I’m experiencing a confusing set of problems with my self-hosted Gitea instance that are directly affecting the development workflow for my Texas Roadhouse menu website project. Everything was running fine for months, but recently my repository started showing intermittent sync failures when pushing commits from local machines. Some pushes go through instantly, while others fail with timeout errors or incomplete object transfers. I’m using Gitea on a small VPS with plenty of free disk space and stable CPU load, so I’m having trouble understanding why the repo behaves so inconsistently now.
The second issue concerns Gitea’s built-in CI/CD integration (via Gitea Actions). My website uses a build pipeline to export optimized static assets—including images and structured menu data—but the runner has started aborting jobs at random stages. Sometimes it stops during dependency installation, and other times it crashes right after the asset optimization step. There are no meaningful error logs except vague “process terminated” messages. This makes it difficult to maintain a reliable build flow for the Texas Roadhouse menu site whenever I push updates or add new menu items.
I also noticed that repository indexing has become extremely slow. Browsing through commits or viewing diffs now takes several seconds, especially for the larger branches that contain the 3D food assets and high-resolution images I use on the website. I suspect it might be related to how Gitea handles large binary files, but I haven’t been able to confirm whether Git LFS is misconfigured or if my server storage layer is causing bottlenecks. The slowdown only started recently, so I’m not sure if this is related to database fragmentation, an LFS issue, or something else.
Another unexpected behavior involves webhook notifications. I’ve set up a webhook that triggers my deployment script whenever a push is made to the main branch. Over the past week, about half the webhook deliveries have been failing, returning 500-series errors even though my deployment server is online. This results in incomplete deployments or situations where the website updates only partially. For a content-heavy site like my Texas Roadhouse menu platform, consistency is essential, so these broken webhooks are creating unnecessary downtime.
I’ve already checked the Gitea logs, but nothing stands out beyond generic warnings about slow responses or failed object writes. Database health looks normal, migration status is clean, and system resources stay well below limits. This makes me think the issue might be related to network problems between the Gitea instance and remote clients, or possibly a corruption somewhere in the repo that Git hasn’t flagged. I’m unsure whether I should re-clone and rebuild the repository from scratch or if there’s a specific Gitea repair tool I should run first.
Overall, I’m trying to determine whether the root cause is related to Gitea configuration, Git LFS handling, CI runner instability, or something deeper in the hosting environment. If anyone here has seen similar behavior especially around repo sync failures, unstable CI builds, or webhook delivery problems I’d really appreciate some guidance. This project is important to me, and maintaining stable version control is crucial for keeping my Texas Roadhouse menu website updated and functioning properly. Sorry for long post