Hacker News: Tractor8626

New comment by Tractor8626 in "Anthropic becomes the 'Apple of AI': Most revenue despite being most expensive"

Tractor8626 — Mon, 17 Aug 2026 18:29:37 +0000

To claim "highest revenue" you need to compare with other companies not with itself in the past...

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Sun, 04 Jan 2026 07:28:22 +0000

Maybe your magic ipv6 configures everything.

My have certain shortcomings: it doesn't assign dns names to hosts and doesn't configure firewall rules

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Sat, 03 Jan 2026 07:53:54 +0000

Yeah. ULA and nat66 would work nicely. Except you would get murdered for asking about nat66.

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Sat, 03 Jan 2026 07:43:49 +0000

How do you setup dynamic dns in your network? Which software do you use?

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Sat, 03 Jan 2026 07:34:48 +0000

No. People fail to configure it. Then start to question why this piece of garbage exist.

It doesn't matter what problems it supposed to solve if it doesn't work.

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Sat, 03 Jan 2026 06:21:19 +0000

One would think that in 30 years there will be some sort of best practises established. Some articles to refer people to. Or at least some people to share their experience and answer practical questions.

And yet there is still only "you doing it wrong, and I won't tell you how to do it right"

New comment by Tractor8626 in "IPv6 just turned 30 and still hasn't taken over the world"

Tractor8626 — Fri, 02 Jan 2026 17:03:23 +0000

Is there yet answer to question "how to get random self-assigned addresses into dns records, firewall rules and switch acls?" ?

New comment by Tractor8626 in "100k TPS over a billion rows: the unreasonable effectiveness of SQLite"

Tractor8626 — Tue, 02 Dec 2025 22:52:26 +0000

> If you're processing financial transactions you want your isolation level to be serialisable. As the order in which the transactions are processed matters.

So you don't know what serializable level is.

> Even with 1ms latency. Amdahl's law will still make you cap out at a theoretical 1000 TPS if you have 100% row lock contention.

So why use 5ms and 10ms for examples?

> I'm confused. I invented transactions?

"Interactive" transactions.

New comment by Tractor8626 in "100k TPS over a billion rows: the unreasonable effectiveness of SQLite"

Tractor8626 — Tue, 02 Dec 2025 21:00:01 +0000

Sqlite is very cool. But what was the point of slowing postgress down?

> But, wait our transactions are not serialisable, which they need to be if we want consistent transaction processing

You either don't know what serializable does or trying to mislead the reader. There is zero reason to use searializable here.

> Let's say you have 5ms latency between your app server and your database.

5ms latency is unrealistic. Unless you use wifi or you database is in another datacenter.

> I'm talking about transactions per second, specifically interactive transactions that are common when building web applications

No they are not common at all. You probably invented them just to make pg look bad.

New comment by Tractor8626 in "Why Strong Consistency?"

Tractor8626 — Thu, 27 Nov 2025 21:52:22 +0000

> read-modify-write is the canonical transactional workload. That applies to explicit transactions (anything that does an UPDATE or SELECT followed by a write in a transaction), but also things that do implicit transactions (like the example above)

Your "implicit transaction" would not be consistent even if there was no replication involved at all. Explicit db transactions exist for a reason - use them.

New comment by Tractor8626 in "Self-hosting a NAT Gateway"

Tractor8626 — Sat, 22 Nov 2025 15:57:39 +0000

> you can absolutely NAT without preventing the "outside" subnets from being allowed to route to the "inside" subnet

Under very specific conditions. Technically if you send packet with destination 192.168.1.10 directly to wan port of router - yes it can route it inside. The problem - how to deliver this packet over internet. You need to be connected to exactly same network segment to pull it off.

And you don't need statefull firewall to deny this kind of packets.

New comment by Tractor8626 in "Self-hosting a NAT Gateway"

Tractor8626 — Sat, 22 Nov 2025 03:22:36 +0000

No. We can't. We encouraging it because it works.

New comment by Tractor8626 in "Self-hosting a NAT Gateway"

Tractor8626 — Sat, 22 Nov 2025 03:19:03 +0000

But NAT acts as a one way door to your private subnet, doesn't it?

New comment by Tractor8626 in "Kubernetes Ingress Nginx is retiring"

Tractor8626 — Fri, 14 Nov 2025 06:58:36 +0000

Cybersecurity is easier? Isn't it all about constantly updating and patching obsolete vulnerable stuff - most annoying part of ops?

New comment by Tractor8626 in "WebDAV isn't dead yet"

Tractor8626 — Sat, 25 Oct 2025 04:30:51 +0000

If you need sftp independent of unix auth - there is sftpgo.

Sftpgo also supports webdav, but for use cases in the article sftp is just better.

New comment by Tractor8626 in "AWS multiple services outage in us-east-1"

Tractor8626 — Mon, 20 Oct 2025 21:24:55 +0000

They supposed to exceed their SLA. SLA is guaranteed worst case.

New comment by Tractor8626 in "The great software quality collapse or, how we normalized catastrophe"

Tractor8626 — Thu, 09 Oct 2025 15:17:02 +0000

Just false nostalgia memory.

20 years ago things werent any better. Software didn't consume gigabytes of ram because there was no gigabytes of ram to consume.

New comment by Tractor8626 in "Blockdiff: We built our own file format for VM disk snapshots"

Tractor8626 — Wed, 01 Oct 2025 16:53:31 +0000

Interesting tool. Something like btrfs send/receive but on a file level and fs agnostic.

New comment by Tractor8626 in "A Postmark backdoor that’s downloading emails"

Tractor8626 — Sat, 27 Sep 2025 21:00:51 +0000

Looks like bcc was added for debugging and was not removed before commit.

Too obvious for backdoor. Replacing bitcoin addresses in email would be more useful)

New comment by Tractor8626 in "The Apache Software Foundation's New Logo"

Tractor8626 — Fri, 12 Sep 2025 04:23:51 +0000

The new logo is cool asf