A while back I was developing a workflow scheduler. However the run duration kept coming back negative. The runs kept finishing like 15 seconds before they started. Eventually I figured out why; the company had configured a policy to use their custom NTP server, but it was 16.7 seconds off. So my Windows laptop's time was in the future, but the WSL VM (which ran the program) used another NTP server and was correct. In the end I just added a bodge that made the program request the time from an external server on startup to calculate an offset.

you are viewing a single comment's thread
view the rest of the comments
[–] 5 points 8 hours ago

3rd line support for a global IT brand you have heard of, on the kind of security product that people don't like to mess with too much once it's installed on their servers.

Bug had been in the tracker over 3 years when I joined, and had defeated anyone who tried it. Segfault that happened to only one customer, unpredictably, and only once or twice a year, and nobody could reproduce it.

After a few weeks of tracing the compiled code line-by-line I found a race condition in the teardown of an object. That's not the silly part.

I wrote the patch for the version of the code the customer was running and was hailed as a mighty hero and everyone on the bus clapped etc. Then I went to forward-port the patch to the newer versions of the program and... it was already fixed there. Some fucken doofus had either fixed the bug without realizing, or, fixed it and just skipped the standard backporting process that would normally patch all the other versions that were still in support.

  • source