I am already using java service wrapper and having another wrapper around the jsw wrapper sounds super hack. Although it would fix the issue. I have gone as far as spreading the applications across more roles to reduce possible configuration load. Do we know for a fact that puppet is purposely unscheduling these service refreshes due to high load. Is this feature/bug documented anywhere? Where would the code be that implements this "smart unschedule"?
than On Thursday, June 19, 2014 6:00:33 PM UTC-7, Trevor Vaughan wrote: > > A fact probably won't help in this case since it seems to be something > that's happening as the service get restarted during the Puppet run and a > fact will only tell you what's happening at the start of the run. > > What you may need to do is to change the timeout and restart values of > your service statements to account for these anomalies. Your restart > statements could call a wrapper script that restarts the service but waits > in the background until enough resources are free and your timeout could be > set sufficiently high that the service resource would have time to execute. > > If you happen to not have anything dependent on this service restarting > (which is usually the case), you could even fire the service statement off > into the background with your wrapper script and allow the Puppet run to > continue unabated. > > This is one of those extremely sticky issues that seems to be particularly > rife in Java-land. > > Thanks, > > Trevor > > > On Thu, Jun 19, 2014 at 6:56 PM, Pete Brown <[email protected] > <javascript:>> wrote: > >> You could write a fact that checks the system load and base your >> service restarts on that. >> I thought there was one but there only seems to be uptime and swapfree. >> >> Here's a ruby gem that might help get you started. >> https://github.com/nethacker/usagewatch >> >> >> On 20 June 2014 04:45, Corey Osman <[email protected] <javascript:>> >> wrote: >> > Is there any "back off" algorithm in puppet that would unscheduled a >> service refresh when the CPU load is too high. Is there a timeout >> associated with refreshing services? >> > >> > My CPU load peaks as high as 18(one min)(6 Processor Count) during >> puppet runs . I have about 20-30 java processes that need restarting >> almost every time so its quite a chore for puppet to restart all these >> services. >> > >> > >> > My puppet runs last about 8-9 minutes for a full >> configuration/deployment. >> > >> > >> > >> > Corey >> > >> > -- >> > You received this message because you are subscribed to the Google >> Groups "Puppet Users" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> an email to [email protected] <javascript:>. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/puppet-users/D0CE60BA-B66D-4CAB-8B41-86E8CF75C31D%40logicminds.biz >> . >> > For more options, visit https://groups.google.com/d/optout. >> >> -- >> You received this message because you are subscribed to the Google Groups >> "Puppet Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/puppet-users/CAJ8DPF71uaVqpS_Ap_NtH1xG8p0dohfKa-5GUiWZ%3D3zh6t3t5Q%40mail.gmail.com >> . >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Trevor Vaughan > Vice President, Onyx Point, Inc > (410) 541-6699 > [email protected] <javascript:> > > -- This account not approved for unencrypted proprietary information -- > -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-users/aded3733-a4ef-4257-9a5c-0741ded9faf0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
