Hi Lukasz, Ilya, I completely agree with Ilya that the core Ignite modules shouldn't deal with that.
That said, I've seen numerous situations when people tried to implement scheduling of tasks for Ignite. Sometimes, they wouldn't do it in a distributed fashion, and it would because the hosting node failed. Sometimes, they would use ignite-schedule and, well, be unhappy with it. I think it is absolutely fine for ignite-extensions to have a distributed scheduler. Lukasz, I would suggest that you drive both activities - remove ignite-scheduler from core modules AND add an improved version to extensions. Thanks, Stan > On 17 Oct 2022, at 20:40, Łukasz Dywicki <l...@code-house.org> wrote: > > Hello Ilya, > Thank you for clarification. I wasn't aware schedule is on strike. ;-) > Mentioned cron-utils is something I used. > My intention was to reduce number of profiles needed to build ignite, but if > this module is marked for removal then both - an extra profile and licensing > inconsistency will disappear. > > Cheers, > Łukasz > > On 17.10.2022 18:16, Ilya Kasnacheev wrote: >> Hello! >> I suggest getting rid of ignite scheduler completely. >> It is poorly thought out, not used by anyone, has a baroque and incomplete >> API. In no way it is doing anything useful for out users in its current >> form. >> Yes we could rewrite it to be more useful, but I think Ignite is not a >> distributed scheduling system. We have too much on our plate already. If >> anybody wants to create a robust scheduler based on Ignite, they are free >> to do so in a dedicated project. >> I think we had consensus to drop it but just too lazy to complete that >> process. >> Regards,
