Re:Swap, swap, swap, and bad places to work
by kevin_nisbet on Friday August 09, 2019 @11:23AM
(#20655175)
This is a plausible route, but it still requires some engineering, specifically tweaking swapiness setting. Otherwise the swap will get used even with plenty of memory available, which in my experience can still cause havoc for GCed processes with high allocation rates on non-ssd disks.
Re:Swap, swap, swap, and bad places to work
by VectorLock on Friday August 09, 2019 @12:41PM
(#20655973)
Why let them live? Why wake myself up? Now your swapping systems are introducing a performance degradation.
Re:Swap, swap, swap, and bad places to work
by mankyd on Friday August 09, 2019 @02:00PM
(#20656819)
Quote: "maybe kill some of them"
Quote: "You also don't want to have a cascading failure where a massive leak makes all your machines fill their ram, and start killing everything like crazy."
Cascading failures are a very real thing that have knocked whole systems offline.
It sounds like the real solution is a balanced solution involving some engineering: kill them if you aren't killing _everything_. Page if the problem is ongoing, not if a couple of machines have a problem.
Either way, you can add swap _and_ kill them. One does not preclude the other.
Why not give them swap, set off pagers, and _maybe_ kill them? There could still be something worth investigating there, and having swap will make that easier
You also don't want to have a cascading failure where a massive leak makes all your machines fill their ram, and start killing everything like crazy.