miller_joe

I’ve been looking for something like this for Claude

I was hoping google cloud artifact registry pull-thru caching would help. Alas, it does not.

I can see an image tag available in the cache in my project on cloud.google.com, but after attempting to pull from the cache (and failing) the image is deleted from GAR :(

Strong flashbacks of Southland Tales (2006). Underrated movie with this as a core premise

Instances are constantly booting up because most instances live <30d. Boot time in terms of how soon a node is fully booted and joined to the EKS apiserver and ready for workloads is approx 2.5-3min. There are lot of parts involved in getting to this point though, some of which would not matter if you're not using EKS. Also this is not something we measure super closely as from a user perspective it is generally imperceptible.

A possibly better metric for your particular case (assuming you're interested in fastest bootup possibly achievable) is from our self-managed github-actions runners. Those boot times are in the 40-50s range. This is consistent with what others see, as far as I know. A good blog on this topic - including how they got boot-to-ready times down to 5s - that you might be interested in from the depot.dev folks: https://depot.dev/blog/github-actions-breaking-five-second-b...

email sent

As a declarative process supervisor i really like systemd. I don't like all of the other things it wants to do though.

Love this idea! ty

I wasn't sure how to load the images back into docker at first. I tried `docker load` but I get this error:

    $ (cd ci-repack && tar cfv - .) | docker load
    ./
    ./oci-layout
    ./index.json
    ./blobs/
    ./blobs/sha256/
    ./blobs/sha256/2ad6ec1b7ff57802445459ed00e36c2d8e556c5b3cad7f32512c9146909b8ef8
    ./blobs/sha256/9f3908db1ae67d2622a0e2052a0364ed1a3927c4cebf7e3cc521ba8fe7ca66f1
    open /var/lib/docker/tmp/docker-import-1084022012/blobs/json: no such file or directory

Then I noticed the `skopeo copy` in one of the github actions workflows. That got me further. The image was able to be pushed to a registry. But I am getting this error when pulling the repacked image:

    failed to register layer: duplicates of file paths not supported

This is great! I’ve been using ipv6 on openbsd for a while now, starting with a hurricane electric tunnel years ago then to native v6 on Comcast and now Sonic. Configuring ipv6 PD has not been supported in base this whole time. I recall using wide-dhcp6c years ago and then switching to the dhcpcd in the article. The situation has improved, slowly, but it will be great to have this in the base system

*EDIT*: dhcp6leased landed in base yesterday: https://www.undeadly.org/cgi?action=article;sid=202406040850...

Vector is fantastic software. Currently running a multi-GB/s log pipeline with it. Vector agents as DaemonSets collecting pod and journald logs then forwarding w/ vector's protobuf protocol to a central vector aggregator Deployment with various sinks - s3, gcs/bigquery, loki, prom.

The documentation is great but it can be hard to find examples of common patterns, although it's getting better with time and a growing audience.

My pro-tip has been to prefix your searches with "vector dev <query>" for best results on google. I think "vector" is/was just too generic.

A nice recent contribution added an alternative to prometheus pushgateway that handles counters better: https://github.com/vectordotdev/vector/issues/10304#issuecom...

I’ve been using this for a few weeks. I converted to a paying customer. I’ve imported some but not yet all of my bookmarks.

I’ve been collecting bookmarks in Evernote and now obsidian for about a decade. I try to add as many tags as I hope will allow me to find an article again later. It’s often many months before I need to find something. My success rate at remembering a term from the title or the right tag is not great. I’ve been pretty impressed with zenfetch’s ability to search and find exactly what I was looking for. And this doesn’t even scratch the surface of what it can do when you want it to synthesize answers from many articles for you.

They’re also working on indexing your saved tweets which I’m excited about. It’s a giant pain to try to find liked tweets.

Edenists.

MLM is the Commonwealth Saga, if I recall. First book “Pandora’s Star”

I just happen to be reading Peter F. Hamilton's Confederation universe (1996–2000) trilogy and in it earth has been dealing with the "Armada storms" for a couple hundred years due to climate change. From the descriptions there are multiple of these storms occurring at any given time. Humans now live in "arcologies" which are giant multi-kilometer domes over population centers. There was no way to reverse the damage done. We can't fly anymore so everything is in a very fast underground train system called "vac-trains". I suppose this may be where we are headed (although I'd rather love to see the vac-train system)

I love the idea. I signed up and now I'd like to import a bunch of bookmarks I have stored in Evernote and Obsidian over the years. I can't find a way to do this though. Is there a way to import all of these?

I suppose I could try the Chrome bookmarks importer if I can import everything into chrome. But I can't find the import option after the initial signup workflow.

