i am SO CLOSE to ps3 debian. this is me chrooted into my install cuz my kernel sucks and is missing cgroup support
August 22, 2025 at 2:13 AM
i am SO CLOSE to ps3 debian. this is me chrooted into my install cuz my kernel sucks and is missing cgroup support
Hooray! I'm now 99.98% sure we fixed this: github.com/compiler-exp...
August 20, 2025 at 3:48 PM
Hooray! I'm now 99.98% sure we fixed this: github.com/compiler-exp...
Linux Control Group Cheat Sheet 🔽
cgroup is a Linux kernel mechanism that limits and accounts for resource usage: CPU, memory, IO, etc. Docker, Kubernetes, and systemd rely on cgroup, and you can also use it directly.
The main cgroup v2 concepts and tools on a single diagram:
cgroup is a Linux kernel mechanism that limits and accounts for resource usage: CPU, memory, IO, etc. Docker, Kubernetes, and systemd rely on cgroup, and you can also use it directly.
The main cgroup v2 concepts and tools on a single diagram:
October 26, 2024 at 3:23 PM
Linux Control Group Cheat Sheet 🔽
cgroup is a Linux kernel mechanism that limits and accounts for resource usage: CPU, memory, IO, etc. Docker, Kubernetes, and systemd rely on cgroup, and you can also use it directly.
The main cgroup v2 concepts and tools on a single diagram:
cgroup is a Linux kernel mechanism that limits and accounts for resource usage: CPU, memory, IO, etc. Docker, Kubernetes, and systemd rely on cgroup, and you can also use it directly.
The main cgroup v2 concepts and tools on a single diagram:
you wouldnt know her, she goes to a different cgroup
August 25, 2026 at 4:25 PM
you wouldnt know her, she goes to a different cgroup
anything you can do with a cgroup, you can do with k8s, and the primary exclusive thing you get from k8s is that you can treat "the abstract idea of a process that might be running currently" as a discrete thing, like systemd services
September 24, 2026 at 8:49 PM
anything you can do with a cgroup, you can do with k8s, and the primary exclusive thing you get from k8s is that you can treat "the abstract idea of a process that might be running currently" as a discrete thing, like systemd services
Desperate times. We rebuilt CE's OS image and now _something_ SOMETIMES takes cpu cgroup delegation away from us. We suspect systemd, and some kind of start-up race condition. But been unable to reliably repro or diagnose.
Does anyone here know anything about such things?
github.com/compiler-exp...
Does anyone here know anything about such things?
github.com/compiler-exp...
Intermittent cgroup cpu controller availability · Issue #1761 · compiler-explorer/infra
On a few (~5%) of our EC2 instances (running identical AMIs), the cpu controller is not enabled in /sys/fs/cgroup/cgroup.subtree_control at boot time, causing nsjail to fail with: [E][2025-08-11T20...
github.com
August 16, 2025 at 9:23 PM
Desperate times. We rebuilt CE's OS image and now _something_ SOMETIMES takes cpu cgroup delegation away from us. We suspect systemd, and some kind of start-up race condition. But been unable to reliably repro or diagnose.
Does anyone here know anything about such things?
github.com/compiler-exp...
Does anyone here know anything about such things?
github.com/compiler-exp...
技術書典で自分のサークル以外からもこんな本出します。cgroupのカーネルコードの解説本です📕
techbookfest.org/product/xuVP...
techbookfest.org/product/xuVP...
cgroupの歩き方 Linux 4.19版：Linuxカーネルもくもく会
Linux 4.19に実装されているcgroupのコア機能の実装を解説します。
本書ではcgroupのデータ構造、cgroupの初期化、タスクの登録・移動・削除などcgroupの機...
techbookfest.org
November 6, 2023 at 3:58 AM
技術書典で自分のサークル以外からもこんな本出します。cgroupのカーネルコードの解説本です📕
techbookfest.org/product/xuVP...
techbookfest.org/product/xuVP...
