actionowl

We've been VERY slowly moving some legacy applications from Rackspace over to AWS. After we migrate one and we're sure everything is good we delete the old VM. We've been doing this all of 2019 and 2020 (probably even a little in 2018). Never once have I had to call to get a VM deleted until today where I received a message saying I had to call them. This involved getting routed to "the team that handles these", then giving my PIN (which is in the web console), waiting on hold, and typing "I agree" in a support ticket they opened. Total waste of time. They claimed additional security but I was able to to do all this from the web console after I called them. They also suggested that I just shut it down for a few days in case there were some weekly jobs running I might care about...

We're down to only 4 VMs at this point and I have a feeling this is an attempt to hold onto some customers. Anyone else had this experience?

I've been unable to reach www.tinycorelinux.net for a couple of weeks now. Anyone know what is going on with their site?

I've been wanting to try out Tiny Core. I'm looking for a very lightweight Linux distribution that comes with an up-to-date browser. That's essentially all I need (boot to a browser, preferably a live CD of some kind). Essentially I'm tired of struggling with rust on a system that is built from sources and want to delegate web activity to a VM.

I don't know if Tiny Core will fill that need but I've been unable to even try and now I'm curious.

At my current company I lead the SRE team, but previously had been a developer and worked on a couple of products, one that ran in production for about 8 years (and is still running, in maintenance-only mode.)

The new development team has decided to write a new system to replace the old one and they frequently ask us to run ad-hoc SQL statements to clean up records and patch up issues caused by bugs.

In my ~8 years of working on the existing app, I can probably count on one hand the number of times we needed to run ad-hoc SQL scripts against production data (almost always after importing data from an other system), everything else was handled with migrations.

I'm seeing about 3-5 ad-hoc scripts needing to be run per-release. Just curious if this is at all normal or if I'm just expecting too much.

If you're wondering why I might have a problem with this at all, it's because at least 50% of the time there is also a problem with the SQL script and it either doesn't do the right thing or it has syntax errors. It feels like it was just "thrown over the wall".