docs: Deployment guide rewrite #853
Conversation
carver
force-pushed
the
deployment-guide
branch
from
5bc1dab
to
540d821
Compare
|
|
||
| ## Communicate | ||
| ### Update docker images | ||
| Docker images are how Ansible moves the binaries to the nodes. Update the docker tags with: |
There was a problem hiding this comment.
It might be worth leaving a short note here about what tags are used for what. eg.
Our CI pushes the
latest&latest-bridgeimages to the docker registry for every pr that's merged into master. Our network uses thetestnet&bridgetags to build from. So, to deploy, we need to pull the latestlatest&latest-bridgetags, rename them, and then push the newly tagged images to the registry.
Ok, that might be a little wordy. But, I'm just thinking it might be helpful to explain why we do these steps so people can have a reference in case something unexpected happens.
| Notify in Discord chat about the new release being complete, and the network nodes being updated. | ||
| ### Run ansible | ||
| - Check monitoring tools to understand network health, and compare against post-deployment, eg~ | ||
| - [Glados](http://glados.ethportal.net/content/) |
There was a problem hiding this comment.
I think adding the trin-bench site here (https://trin-bench.ethdevops.io/login) to evaluate pre/post deploy network health is also valuable.
There was a problem hiding this comment.
👍🏻 I'll put a link straight to the trin metrics dashboard. Some follow-up work would be cool to write up some hints about what to look for in Grafana, and link to that here.
| You might see this during a deployment: | ||
| > fatal: [trin-ams3-18]: UNREACHABLE! => {"changed": false, "msg": "Failed to connect to the host via ssh: ssh: connect to host 178.128.253.26 port 22: Connection timed out", "unreachable": true} | ||
|
|
||
| Retry once more. If it times out again, ask `@paulj` to reboot the machine. |
There was a problem hiding this comment.
Ooooh, nice! Did this actually get resolved? I'm quite familiar with that machine always failing and typically just let it fail since it was the only one that failed, so i'm glad to see that it was resolved!
There was a problem hiding this comment.
Well, "resolved" as in he rebooted it, and the next deploy succeeded. But, maybe it will fail again at the next deploy? 🤷🏻♂️
After deploying last week
carver
force-pushed
the
deployment-guide
branch
from
540d821
to
1a45409
Compare
Fleshed out a lot of instructions, after deploying last week
Fixes #846
To-Do