Dear #askfedi, #devuan edition,
did someone had problems with #docker on #linux 6.12.57+deb13? Mine fails to start after trying to
`/sys/fs/cgroup/ mount -n -t cgroup -o cpuset cgroup cpuset`
with
```
mount: /sys/fs/cgroup/cpuset: fsconfig() failed: cgroup: Unknown subsys name 'cpuset' […]
did someone had problems with #docker on #linux 6.12.57+deb13? Mine fails to start after trying to
`/sys/fs/cgroup/ mount -n -t cgroup -o cpuset cgroup cpuset`
with
```
mount: /sys/fs/cgroup/cpuset: fsconfig() failed: cgroup: Unknown subsys name 'cpuset' […]
Original post on mastodon.social
mastodon.social
November 29, 2025 at 7:33 PM
"Flatten The Pick" Linux Patches Progress For Better cgroup Scheduling While Linux Gaming - https://www.phoronix.com/news/Flatten-The-Pick-v3
"Flatten The Pick" Linux Patches Progress For Better cgroup Scheduling While Linux Gaming
A month ago I wrote about Linux scheduler work to help boost gaming performance on old "potato" hardware with Intel engineer Peter Zijlstra noting that Linux cgroup scheduling has continued to be "a pain in the arse." This work continues advancing with a third iteration of these "flatten the pick" patches being posted...
www.phoronix.com
June 7, 2026 at 11:14 AM
"Flatten The Pick" Linux Patches Progress For Better cgroup Scheduling While Linux Gaming - https://www.phoronix.com/news/Flatten-The-Pick-v3
書きました。間違いがあったら優しく教えてください😇
https://gihyo.jp/article/2023/07/linux_containers-0053
https://gihyo.jp/article/2023/07/linux_containers-0053
第53回 Linuxカーネルのコンテナ機能 - cgroup v2から使うCPUの帯域幅制限（1��...
前回は、cgroup v2から使うioコントローラの話題を取り上げました。今回はcgroup v2から使うCPUコントローラの話をしたいと思います。
gihyo.jp
August 1, 2023 at 1:38 AM
書きました。間違いがあったら優しく教えてください😇
https://gihyo.jp/article/2023/07/linux_containers-0053
https://gihyo.jp/article/2023/07/linux_containers-0053
Why Set GOMEMLIMIT for Go Apps in Containers/Kubernetes Pods? 🔽
- Linux containers isolate network and fs but share host CPU cores/RAM.
- cgroup memory limits aren’t considered by the Go runtime.
- The Go GC assumes the host’s total memory.
- Ignoring cgroup limits risks terminations by OOM Killer.
- Linux containers isolate network and fs but share host CPU cores/RAM.
- cgroup memory limits aren’t considered by the Go runtime.
- The Go GC assumes the host’s total memory.
- Ignoring cgroup limits risks terminations by OOM Killer.
December 22, 2024 at 5:55 PM
Why Set GOMEMLIMIT for Go Apps in Containers/Kubernetes Pods? 🔽
- Linux containers isolate network and fs but share host CPU cores/RAM.
- cgroup memory limits aren’t considered by the Go runtime.
- The Go GC assumes the host’s total memory.
- Ignoring cgroup limits risks terminations by OOM Killer.
- Linux containers isolate network and fs but share host CPU cores/RAM.
- cgroup memory limits aren’t considered by the Go runtime.
- The Go GC assumes the host’s total memory.
- Ignoring cgroup limits risks terminations by OOM Killer.
Yup. This was a good test because it was way down in the guts of a freshly booted VM using my custom init replacement, inside a container, dealing with cgroup namespaces and reproducibility problems and it only happened on arm64. Impressive deduction skills.
August 24, 2026 at 1:11 AM
Yup. This was a good test because it was way down in the guts of a freshly booted VM using my custom init replacement, inside a container, dealing with cgroup namespaces and reproducibility problems and it only happened on arm64. Impressive deduction skills.
I’ve filed a proposal to make GOMAXPROCS aware of container CPU limits by default! Please take a look and provide feedback if this is something you care about!
go.dev/issue/73193
go.dev/issue/73193
proposal: runtime: CPU limit-aware GOMAXPROCS default · Issue #73193 · golang/go
Overview Change the Go runtime on Linux to use CPU cgroup quota limits to set the default value of GOMAXPROCS. This is a concrete proposal for the ideas discussed in #33803. I've included a lot of ...
go.dev
April 7, 2025 at 5:59 AM
I’ve filed a proposal to make GOMAXPROCS aware of container CPU limits by default! Please take a look and provide feedback if this is something you care about!
