From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0) Description of problem: rhr2-externals-0.9-11 rhr2-tests-0.9-11 rhr2-0.9-11 on all pSeries boxes(e.g. p630, p650), the first scsi tape is identitied as "st" by rhr2 hardware detection(in hardware.conf). When we choose to test "Tape", it gives us the following message in error.log: WARNING: does not appear to be valid /dev/st: No such file or directory In fact, st should be "st0", then the test is OK. We noticed that this tape test call "kudzu -q" first, in the generated "hwconf" file, tape device is "st". Version-Release number of selected component (if applicable): rhr2-tests-0.9-11 How reproducible: Always Steps to Reproduce: 1. install rhr2 2. run redhat-ready, pembroke and discovery 3. run tape test Additional info:
I can work around that in the detection sripts; but this should as quickly as possible be opened up as a kudzu bug (our data is based on the kudzu output so we really should fix the root cause) expecially since rhn and the installer rely on the output of kudzu as well (I'm not certain the true impact it would/will have on those)
Hi Rob, bug 107361 is opened against kudzu, and the fix provided is just fine. Thanks!
Excellent, from how I read the kudzu bug; it appears the kudzu update isn't going to make the Gold CD; but instead would be released as either an errata and/or on Update 1; this means in the interim you'll need to edit /etc/rhr/hardware.conf and correct the TAPE line since I am not planning on writting a work around I'll just have to remove later. Does this sound acceptable?
That's fine, Rob. We'll reverify it when QU 1 comes out.
Hi Rob, fix is included in QU1. So it can be closed.