Thanks Ilya, I’ll upgrade and see if that works soon. Right now we’ve switched to only using partitioned caches to avoid this issue. If it’s something that isn’t fixed with 2.7 then I’ll create a bug and attach code to reproduce it.
I’m really surprised that no one else is complaining about it. The repro is very straight forward.. On Mon, Dec 24, 2018 at 06:36 Ilya Kasnacheev <ilya.kasnach...@gmail.com> wrote: > Hello! > > There could be fixes for expiration between 2.6 and 2.7. > > However, if it will still be the case, we would be glad if you could post > reproducer for this behavior. > > Note that you can almost emulate replicated cache by creating partitioned > cache with large number of backups, maybe you should be using that in the > meantime. > > Regards, > > -- > Ilya Kasnacheev > > > пт, 21 дек. 2018 г. в 23:28, scottmf <scott...@gmail.com>: > >> I forgot to mention this is on Ignite 2.6.0 and i'm running with a 3 node >> cluster. >> >> I've been experimenting some more and I'm finding that using PARTITIONED >> caches works fine, but replicated caches stop expiring usually within 10 >> minutes. The behavior is very consistent. >> >> I plan on trying this with Ignite 2.7.0... >> >> >> >> -- >> Sent from: http://apache-ignite-users.70518.x6.nabble.com/ >> >
