[Puppet-users] Unmanageable state 'maintenance' on Sol10 service

[Matthew Flanagan]

On 4/3/07, Luke Kanies <luke at madstop.com> wrote:
> On Mar 22, 2007, at 8:06 PM, Matthew Flanagan wrote:
>
> > On 3/21/07, Luke Kanies <luke at madstop.com> wrote:
> >> On Mar 18, 2007, at 4:53 PM, Charlie Schluting wrote:
> >>>
> >>> I, too, believe it'd be very nice for puppet to just run 'svcadm
> >>> clear
> >>> $FMRI' in this case. Like someone else said, if the service is
> >>> really
> >>> broken, SMF will just put it back in maint mode.
> >>>
> >>> Often times, the service is fixed and just needs to be cleared of
> >>> the
> >>> maintenance state. I've been in the same situation a few times,
> >>> and I
> >>> had to ssh to each box to fix it.
> >>
> >> I'm perfectly willing to do this, I just figured it was dangerous or
> >> something.
> >>
> >> Anyone disagree with doing this?
> >
> > I think it is safe to do this.
>
> Is someone willing to file this as an enhancement request, with the
> specific command to use to clear the service?  How many times should
> this be tried (i.e., I expect I'll have to clear, reparse the output,
> and check again)?  Or should I just try once and assume the service
> successfully clears?

I would only try and clear it once. SMF's svc.startd(1M) handles
retrying and only puts a service into maintenance state if:

"three method failures happen in a row, or if the service is
restarting more than once a second"

>
>   --
>   You only have to be open minded if you're wrong.
>   ---------------------------------------------------------------------
>   Luke Kanies | http://reductivelabs.com | http://madstop.com
>
>
> _______________________________________________
> Puppet-users mailing list
> Puppet-users at madstop.com
> https://mail.madstop.com/mailman/listinfo/puppet-users
>


-- 
matthew
http://wadofstuff.blogspot.com