James Gregory

Also known as: how to move hundreds of files in S3 that you accidentally put in the wrong place because you misconfigured a Kinesis Firehose.

Post http://kapacitor.default:9092/write?consistency=&db=telegraf&precision=ns&rp=autogen: dial tcp: lookup kapacitor.default on 100.1.1.1:53: no such host service=subscriber

If you’re seeing something like this error in your InfluxDB logs, and don’t know what it means: Kapacitor has created one or more subscriptions in your InfluxDB database, and InfluxDB is trying to POST to the Kapacitor endpoint; however, Kapacitor is unreachable. Kapacitor might be unreachable because it’s down, or you have a network partition or other connectivity issue, or in my case you’ve actually just destroyed your Kapacitor instance.

Are programs in your Docker container complaining of no free space?
Does your host have loads of space?
And your container does too?

It could be inode exhaustion!

Not all of our packages can be pushed to the public NPM repository. Proprietary code and uninteresting code we want to keep internal, but until recently the package distribution story for this code has been worse than open-sourcing it. You either modularise it and publish to the world, or you have a bad time.

Over the past few years I’ve moved closer and closer to virtualising my entire development environment. It started with developing for Windows whilst using a Mac, continued when I refused to sully my machine with some godforsaken Oracle product, and has now reached completion with the arrival of a new laptop and an unwillingness to install RVM again.