On Mon, Dec 23, 2013 at 6:23 PM, Klaus Aehlig <[email protected]> wrote:
> Now that all jobs are monitored with inotify, increase the polling > interval. > > Signed-off-by: Klaus Aehlig <[email protected]> > --- > src/Ganeti/Constants.hs | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/src/Ganeti/Constants.hs b/src/Ganeti/Constants.hs > index 9b781e4..7ded77b 100644 > --- a/src/Ganeti/Constants.hs > +++ b/src/Ganeti/Constants.hs > @@ -3931,8 +3931,10 @@ partReserved = 0.02 > -- | Time intervall in seconds for polling updates on the job queue. This > -- intervall is only relevant if the number of running jobs reaches the > maximal > -- allowed number, as otherwise new jobs will be started immediately > anyway. > +-- Also, as jobs are watched via inotify, scheduling usually works > independent > +-- of polling. > luxidJobqueuePollInterval :: Int > -luxidJobqueuePollInterval = 3 > +luxidJobqueuePollInterval = 307 > Thats an unusal magic number. Please add a comment why this is a good choice. > > -- | Maximal number of jobs to be running at the same time. Once this > number is > -- reached, new jobs will just be queued and only started, once some of > the > -- > 1.8.5.1 > > Rest LGTM, thanks -- -- Helga Velroyen | Software Engineer | [email protected] | Google Germany GmbH Dienerstr. 12 80331 München Registergericht und -nummer: Hamburg, HRB 86891 Sitz der Gesellschaft: Hamburg Geschäftsführer: Graham Law, Christine Elizabeth Flores
