sysctl vm.overcommit_memory=1 echo never > /sys/kernel/mm/transparent_hugepage/enabled echo 511 > /proc/sys/net/core/somaxconn 1:M 26 Nov 2023 21:34:33.840 # WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128. 1:M 26 Nov 2023 21:34:33.840 # Server initialized 1:M 26 Nov 2023 21:34:33.840 # WARNING overcommit_memory is set to 0! Background sa........

Assign way more replicas than you have of memory on all nodes and watch the Swarm crash which can easily reproduce in a small VMfor testing. root@Deb11Docker01:~# docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAM........

If you have swarm services and dockerd is creating a high load even with the containers just being idle, the easiest solution is to upgrade to a newer docker version. For example an identical config of 3 nodes, with Redis 5 with 30 replicas produces a load of about 1.45 in Debian 10 with Docker18.09.1 If I create the same setup on Debian 11, with Docker 20.10.5+dfsg1 then the CPU usage is low. One other difference I wondered is the kernel. In my test setup........

Bonding is an excellent way to get both increased redundancy and throughput. It is similar to the "Network Teaming" feature in Windows. There are a few different modes but we will use mode 6, I think it's the best of both worlds, as it is not just a failover, but it provides round robin, so you will get redundancy and load balancing. So if you have a 1G single port, you will have a combined throughput of 4G at this point. Just bear in mind that the true thr........

This is sure simple if you follow the guide but it took a lot of hacking around to make this work on Debian/Ubuntu! Now before you ask why bother running wine and python, the reason is because Python executables are NOT cross-platform. If you run pyinstaller in Linux, that binary will only run on Linux and the same if you do it in Windows. So it is preferable if you have a single environment that you can create Linux and Windows binaries from rather than running 2 separate........

The strange thing is that usually the first install or two will work on any new machine but then it suddenly won't. I had this experience on QEMU 2.13 on a different machine. There is something finicky or buggy about the CUCM installer even when choosing the same virtual hardware specs. qemu-kvm command: /usr/libexec/qemu-kvm -version QEMU PC emulator version 0.12.1 (qemu-kvm-0.12.1.2-2.506.el6_10.1), Copyright (c) 2003-2008 Fabrice Bellard ........

M4A is a weird format, so you have to be creative here is a quick copy of what I did. Basically you need to convert to .wav to make use of them and thenI converted the resulting .wav into an mp3 (nice small file size and basically universally playable): ======= sudo apt-get install mpg321 mp3gain faad normalize lame faad "Voice 002 (copy).m4a" faad "Voice 002 (copy).m4a"........

The folder I was trying to archive is about 72GB, but much like rsync at about 17GB it chokes because of the filesize. What's with so many common and essential Linux tools having such limitations? I guess it is likely that the authors never wrote their code with the idea that files would be so large but it's still very annoying. It's important to stay on top of these limitations on production servers because I didn't realize what happened until I checked the file with "........