March 25, 2026 at 12:00 am
Comments posted to this topic are about the item Doing the Little Things
March 25, 2026 at 10:03 am
I inherited a data pipeline with a daunting amount of infrastructure, code, database artefacts, and huge volumes of logs.
I found that the CICD pipeline was creating artefacts for each run but only cleaning up a handful afterwards, and what cleaning it did was fragile.
Eventually, I discovered that a large part of the code base was old, dead code. I lost a lot of time trying to understand this code base and that time could have been better spent. For this reason, I'm a big fan of code and infrastructure hygiene.
In a DR situation, you want to be recovering useful systems, not zombies. In a messy estate, how do you know what is a zombie and what is not?
There was a post on LinkedIn for ex-service personnel asking what has stayed with them from their time in the armed forces, even if they left decades ago. An ex-Royal Marine offered up "today a top button, tomorrow a submarine hatch". All the comments on the thread spoke of attention to small details and disciplines that remain hard-wired years later.
March 25, 2026 at 2:53 pm
If you ever want to see a serious FOD walk, look up videos of aircraft carrier crews doing an FOD check of the flight deck. Crewmen lined up across the width of the deck nearly shoulder-to-shoulder and 3 or 4 rows deep taking slow steps.
I've automated a fair number of the "little things" that need to be handled to keep the servers happy. Audit files are handled by a daily batch job, backups are stored in Azure Blob storage with a Storage Lifecycle Policy on the container to migrate them through the storage tiers and eventually remove the old ones, Redgate Monitor warns when a drive starts getting full up and SQL Agent Alerts backstop Monitor (belt and suspenders,) and I still go and manually check on occasion.
While I've never been in the military, I think I picked up the habit of paying attention to the "fiddly bits" when I would build my own PCs, way back in the day when you had to set jumpers to tell the motherboard what processor you had, and if you set them wrong, best case the PC wouldn't boot up or would be under-clocked, worst case you fried your rather expensive CPU or RAM.
March 25, 2026 at 4:47 pm
Eventually, I discovered that a large part of the code base was old, dead code. I lost a lot of time trying to understand this code base and that time could have been better spent. For this reason, I'm a big fan of code and infrastructure hygiene.
...
Code hygiene is certainly a little thing that helps.
March 25, 2026 at 4:48 pm
If you ever want to see a serious FOD walk, look up videos of aircraft carrier crews doing an FOD check of the flight deck. Crewmen lined up across the width of the deck nearly shoulder-to-shoulder and 3 or 4 rows deep taking slow steps.
I've automated a fair number of the "little things" that need to be handled to keep the servers happy. Audit files are handled by a daily batch job, backups are stored in Azure Blob storage with a Storage Lifecycle Policy on the container to migrate them through the storage tiers and eventually remove the old ones, Redgate Monitor warns when a drive starts getting full up and SQL Agent Alerts backstop Monitor (belt and suspenders,) and I still go and manually check on occasion.
While I've never been in the military, I think I picked up the habit of paying attention to the "fiddly bits" when I would build my own PCs, way back in the day when you had to set jumpers to tell the motherboard what processor you had, and if you set them wrong, best case the PC wouldn't boot up or would be under-clocked, worst case you fried your rather expensive CPU or RAM.
I saw something similar in Hawaii. A friend was a pilot and let us walk across the runway, where they have FOD area signs and reminders to be careful. He showed us how they'd do this with just 3-4 of us, but said every day they would have dozens of airmen doing it.
March 26, 2026 at 5:29 pm
My attention to little things is a constant source of irritation to my wife. Forty-two years in IT instilled habits of noticing details, considering the source, anticipating the potential effects, analyzing the risks, and establishing preventive solutions.
Rick
If you do a half-assed job of things, folks will ask 'why did this ass only do half the job?'
March 29, 2026 at 1:52 pm
I'm not a DBA, so initially I thought this editorial wouldn't apply to me. But I believe I am wrong. One of my jobs is being one of our GitHub Administrators. At this point we don't use what's called GitHub Runners (CI/CD that run on GitHub.com). Instead, we use what's called GitHub Self-Hosted Runners, which means they're GitHub runners that run on our internal servers. I configured one to do this 3 years ago.
Anyway, I had been under the mistaken impression that these self-hosted runners clean up after themselves. Recently I've learned they might not. The server it runs on has a lot of disk space, so we've been fine. So far. Thank you for suggesting writing some routines to clean up older artifacts.
Kindest Regards, Rod Connect with me on LinkedIn.
Viewing 7 posts - 1 through 7 (of 7 total)
You must be logged in to reply to this topic. Login to reply