#nerdctl — Public Fediverse posts
Live and recent posts from across the Fediverse tagged #nerdctl, aggregated by home.social.
-
Managing containers using #nerdctl compose in an IPv6-only environment is aggravating, not least due to non-intuitive default behavior like all v6 nameservers being stripped from /etc/resolv.conf .
But I choose not to blame the Nerdctl developers, as their implementation intentionally tracks that of Docker Engine. Docker themselves are responsible for royally messing up network understanding for a whole generation of developers, ruining the excellent use-case of #IPv6 for scaled-up containers.
-
Managing containers using #nerdctl compose in an IPv6-only environment is aggravating, not least due to non-intuitive default behavior like all v6 nameservers being stripped from /etc/resolv.conf .
But I choose not to blame the Nerdctl developers, as their implementation intentionally tracks that of Docker Engine. Docker themselves are responsible for royally messing up network understanding for a whole generation of developers, ruining the excellent use-case of #IPv6 for scaled-up containers.
-
Managing containers using #nerdctl compose in an IPv6-only environment is aggravating, not least due to non-intuitive default behavior like all v6 nameservers being stripped from /etc/resolv.conf .
But I choose not to blame the Nerdctl developers, as their implementation intentionally tracks that of Docker Engine. Docker themselves are responsible for royally messing up network understanding for a whole generation of developers, ruining the excellent use-case of #IPv6 for scaled-up containers.
-
Managing containers using #nerdctl compose in an IPv6-only environment is aggravating, not least due to non-intuitive default behavior like all v6 nameservers being stripped from /etc/resolv.conf .
But I choose not to blame the Nerdctl developers, as their implementation intentionally tracks that of Docker Engine. Docker themselves are responsible for royally messing up network understanding for a whole generation of developers, ruining the excellent use-case of #IPv6 for scaled-up containers.
-
Managing containers using #nerdctl compose in an IPv6-only environment is aggravating, not least due to non-intuitive default behavior like all v6 nameservers being stripped from /etc/resolv.conf .
But I choose not to blame the Nerdctl developers, as their implementation intentionally tracks that of Docker Engine. Docker themselves are responsible for royally messing up network understanding for a whole generation of developers, ruining the excellent use-case of #IPv6 for scaled-up containers.
-
I feel I'm on the cusp of an adventure I really don't want to start. In short: #nerdctl + Compose + #IPv6
I suspect it will be awful, seeing as nerdctl duplicates the #docker commands, which only recently supported IPv6. And then there's the whole business of IPAM for containers, whether they use NAT66 and ULAs -- ugh -- or rely on SLAAC and native networking.
I wanted Compose for #k8s like orchestration but simplified for single machine operations. Why must this IPv6 journey be so hard?
-
I feel I'm on the cusp of an adventure I really don't want to start. In short: #nerdctl + Compose + #IPv6
I suspect it will be awful, seeing as nerdctl duplicates the #docker commands, which only recently supported IPv6. And then there's the whole business of IPAM for containers, whether they use NAT66 and ULAs -- ugh -- or rely on SLAAC and native networking.
I wanted Compose for #k8s like orchestration but simplified for single machine operations. Why must this IPv6 journey be so hard?
-
I feel I'm on the cusp of an adventure I really don't want to start. In short: #nerdctl + Compose + #IPv6
I suspect it will be awful, seeing as nerdctl duplicates the #docker commands, which only recently supported IPv6. And then there's the whole business of IPAM for containers, whether they use NAT66 and ULAs -- ugh -- or rely on SLAAC and native networking.
I wanted Compose for #k8s like orchestration but simplified for single machine operations. Why must this IPv6 journey be so hard?
-
I feel I'm on the cusp of an adventure I really don't want to start. In short: #nerdctl + Compose + #IPv6
I suspect it will be awful, seeing as nerdctl duplicates the #docker commands, which only recently supported IPv6. And then there's the whole business of IPAM for containers, whether they use NAT66 and ULAs -- ugh -- or rely on SLAAC and native networking.
I wanted Compose for #k8s like orchestration but simplified for single machine operations. Why must this IPv6 journey be so hard?
-
There was #rancher and #podman, now #AWS shares #finch, a new open-source solution build on top of #lima #nerdctl #containerd and #buildkit for building, running and publishing Linux containers on MacOS. https://aws.amazon.com/fr/blogs/opensource/introducing-finch-an-open-source-client-for-container-development/
-
There was #rancher and #podman, now #AWS shares #finch, a new open-source solution build on top of #lima #nerdctl #containerd and #buildkit for building, running and publishing Linux containers on MacOS. https://aws.amazon.com/fr/blogs/opensource/introducing-finch-an-open-source-client-for-container-development/
-
There was #rancher and #podman, now #AWS shares #finch, a new open-source solution build on top of #lima #nerdctl #containerd and #buildkit for building, running and publishing Linux containers on MacOS. https://aws.amazon.com/fr/blogs/opensource/introducing-finch-an-open-source-client-for-container-development/