go.dev/issue/73193
go.dev/issue/73193
Linux 7.1 sched_ext Brings cgroup Sub-Scheduler Groundwork, Idle SMT Sibling Improvement - https://www.phoronix.com/news/Linux-7.1-sched-ext
Linux 7.1 sched_ext Brings cgroup Sub-Scheduler Groundwork, Idle SMT Sibling Improvement
The extensible scheduler "sched_ext" code for allowing Linux scheduling behavior to be defined via BPF programs is seeing some useful improvements with the in-development Linux 7.1 kernel...
www.phoronix.com
April 16, 2026 at 9:36 PM
Linux 7.1 sched_ext Brings cgroup Sub-Scheduler Groundwork, Idle SMT Sibling Improvement - https://www.phoronix.com/news/Linux-7.1-sched-ext
Linux 6.18 sched_ext Preps For Cgroup Sub-Scheduler Support - https://www.phoronix.com/news/Linux-6.18-sched-ext
Linux 6.18 sched_ext Preps For Cgroup Sub-Scheduler Support
The sched_ext scheduler framework that allows creating kernel thread schedulers via BPF programs is ready with some updates for the Linux 6.18 kernel...
www.phoronix.com
September 27, 2025 at 10:20 AM
Linux 6.18 sched_ext Preps For Cgroup Sub-Scheduler Support - https://www.phoronix.com/news/Linux-6.18-sched-ext
Just landed: Learn Kubernetes weekly 179! My top picks:
🤔 cgroup v1 CPU Shares to v2 CPU Weight
💰 Kubernetes for Pentesters
⚙️ Observability in EKS
👨🔧 Prometheus SLO Generator Sloth
Read it here: https://kube.today/issues/179
🤔 cgroup v1 CPU Shares to v2 CPU Weight
💰 Kubernetes for Pentesters
⚙️ Observability in EKS
👨🔧 Prometheus SLO Generator Sloth
Read it here: https://kube.today/issues/179
April 15, 2026 at 12:11 PM
Just landed: Learn Kubernetes weekly 179! My top picks:
🤔 cgroup v1 CPU Shares to v2 CPU Weight
💰 Kubernetes for Pentesters
⚙️ Observability in EKS
👨🔧 Prometheus SLO Generator Sloth
Read it here: https://kube.today/issues/179
🤔 cgroup v1 CPU Shares to v2 CPU Weight
💰 Kubernetes for Pentesters
⚙️ Observability in EKS
👨🔧 Prometheus SLO Generator Sloth
Read it here: https://kube.today/issues/179
thats actually systemd's fault
systemd in its infinite wisdom will kill a user session by default if any process within its cgroup gets OOM killed by the kernel
systemd in its infinite wisdom will kill a user session by default if any process within its cgroup gets OOM killed by the kernel
March 6, 2025 at 11:50 AM
thats actually systemd's fault
systemd in its infinite wisdom will kill a user session by default if any process within its cgroup gets OOM killed by the kernel
systemd in its infinite wisdom will kill a user session by default if any process within its cgroup gets OOM killed by the kernel
Kubernetes v1.34: Autoconfiguration for Node Cgroup Driver Goes GA-
Kubernetes v1.34: Autoconfiguration for Node Cgroup Driver Goes GA
Historically, configuring the correct cgroup driver has been a pain point for users running new Kubernetes clusters. On Linux systems, there are two different cgroup drivers: cgroupfs and systemd. In...
kubernetes.io
September 15, 2025 at 6:06 PM
Kubernetes v1.34: Autoconfiguration for Node Cgroup Driver Goes GA-
Sub-Scheduler Support Could Be One Of The Most Exciting Features To Come For Linux 7.1 - https://www.phoronix.com/news/cgroup-sub-scheduler-sched-ext
Sub-Scheduler Support Could Be One Of The Most Exciting Features To Come For Linux 7.1
While there are many great Linux 7.0 features with that still-young development cycle, looking ahead to Linux 7.1 this summer there's an interesting feature on track: cgroup sub-scheduler support for sched_ext...
