Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)GA
Posts
2
Comments
93
Joined
2 yr. ago

  • I'm sold on the sales pitch for it, but deflect.ca resolves to a Hetzner box in their Washington DC datacenter. I have a 112 ms ping to it from Toronto.

    How do you sell a CDN when your own website is hosted out of the country at a PoP that's that far away?

  • If I can ramble a bit more - forget the Anycast bit. If you run your own DNS server(s), you can just configure them to respond based on the geographic location of the requester. PowerDNS is pretty easy to set up for this. You could run your own DNS just for the image domain. You basically run PowerDNS authoritative server, set up your zones and the geoip stuff, then slap dnsdist in front of it to be publicly exposed. dnsdist has anti-DDoS features and loadbalancing in it, in case you need it down the road.

    Since it's just for static images, you can have a higher TTL so you don't need to worry about distributing the DNS servers. (ie. the DNS lookup might not be super fast since it could go across the country, but it doesn't matter since that lookup is only going to happen every TTL period on each client, which can be high.)

  • One suggestion to consider for Lemmy.ca is to move your images and other easily-cacheable content to a different domain or subdomain, to give you more flexibility.

    eg. If you serve your static assets off of lemmyimages.ca, then you can have only that behind a CDN, Cloudflare, or some other hosting with DDoS scrubbing. It gives you more flexibility to cope with various situations.

    2tb a week isn't much (6 mbps on average?). It's pretty easy to set up nginx as a caching reverse proxy and spin that up on a couple of VPSes, but the annoying bit is you need to anycast your own IP address space in order for it to be functional as a CDN.

    I'm not aware of any Canadian-owned CDNs either... OVH has one but they're pretty crappy as a company. Beware of whitelabelled CDNs too, even some of the CDNs provided by big cloud hosting companies are actually whitelabelled from another company.

  • It's not about software or data. It's about control over the supply chain - cars are essential to our economy and way of life in North America (like it or not). It's the same reason we protect the milk supply. You don't want another country to be able to turn it off in a conflict.

  • The thing is, if Trump wants to kill Canada's role in US car manufacturing, then it will cost him the car markets in Mexico and Canada. If there's no jobs here to protect, then we'll just drop the tariffs on Chinese EVs. (This is speaking like 20 years down the road). We'll all be driving Chinese cars in that scenario. The tariffs are a total lose-lose situation, so dumb.

  • It's great that you've pointed this out and I hope there's more awareness about it. In practice, it's not hard: Buy from candian-owned small businesses who manufacture in Canada.

    In my not-at-all-humble opinion, most of your examples are all shit you shouldn't be buying in the first place:

    • We have copious amounts of local craft beer. Never buy the big brands, they're all swill. If you actually drink craft beer, you'll know which brands, like Creemore and Millstreet, are fake craft.
    • Nobody should be buying anything from Coca-Cola in the 21st century. We've known pop is terrible for your health for like 50 years. They're a shit company who is the biggest polluter of plastics in the world.
    • Canada Goose is for tools with no taste. By the time any trendy fashion company gets bought out, it's not cool anymore.

    All we have to do as a nation is just put that little extra effort into learning about what we're buying and making different choices, and it's actually great that we're all doing that because we should have been doing it all along.

  • Yeah, this might be the way to go. OpenWRT supports hardware NAT with many of these ARM-based routers like many of the MediaTek-based ones, which gives them super high throughput at very low CPU usage. The efficiency blows x86 out of the water. The ability to migrate your OpenWRT config to new hardware (real or virtual) in the future means you kinda get the best of both worlds....

  • Do not use an SSD for cold storage - it will fail. SSDs need to be plugged in every once to refresh the charge in their NAND, otherwise they'll lose the data.

    This is not a theoretical thing - I've had a good Samsung 850 Pro drive fail while being off for 2 years.

  • Even if the virtualized router is down, I'll still have access to the physical server over the network until the DHCP lease expires. The switch does the work of delivering my packets on the LAN, not the router.

    Thanks for the tip about the pfSense limit. After running pfSense for like 8 years, my opinion is that is flush with features but overall, it's trash. Nobody, not even Netgate, understands how to configure limiters, queues, and QoS properly. The official documentation and all the guides on the internet are all contradictory and wrong. I did loads of testing and it worked somewhat, but never as well as it should have on paper (ie. I got ping spikes if I ran a bandwidth test simultaneously, which shouldn't happen.) I don't necessarily think OpenWRT is any better, but I know the Linux kernel has multithreaded PPPOE and I expect some modern basics like SQM to work properly in it.