Looks great. Another tool in this space is https://github.com/humanlogio/humanlog

I've been experimenting with working in a gcp VM for a few months and like it quite a bit. Between ssh+tmux+nvim and vscode-remote I have a remote and local dev exp that is basically identical.

Syncthing is my key to keeping my ~/git tree in sync between the two environments. Just have to be careful not to try to flip back and forth quickly between local/remote as syncthing delays can be ~10s.

I’ve been using atuin for a while and have been thinking about something like this. Please do post to GitHub!

This is great! One small suggestion: try it with the newer veb bridge device. It should be a bit faster.

no, i am able to saturate a 2.5gbps connection with NAT and filtering with or without ignoring the interrupt storm. Perf was never an issue, just higher temps (it's a fanless box) and probably energy draw (i never measured that so i am not sure how much)

greetings fellow traveler. I am also a big fan of openbsd for home router/firewall. Been running it since the early 2000s.

I've gone through a lot of hardware in that time. From an old dec alpha (kernel panic'd often) to only x86 these days.

In recent years I've started to use these low-cost, fanless, intel machines from aliexpress. They're usually pretty great. Recently I upgraded (ServeTheHome's reviews were very helpful!) from a box I got in 2018 from aliepxress that had 6x 1gbe ports to a newer box with 2.5gbe since my fiber ISP has 10gbe ports right now (and there are no cheap, fanless, low-power intel boxes with 10gig nic's). Specifically a Celeron N5105 (N6005 was not available earlier this year, so I canceled and ordered N5105) with 4x 2.5gbe nic's, this one[1] from Topton, though I believe other vendors on aliexpress are all selling basically the same thing.

The unit works pretty well, however, I noticed with OpenBSD 7.0 and 7.1 that one CPU core was constantly pegged at 100% servicing interrupts. Running `top -S` showed the acpi kernel thread was the culprit, and `systat` showed `acpi0` as the source of interrupts. The box worked fine and could route (with NAT and filtering) the full 2.5gbps internet connection fine, but it would run hotter and probably use more electricity than needed.

I enabled ACPI_DEBUG which did not make for a usable box at all due to the amount of debug prints being generated but it was enough to verify the interrupt spam coming from the ACPI GFE _L6F handler. A google search shows this is not particularly uncommon and is due to buggy motherboard firmware. Both Linux and FreeBSD have a way to disable interrupts on a live system. OpenBSD does not however. And FWIW, I booted an opnsense live CD based on FreeBSD and this interrupt storm was not happening. So I assume FreeBSD may have seen this particular bug or just happens to handle it already. This box should be fine with any of the FreeBSD based OS's. Which is often bundled on these firewalls you buy from aliexpress.

The usual fix for these interrupt storms from buggy BIOS's is an eventual fix from the vendor. In this case though I don't expect to ever seen a BIOS update for this box, and there really is not a website to go download one from anyway. The vendor on aliexpress assured me that if they ever release a new BIOS that they would tell me. I doubt this will ever happen.

In any case, I have "solved" the issue finally with a small kernel patch I found on the excellent openbsd mailing lists. The issue in that case was different, older hardware, but the symptoms were the same.

Long story short I am posting all of this here so that if someone else runs across this same issue they will be able to resolve it much faster than I was.

Patch for OpenBSD 7.1 kernel to ignore the L6F GFE interrupt storms: https://gist.github.com/joemiller/9f5698c5634d4a93d101985dc5...

[1] https://www.aliexpress.us/item/3256803807107706.html

i am also wondering about this and happen to be in the middle of testing some L2 tunneling (etherip mostly, might also test vxlan or egre) with ipsec and wireguard in a set of VM's on virtualbox.

the SMP improvements in the networking stack caught my eye and i am in the process of trying to upgrade these vm's and re-test (iperf3). if i find anything interesting i'll report back

I also ran into this with Comcast ipv6. It seemed to get better over the last couple years so maybe they change your PD less frequently now.

One solution I was considering, but never implemented, was to use a ULA addres range internally and doing 1:1 NAT mapping on the external delegation to the internal ULA range

I learned more from this comment than the article, and I thank you

Hey, cool idea. I like it. I made a Go implementation for fun/practice - https://github.com/joemiller/prefixed-api-key

This is great!

Do you recall the name of that monitor from the 90s? Sounds interesting

Can it run Messages.app? This is the one thing that constantly gives me pause on the linux-desktop switching

EDIT: It says "some" GUI apps are working / experimental, so probably Messages is not there yet, just curious if anyone has tried.

great job! the results on the images i tried were impressive. i feel like i have tried similar tools in the past and don't recall the results being as good.

amazing. that's my feedback =)

https://allma.io/ is also in the 2nd category and is a pretty awesome product so far, IMO.

I've been using TabNine for several years and love it. I'd be interested in hearing a comparison between TabNine and GH Copilot from anyone who has used both.