Hi! That's interesting - thanks for finding this!
If this is the case then we should at first try to update. Gonna create a ticket for it. Otoh quartz is already 700kB in size. That's quite a lot for such a simple stuff like starting a few tasks every minute. LieGrue, strub > Am 05.01.2018 um 20:10 schrieb John D. Ament <[email protected]>: > > On Fri, Jan 5, 2018 at 2:00 PM John D. Ament <[email protected]> wrote: > >> On Fri, Jan 5, 2018 at 1:21 PM Mark Struberg <[email protected]> >> wrote: >> >>>> Shouldnąt we just disable the update check in our quartz configuration? >>> >>> Well there is a flag to disable this behaviour, but a.) it's hard to set >>> (requires -D) and it doesn't disable 100%. >>> The code still does some http calls out :/ >>> >> >> You can disable it programmatically. I've done wireshark checks, there is >> no other HTTP calls made out. Here's a quick patch that does it: >> >> >> https://github.com/johnament/deltaspike/commit/5a4fec98ff3f8f6ed6b54c36332bb8621cd3b09d >> > > Here's a more interesting thing, looks like for Quartz 2.3 they changed > from an opt out to an opt in - > https://github.com/quartz-scheduler/quartz/commit/dfe1e5a3cc248e2a46a6ea55567aaa6dc8e15ca5 > > John > > >> >> >> John >> >> >>> >>> That's really bad, and the terracotta community (or rather the firm >>> behind it) declined to disable it by default since 2010 :/ >>> >>> @Tomas, yes there is additional effort to maintain it. But imo it's worth >>> it. >>> >>> @Romain, John the question for me is rather where we do like to keep the >>> code. >>> Either here in DeltaSpike or at geronimo? The reason is that we might >>> also later use this in other projects (TomEE) as well. >>> >>> For now I'd just start to play a bit with it over here and then we can >>> still move it around later. >>> >>> LieGrue, >>> strub >>> >>> >>>> Am 05.01.2018 um 17:44 schrieb Arne Limburg < >>> [email protected]>: >>>> >>>> Hi Mark, >>>> >>>> Shouldnąt we just disable the update check in our quartz configuration? >>>> >>>> Cheers, >>>> Arne >>>> >>>> >>>> Am 05.01.18 17:39 schrieb "Mark Struberg" unter >>>> <[email protected]>: >>>> >>>>> Hi folks! >>>>> Since I've now had a few complaints about Quartz 'phoning home' >>> (totally >>>>> useless update check), I'm really inclined to just kick out quartz and >>>>> implement the Scheduler ourselves. >>>>> Implementing a proper Scheduler is not that complicated anyway, so do >>> we >>>>> like to roll this ourselves? >>>>> Or do you think I underestimate the effort? >>>>> >>>>> LieGrue,strub
