[Spacewalk-list] unable to cancel pending actions!

[Andy Ingham]

Grant --

Thanks for the offer.  I am testing a workaround of "breaking" OSAD
connectivity and disabling actions on the client end until the point when
it is "ok" for the action to get picked up and executed.

I'm game for the pgsql option if my testing doesn't work, though.

Best,
Andy

On 2/25/14 9:39 AM, "Grant Gainey" <ggainey at redhat.com> wrote:

Hey Andy,

There's a fix that keeps you from getting into this state in the next
version of SW. Unfortunately, the fix won't help once you're already
broken.

I have some Oracle SQL that will fix this, but it'll take me a bit to
translate to pgres - the Oracle code uses START WITH ...CONNECT BY, which
I have to convert to a Postgresql WITH RECURSIVE, which is kind of
exciting.

Assuming I can pull it together - are you comfortable running direct SQL
against your SW's postgres?

G

----- Original Message -----
> I believe the ability to select MULTIPLE systems for a package action and
> link that with an arbitrary script on ALL those selected systems (such as
> a post-update reboot) was new in spacewalk 2.0 (in prior versions, it was
> only possible to link a package action with a remote command on one
>system
> at a time).
> 
> It seems that the problem I'm encountering is directly related to this
>new
> functionality, in that I am unable to CANCEL "linked" actions that were
> set up IN BULK.  (Regardless of whether I try to cancel individual
> actions, individual systems, all events together, via GUI, or via
> spacecmd).
> 
> The reboots need to be cancelled, so I'm guessing I'll have to DELETE the
> systems from spacewalk and re-add them.
> 
> Any other ideas, short of this drastic action?
> 
> Andy
> 
> 
> 
> On 2/24/14 4:45 PM, "Andy Ingham" <andy.ingham at duke.edu> wrote:
> 
> Either in the GUI or via spacecmd.
> 
> spacecmd {SSM:0}> schedule_cancel 30953 30741 30740 30739 30738 30737
>30736
> ERROR: redstone.xmlrpc.XmlRpcFault: unhandled internal exception: null
> 
> 
> 
> 
> I've tried restarting both the spacewalk stack AND postgres
> 
> 
> Errors in /var/log/tomcat6/catalina.out seem to indicate it is unhappy
> about the fact that the events are (package updates & arbitrary scripts)
> scheduled together.
> 
> I REALLY need to be able to cancel these events sometime in the next 24
> hours.
> 
> Is there any way to force them to clear?
> 
> 
> I just upgraded to Spacewalk 2.0 (still with a local postgres 8.4
>backend)
> 
> 
> 
> Help!
> 
> Andy
> 
> 
> _______________________________________________
> Spacewalk-list mailing list
> Spacewalk-list at redhat.com
> https://www.redhat.com/mailman/listinfo/spacewalk-list
> 
> 
> _______________________________________________
> Spacewalk-list mailing list
> Spacewalk-list at redhat.com
> https://www.redhat.com/mailman/listinfo/spacewalk-list
>