >>> > > > > Ah, okay. I just saw some odd variables in some of the test scripts > > that look for ../../proto. > > > If go to src/ and type make install, it creates a proto area and and > SVR4 pkg SUNWipkg. I wonder if the tests depend on proto being there. > > When I run make test I see a bunch of FAILED ASSERT > ASS 1: pkg status should fail in an empty image > PASS 2: Send empty package [EMAIL PROTECTED], install and uninstall. > *** case 3: expected exit status: 0, saw 1. > *** FAILED ASSERT: Send package [EMAIL PROTECTED], containing a directory and > a > file, install, search, and uninstall. > *** SUB-ASSERT: pkg status -a, install foo, verify should succeed > > However running it as root (i.e. sudo make test) yields better results.
Yes, after I figured out that I needed to do a "make install" first before I could run make test, it all came together. I'm not quite used to the proto are approach yet. Also, as you mentioned, the results were far better when run as the privileged user (pfexec shell in my case) than when run as my own user. Some tests still fail, but it looks like "known issues." Thanks again. Cheers, -- Shawn Walker, Software and Systems Analyst http://binarycrusader.blogspot.com/ "To err is human -- and to blame it on a computer is even more so." - Robert Orben _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
