Adrián, Kickstarting CentOS worked fine. I did not add any rhn* tools to the base channel so at the end of the installation two commands that are in ks.cfg fail: rhnreg_ks and rhn_check. After installing the rpm's manually (as documented in https://fedorahosted.org/spacewalk/wiki/RegisteringClients) I was able to register the CentOS with Spacewalk.
Regards, Willem. On Thu, Nov 20, 2008 at 3:02 PM, Willem Bos <[EMAIL PROTECTED]> wrote: > Adrián, > > Thanks. I did check the entitlements when I created the activation key but > then I changed the value of the 'key'-field to something I could > remember/recognize. The logging however did not reflect this change. So - > just now - I deleted the key, created a new one and associated it with the > kickstart profile. Requesting the kickstart file from my browser works > (jay!) I'll try kickstarting a VM in a moment... > > Willem. > > > On Thu, Nov 20, 2008 at 10:57 AM, Adrián Márques <[EMAIL PROTECTED]>wrote: > >> Hopefully this could save you some time: >> >> Go and edit your activation key, (from heart) there should be 4 checkboxes >> were you mark if the system associated with that key have the entitlements >> provisioning, monitoring, virtualization or virtualization platform. Make >> sure 'provisioning' is checked. That should take care of your >> PermissionException. >> >> I still haven't had the time to get kickstarting working myself, so I >> could be dead wrong here, but in order to kickstart systems you'll need the >> rhn-kickstart package, which (for RHN) seems to be located in the Tools >> channel. I think you should also need rhnlib, and that you can get them both >> at the location where all spacewalk client packages are hosted. I don't know >> if these packages should be placed in a particular channel or just pushed >> into the base one associated with your kickstart profile. >> >> Please let me now if you fixed your problem and how, since I'm also >> interested in setting up CentOS kickstarting. >> >> I hope this helps somewhat. >> >> Regards. >> >> Adrián. >> >> >> Willem Bos escribió: >> >>> Hi James, >>> >>> I just checked /var/log/tomcat5/catalina.out and it contains errors : >>> >>> 2008-11-18 15:36:03,700 [TP-Processor6] ERROR >>> com.redhat.rhn.manager.kickstart.KickstartScheduleCommand - Tools channel >>> not found! This means we can't find the rhn-kickstart package. >>> >>> 2008-11-18 15:36:03,707 [TP-Processor6] ERROR >>> com.redhat.rhn.common.errors.PermissionExceptionHandler - Permission Error >>> com.redhat.rhn.common.security.PermissionException: The activation key >>> '1-1b5ded8329c9661f44cfc5656fca7a50' needs provisioning capabilities to be >>> able to facilitate the config channel functionality >>> >>> Now, I'm new to Spacewalk so I'll have to check te manual. I'll probably >>> get to that tomorrow. >>> >>> Regards, >>> Willem. >>> >>> On Wed, Nov 19, 2008 at 7:21 PM, James Fillman <[EMAIL PROTECTED]<mailto: >>> [EMAIL PROTECTED]>> wrote: >>> >>> I'm having the same problem too. I've got Spacewalk 0.3 running on >>> a CentOS 5.2 server. I was successfully kickstarting and >>> registering centos-52-i386 clients 2 days ago. As soon as I >>> created a new CentOS Org, setup the necessary channels, installed >>> the kickstart tree, created an activation key, and updated the >>> kickstart profile, I can no longer pull the down a kickstart file >>> from ANY of my orgs. I keep getting the HTTP 401error. It's not a >>> DNS issue as I'm using the FQDN. >>> >>> >>> [EMAIL PROTECTED] ~]$ wget >>> >>> http://spacewalk.cucbc.com/kickstart/ks/org/3x7a18626c82422f653f552abb32918f20/label/centos-32 >>> >>> --2008-11-19 10:04:20-- >>> http://spacewalk.cucbc.com/kickstart/ks/org/3x7a18626c82422f653f552abb32918f20/label/centos-32 >>> >>> Resolving spacewalk.cucbc.com... 172.17.109.30 <http://172.17.109.30> >>> >>> Connecting to spacewalk.cucbc.com >>> <http://spacewalk.cucbc.com>|172.17.109.30 >>> <http://172.17.109.30>|:80... connected. >>> >>> HTTP request sent, awaiting response... 