Dear Alexandros, as I'm monitoring cpu+mem load on fix+max for weeks while running my jobs,I noted for >1day the 256G VM on fix using ~all of its mem while having aconstant cpu load < 30%. This is after I killed all but 2 of my jobs. My2 jobs are stopped for now by kill -STOP, to be -CONT. Checking with iotop, there is no IO problem, and adding all mem, no mem problem(also seen by ~10G used for buff). fix 09102020 top - 18:24:44 up 208 days, 2:29, 1 user, load average: 7.04, 8.13, 9.21 Tasks: 670 total, 1 running, 665 sleeping, 3 stopped, 1 zombie %Cpu(s): 8.9 us, 0.8 sy, 0.0 ni, 84.2 id, 5.6 wa, 0.0 hi, 0.4 si, 0.0 st MiB Mem : 515779.0 total, 15306.1 free, 489885.8 used, 10587.2 buff/cache MiB Swap: 192512.0 total, 76012.1 free, 116499.9 used. 22543.8 avail Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 5760 libvirt+ 20 0 265.8g 258.9g 7892 S 21.5 51.4 9035:27 qemu-system-x86 20070 libvirt+ 20 0 133.4g 95.9g 8464 S 302.3 19.0 70430:11 qemu-system-x86 8475 libvirt+ 20 0 129.5g 85.5g 5024 S 19.2 17.0 35409:13 qemu-system-x86 16254 libvirt+ 20 0 21.0g 8.7g 6416 S 8.6 1.7 18042:44 qemu-system-x86 33780 libvirt+ 20 0 20.9g 6.8g 5752 S 8.9 1.3 17161:44 qemu-system-x86 14947 reczko 20 0 7739632 6.4g 2048 T 0.0 1.3 1047:48 batchman 26183 libvirt+ 20 0 20.9g 4.0g 6408 S 10.9 0.8 19247:54 qemu-system-x86 1280 libvirt+ 20 0 20.9g 3.6g 7712 S 10.6 0.7 3846:18 qemu-system-x86 13808 reczko 20 0 7739632 821496 2100 T 0.0 0.2 937:02.68 batchman (+ 258.9 95.9 85.5 8.7 6.8 6.4 4.0 3.6) 469.79999999999995