• 0 Posts
  • 36 Comments
Joined 4 months ago
cake
Cake day: January 20th, 2025

help-circle
  • Since they handle redundancy and backups I think it’s fine staying with them (+ great product)

    This. I love self hosting services, but anything that I 100% can’t live without isn’t one of them. Because I don’t have the funds for proper redundancy/high availability, and my backup practices at home are… Not ideal. I’ve had a couple brushes with data loss due to gaps in backups, lack of monitoring for impending hardware failures, and had 2 disks suddenly die together in a raid array, all in over a decade of self hosting.

    I have cold backups of most of my critical services, but they’re not nearly regular enough for me to trust my passwords to myself.







  • That’s what I do at work, even though I’m salary.

    Management decided to hire a new guy and then have a round of layoffs within 6 months, effectively canning someone to replace him. Since then, we’ve had multiple times where we have hundreds of tickets sitting unassigned because there’s more work than people. So shit sits and falls through the cracks until someone has time or something is on fire.

    It fucking sucks, but eventually the bean counters will see that we actually needed that extra body…


  • It all depends on how time tracking is implemented.

    Tell me I have to account for my time in 15-30 min increments? Fine, I can put it into a spreadsheet and track it.

    Tell me I have to track real time spent? Get entirely fucked, and I hope you’re ok with spending time fixing my time punches because I absolutely am going to forget to open or close a time entry because I’m working on 3-4 tasks at any given time.

    I’ve done both, and while I won’t intentionally sabotage the latter, my rampant ADHD and terrible memory have got my back on that one.



  • Maybe there’s more implementation nuance here but it seems like an opinionated rule that has zero effect on performance unless that code is being called thousands of times every second

    It’s good practice to get in the habit of coding to only do the things you want/need to do rather than hoping the compiler does it for you.

    This particular constructor call may be light, but there may be constructors that have a lot of overhead. Or you might be running alongside 1000 other processes who said the same thing and you start to see performance degradation.






  • Do you remember what you fixed when you fixed it on the window side? Asking because what you’re describing almost sounds like you have a bad driver, which would explain why your Linux side would also have a similar problem, IE locking up completely auddenly, if it had the same bad driver and interacted with the hardware the same way causing a similar crash.

    Honestly, if it’s fixable in the windows it’s definitely fixable Linux. It just might take a little bit more extra work to figure it out.





  • And depending on your audience or how often you (don’t) go back and review it, it can help to comment the things that are self explanatory.

    I write a lot of scripts for my team and have to make them “maintainable” by the people in my department (who are as familiar with the concept, as your buddy that calls the gym the James), so I will regularly over comment so that any ape can come along and hopefully know what the script does.


  • It’s not like the Java server is slow tho, it only becomes a problem when mods are added

    Lol, the java server is dogshit slow, it just doesn’t show until you have a big enough map if you’re not running mods. I played on a semi-public server and was a mod back in the day, and wed regularly have to do world resets because it started slowing down when people got far enough away from the spawn. Hell, even with mods it doesn’t start to slow down immediately, it waits until you’ve done a lot in your world before it starts chugging like dad after work.