[Moabusers] moab not clearing old triggers

[wightman]

There is a new snapshot that should have this issue fixed.

Let us know if you still have problems with node triggers.

Thanks,

- Douglas

On Thu, 2006-09-28 at 15:19 -0600, Lloyd Brown wrote:
> Alright.  I'll do that in the future.  I just wasn't sure if it was a
> bug, or some parameter I was missing.  Goodness knows I miss simple
> things like that a lot.
> 
> Thanks,
> Lloyd
> 
> wightman wrote:
> > Lloyd,
> > 
> > In most cases the moabusers mailing list is for situations where fellow
> > moab users can talk about policies and such.  
> > 
> > For bug reports and the like we recommend that customers use either the
> > email address "support at clusterresources.com" or you can log into
> > http://support.clusterresources.com and file a trouble ticket.
> > 
> > For the fastest response from Cluster Resources regarding a bug we
> > recommend using the trouble ticket system.  It's easy to track emails
> > back and forth and it is also the only "queue" that is regularly checked
> > to verify no outstanding issues.
> > 
> > In regards to this issue.  We have verified the behavior and are aware
> > of a fix.  However, we are running a few regressions tests on the fix to
> > verify that nothing has been broken.
> > 
> > We will let you know when a new patch is ready.
> > 
> > Thanks for the bug report.
> > 
> > - Douglas
> > 
> > On Thu, 2006-09-28 at 15:05 -0600, Lloyd Brown wrote:
> >> Nothing, huh?  Can someone at least verify the behavior?
> >>
> >> Thanks,
> >> Lloyd Brown
> >>
> >> Lloyd Brown wrote:
> >>> So, I'm using Moab to do triggers to notify when a node goes down.  The
> >>> specification looks something like this in the moab.cfg:
> >>>
> >>>> NODECFG[DEFAULT] TRIGGER=AType=exec,Action="/opt/moab/node_down_notification.pl $OID",EType=fail,MultiFire=TRUE,RearmTime=5:00
> >>> Somehow, though, whenever moab restarts, I get a bunch of the nodes with
> >>> a duplicate trigger.  I mean that when I start with a clean slate,
> >>> having deleted all the triggers by hand (mschedctl -d
> >>> trigger:trigger_id), and I re-cycle moab (mschedctl -R), I get a
> >>> "default" trigger, as well as a copy for each node, in this case, mdiag
> >>> -T | wc -l shows 624 total lines.  If I then immediately re-cycle again,
> >>> I get 1029 lines.  Not all the nodes seem to have duplicates, but it
> >>> looks like there's some arbitrary limit on triggers that limits you to
> >>> 1024 (plus 5 lines of header/footer to the mdiag -T command).  I don't
> >>> have any problem staying under that limit for now, but I wanted to
> >>> understand why the old copy of the triggers isn't getting deleted.  Is
> >>> there some variable I can set so we don't get these duplicates any time
> >>> moab re-starts?
> >>>
> >>> Thanks,
> >>> Lloyd Brown
> >>> _______________________________________________
> >>> moabusers mailing list
> >>> moabusers at supercluster.org
> >>> http://www.supercluster.org/mailman/listinfo/moabusers
> >>>
> >>>
> >>>
> >> _______________________________________________
> >> moabusers mailing list
> >> moabusers at supercluster.org
> >> http://www.supercluster.org/mailman/listinfo/moabusers
> > 
> > 
> > 
>