jonjonsonjr

4 ms·

1.
by jonjonsonjr 1y ago
Something must be in the air. I've been working on a gzip/deflate visualizer recently as well: https://jonjohnsonjr.github.io/deflate/ This is very work in progress, but for folks looking for a deeper explana
2.
by jonjonsonjr 1y ago
Sorry about that. I usually don't let it scale up at all so that I don't have to worry about costs. I've raised some limits for now so hopefully you can get through.
3.
by jonjonsonjr 1y ago
Hi, this is my project! I was surprised to see it posted here. It's a debugging tool for container images that I host for myself and some friends/coworkers. Some of its features are intentionally not very discoverable, partially t
4.
by jonjonsonjr 2y ago
Even worse, in the general case, you should really decompress the whole tarball up to the end because the traditional mechanism for efficiently overwriting a file in a tarball is to append another copy of it to the end. (This is similar to
5.
by jonjonsonjr 2y ago
I tend to agree with this sentiment but this year I came across two situations where the use of cat is actually harmful and not just useless: https://github.com/jonjohnsonjr/til/blob/main/post/readat
6.
by jonjonsonjr 2y ago
The proof is that they can tell you the number of leaves (again) after you have secretly removed some. Since only you know the number you removed, if the difference between their counts matches your number, it is likely that they can indeed
7.
by jonjonsonjr 3y ago
We have built something very similar to what you are describing: https://github.com/chainguard-dev/apko
8.
by jonjonsonjr 4y ago
I gave a very compressed lightning talk last year about this: https://youtu.be/ExyWAhS2zBA
9.
GKE image streaming for fast application startup and autoscaling(cloud.google.com)
11 points by jonjonsonjr 5y ago|4 comments
10.
by jonjonsonjr 5y ago
Digests cryptographically guarantee that you get the correct content, which prevents both malicious tampering (mitm, stolen credentials, etc) or accidental mutations. This is why "immutable tags" are a bad substitute and an oxymor
11.
by jonjonsonjr 5y ago
Can you explain why? I've found that it is a lot simpler to use git bisect if every commit builds.
12.
by jonjonsonjr 5y ago
I don't think I'd ever call a Dockerfile declarative.
13.
by jonjonsonjr 5y ago
Kind of similar idea to emojisum https://github.com/emojisum/emojisum
14.
by jonjonsonjr 5y ago
Nice! I was thinking about building something similar -- just filed an issue for how you might extend this to work for migrating registries. Pleasantly surprised to come across this PR: https://github.com/safe-waters/do
15.
by jonjonsonjr 6y ago
> Comparing image digests won’t work; they will never match. This is a strong assertion with no further explanation. It reads like a generic truth about container images, but it's certainly possible to achieve this, as referenced la
16.
by jonjonsonjr 7y ago
Your first point is incorrect. That was true of v1 docker images, but layers have been content-addressable for a while now. Your second point is absolutely correct - we strip timestamps from everything which tends to confuse folks :)
17.
by jonjonsonjr 8y ago
If jib doesn't work against ECR, please file a bug. There isn't anything specific to Google here -- it's all based on standard container tooling. The only thing you should need to do is configure the ECR credential helper: h
18.
by jonjonsonjr 9y ago
You might be interested in distroless[1] base images. The repo links to a talk that goes into more depth, but the basic idea is to a use minimal language-specific base for your runtime instead of e.g. statically linking all of ubuntu into y
19.
by jonjonsonjr 10y ago
This thread might be interesting to you: https://groups.google.com/forum/#!topic/kythe/Nwt9TrefyZI
20.
by jonjonsonjr 11y ago
This is from 2012.
21.
by jonjonsonjr 12y ago
Are these related? https://github.com/garybernhardt/raptor