On 2020-01-30 6:05 a.m., stakanov wrote:
Unevictable, unreclamable, unresponsive........seems to be more a grave than a swap. Complains about a qt problem but I do not understand which kind of. While the system swaps so heavily the memory is free, not even a problem of cached memory. The system has 8GB and when this happens it has up to 25% of unallocated(!) memory, completely free. So this is definitely not normal. If I force back all to ram I have still 15 to 20% of memory free and unallocated.
All the evidence is that the swap is a ratchet mechanism. Once something is swapped out it being swapped back in makes no difference to the amount of swap allocated. Or maybe it just works in terms a high water mark. That you CAN do a swapoff and the system works fine after with zero in swap is very significant. Sometimes it seems Linux does 'pre-emptive' swapping', making guesses, rightly or wrongly, about what might not be needed in the near future. The more you jack up 'swapiness' the more likely that is. I understand the reasoning behind pre-emptive swapping but disagree with the wisdom. Once again I point out that code does not get swapped out, only volatile data. Or that's how it should be. I strongly suggest running 'vmstat -SM -a 5' in a terminal window. Oh, and read the man page as well :-) -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org