Our great sponsors
-
-
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
>As now we've introduced some peers code review, automatic testing on most critical stuff (but since the codebase sucks these aren't really reliable tests)
They may not be "reliable", but these are your safety net, or harness, so you don't fall. I wrote about similar issues, for instance here: https://news.ycombinator.com/item?id=26591067 and, given your promotion, here: https://news.ycombinator.com/item?id=37211796. It contains a few steps starting from "So...".
You can add monitoring, something like Sentry (https://sentry.io) will capture exceptions that were not handled that you have not seen because the stack trace is buried in hundreds of pages of logs or something. It groups them by exception and counts them. It's pretty awesome. (https://docs.sentry.io). It supports around 108 platforms (Java, Python, JavaScript, etc.). This lets you see the exceptions and makes prioritizing easier (which ones are the most frequent, which ones impact the most, etc.).
If you don't have them already, issue templates are really useful and the comment I linked to explains why, but here's an example of an issue template (again, you can configure them for different types of issues so team members select from a dropdown for a bug or a feature):
>As now we've introduced some peers code review, automatic testing on most critical stuff (but since the codebase sucks these aren't really reliable tests)
They may not be "reliable", but these are your safety net, or harness, so you don't fall. I wrote about similar issues, for instance here: https://news.ycombinator.com/item?id=26591067 and, given your promotion, here: https://news.ycombinator.com/item?id=37211796. It contains a few steps starting from "So...".
You can add monitoring, something like Sentry (https://sentry.io) will capture exceptions that were not handled that you have not seen because the stack trace is buried in hundreds of pages of logs or something. It groups them by exception and counts them. It's pretty awesome. (https://docs.sentry.io). It supports around 108 platforms (Java, Python, JavaScript, etc.). This lets you see the exceptions and makes prioritizing easier (which ones are the most frequent, which ones impact the most, etc.).
If you don't have them already, issue templates are really useful and the comment I linked to explains why, but here's an example of an issue template (again, you can configure them for different types of issues so team members select from a dropdown for a bug or a feature):