www.phoronix.com
February 25, 2026 at 4:12 PM
Sub-Scheduler Support Could Be One Of The Most Exciting Features To Come For Linux 7.1 - https://www.phoronix.com/news/cgroup-sub-scheduler-sched-ext
技術書典15では、自分のサークルだけでなく、お隣のサークル（き11）からもこんな本出しますのでよろしくです！
cgroupのカーネルコードの解説本です！
cgroupのカーネルコードの解説本です！
October 31, 2023 at 4:21 AM
技術書典15では、自分のサークルだけでなく、お隣のサークル（き11）からもこんな本出しますのでよろしくです！
cgroupのカーネルコードの解説本です！
cgroupのカーネルコードの解説本です！
"Kubernetes has made infra easy for devs! Just use our slick APIs!"
dev: what is this cgroup error
me: oh that's a Linux thing—
dev: I HAVE TO LEARN LINUX NOW???
dev: what is this cgroup error
me: oh that's a Linux thing—
dev: I HAVE TO LEARN LINUX NOW???
August 20, 2025 at 5:34 PM
"Kubernetes has made infra easy for devs! Just use our slick APIs!"
dev: what is this cgroup error
me: oh that's a Linux thing—
dev: I HAVE TO LEARN LINUX NOW???
dev: what is this cgroup error
me: oh that's a Linux thing—
dev: I HAVE TO LEARN LINUX NOW???
"Kubernetes 1.35 “Timbernetes” Introduces Vertical Scaling" thenewstack.io/kubernetes-1...
Kubernetes 1.35 "Timbernetes" Introduces Vertical Scaling
This release also features 60 enhancements, including a specialized KYAML format and gang scheduling for AI workloads, while phasing out legacy components like Ingress Nginx and cgroup v1.
thenewstack.io
December 17, 2025 at 8:51 PM
"Kubernetes 1.35 “Timbernetes” Introduces Vertical Scaling" thenewstack.io/kubernetes-1...
This case study traces why Kafka started reading from disk after moving from EC2 to EKS, and lands on cgroup v2 reclaim pressure
➤ https://ku.bz/0P1BSLKKz
➤ https://ku.bz/0P1BSLKKz
September 19, 2026 at 3:16 PM
This case study traces why Kafka started reading from disk after moving from EC2 to EKS, and lands on cgroup v2 reclaim pressure
➤ https://ku.bz/0P1BSLKKz
➤ https://ku.bz/0P1BSLKKz
LXCで学ぶコンテナ入門 －軽量仮想化環境を実現する技術 「第63回 Linuxカーネルのコンテナ機能 ― cgroupの改良版cgroup v2［8］」公開
gihyo.jp/article/2026...
gihyo.jp/article/2026...
第63回 Linuxカーネルのコンテナ機能 ― cgroupの改良版cgroup v2［8］ | gihyo.jp
cgroup v2がstableとされた4.5カーネルがリリースされたのは2016年で、もう10年が経ちました。この連載でcgroup v2を初めて取り上げたのは2017年で、そこからでも8年以上経っています。
gihyo.jp
June 24, 2026 at 12:15 AM
LXCで学ぶコンテナ入門 －軽量仮想化環境を実現する技術 「第63回 Linuxカーネルのコンテナ機能 ― cgroupの改良版cgroup v2［8］」公開
gihyo.jp/article/2026...
gihyo.jp/article/2026...
This week on the Learn Kubernetes Weekly:
🌪️ Centralizing controller upgrades
🆙 Major update ingress controller
📦 OCI introduction
🚦 HA with dynamic DNS
🤗 Embracing cgroup v2
Read it now: https://learnk8s.io/issues/119
🌟 Loft Labs
🌪️ Centralizing controller upgrades
🆙 Major update ingress controller
📦 OCI introduction
🚦 HA with dynamic DNS
🤗 Embracing cgroup v2
Read it now: https://learnk8s.io/issues/119
🌟 Loft Labs
February 19, 2025 at 12:36 PM
This week on the Learn Kubernetes Weekly:
🌪️ Centralizing controller upgrades
🆙 Major update ingress controller
📦 OCI introduction
🚦 HA with dynamic DNS
🤗 Embracing cgroup v2
Read it now: https://learnk8s.io/issues/119
🌟 Loft Labs
🌪️ Centralizing controller upgrades
🆙 Major update ingress controller
📦 OCI introduction
🚦 HA with dynamic DNS
🤗 Embracing cgroup v2
Read it now: https://learnk8s.io/issues/119
🌟 Loft Labs