Laravel 13.30, weekly updates and tip

Laravel 13.30

This week brings us to Laravel 13.30. Here are the highlights:

  • Allow retrieving command by name without resolving all commands in #61361
  • Add chunkBy to collections in #61357
  • Add macroable to CloudManager in #61311
  • Add NotificationSkipped event in #61331
  • Add createPayloadUsing directly to the queue manager in #61367

You may review the full branch diff on GitHub for a complete list of changes.

Weekly Journal

Last week I livestreamed trying to understand using an MCP server. Admittedly I didn't get very far. Laravel, of course, offers a first party package (laravel/mcp). But the terminology and architecture is a bit over my head. So, I'll carry on in this week's livestream.

I got pretty sick Thursday and didn't really feel better until yesterday. So I didn't do much during that time. Just rested. Tried to help with the girls when I could. Watched a little Ted Lasso to keep the spirits up.

Today was mostly wasted on a Digital Ocean bug. When a new release is tagged, I spawn dynamic workers. Today, I received a 422 that the droplet sizes didn't exist in my region. Of course, there was nothing documented about this in their changelog or status page. But it wouldn't be the first time a service changed things without notice. For example, last month GitLab apparently moved to JWT tokens.

To relieve the queue workload, I temporarily bumped the size of my snapshot. Which cost a few more cents per hour. Not a big deal in absolute terms, but relatively it was double the cost of the original size.

Now I'm all about efficiency and minimalism. So I sent Claude down a rabbit hole to "crunch the numbers" on the server specs under the automation workload, per droplet pricing, and queue throughput. It did its calculus and determined that a larger box with more vcpus would actually be cheaper. So, I had it build that.

By the time I verified its work and was ready to rebuild the snapshot, my dynamic workers appeared. So it seemed to be some kind of temporary outage or bug. We don't think this stuff happens with these big services. But it happens more often than you think. Look at Cloudflare and GitHub in recent months. At the end of the day, they're developers too - making the same mistakes we do.

Weekly Tip

"Don't make mistakes."

View Archives →