401 Unauthorized >>> >>> Authorization failed. >>> >>> [EMAIL PROTECTED] ~]$ >>> >>> >>> I'm currently scratching my head over this too. Like I said, this >>> was all working great before I went to add another Org. Now that I >>> have, I can't pull down any ks files. >>> >>> >>> Any more idea's?? >>> >>> >>> >>> Regards, >>> >>> James Fillman >>> Systems Architect >>> >>> Central 1 Credit Union >>> 1441 Creekside Drive >>> Vancouver, British Columbia, Canada >>> V6J 4S7 >>> >>> [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> >>> Tel: 604.730.6328 >>> >>> >>> ------------------------------------------------------------------------ >>> >>> *From:* [EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]> >>> [mailto:[EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]>] *On Behalf Of *Willem Bos >>> *Sent:* Wednesday, November 19, 2008 9:59 AM >>> >>> *To:* [email protected] <mailto:[email protected]> >>> *Subject:* Re: [Spacewalk-list] "bad HTTP response code" when >>> kickstartingCentOS 5.2 >>> >>> >>> Hi Adrián, >>> >>> Thanks for your response. The fqdn is present in the host file of >>> the Spacewalk server and test failed there. During the >>> installation of the client there's no shell on VT2 so I'm unable >>> to perform the test in that stage. >>> >>> Thanks again, >>> Willem. >>> >>> On Wed, Nov 19, 2008 at 1:39 PM, Adrián Márques >>> <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> wrote: >>> >>> I can't provide any insight into your problem, but I'd recommend >>> checking out whether it actually is the fqdn before taking the >>> time to set up a DNS server. >>> Just add your server's fqdn to /etc/hosts in your client machine >>> and see if a request addressed to the fqdn works or not. If not, >>> mail back and someone else might be able to help you out. >>> >>> Good luck. >>> >>> Adrián. >>> >>> [EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]> escribió: >>> >>> >>> Hi Willem, >>> >>> I am unsure about the checks that spacewalk does when receiving a >>> request for a kickstart. Login failed for me if I did not have a >>> fqdn. I think I might have used ip on a previous satellite install >>> I worked on but I dont have access to it anymore to test that. >>> >>> You are using a enterprise level tool that just works better if it >>> has the supporting services like dns working. It will take you >>> 30mins to setup a dns server somewhere on a box and at least help >>> eliminate dns related issues. >>> >>> Regards >>> >>> *From:* [EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]> >>> [mailto:[EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]>] *On Behalf Of *Willem Bos >>> *Sent:* 18 November 2008 16:18 >>> *To:* [email protected] <mailto:[email protected]> >>> *Subject:* Re: [Spacewalk-list] "bad HTTP response code" when >>> kickstarting CentOS 5.2 >>> >>> Gerhardus, >>> >>> Thanks. But what about wget, shouldn't that work? >>> >>> Regards, >>> Willem. >>> >>> On Tue, Nov 18, 2008 at 4:11 PM, >>> <[EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]> >>> <mailto:[EMAIL PROTECTED] >>> <mailto:[EMAIL PROTECTED]>>> wrote: >>> >>> Hi >>> >>> It is best to get hostnames working as spacewalk is fussy about this. >>> >>> Regards >>> >>> >> _______________________________________________ >> Spacewalk-list mailing list >> [email protected] >> https://www.redhat.com/mailman/listinfo/spacewalk-list >> >> >
_______________________________________________ Spacewalk-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/spacewalk-list
