Frequent 500 Internal Server Errors - Production System Down Daily
PROJECT ID: 6951f2xxxxx1ee9e750a REGION: Singapore (sgp.cloud.appwrite.io) PLAN: Pro
Our production application experiences 500 Internal Server Errors multiple times every day, making the system unreliable for our users (200+ students and staff).
Error Pattern:
- Frequency: 1-2+ times per day
- Timing: Random, not at specific times
- Duration: Varies (seconds to minutes)
- Recovery: Automatic, after retry
- Impact: Complete application unavailability
ERROR DETAILS:
Console Error: Uncaught (in promise) TypeError: Failed to fetch at analytics.ts:189:35 at di (analytics.ts:74:35) at ok (query-svelteKit:75:35) at marker.svelteKit:175:5 at shell.svelteKit:205:20
text
Network Error: POST https://0106364j.ingest.us.sentry.io/...: net::ERR_NAME_NOT_RESOLVED POST https://0106364j.ingest.us.sentry.io/...: net::ERR_NAME_NOT_RESOLVED GET https://fonts.ononsrik.org/...: net::ERR_NAME_NOT_RESOLVED
text
HTTP Status: 500 Endpoint: https://sgp.cloud.appwrite.io/v1
WHAT WE'VE DONE:
✅ Verified API keys are valid and rotated after exposure ✅ Confirmed database connections are working (sporadic failures) ✅ Checked collection indexes and query optimization ✅ Verified Storage bucket permissions ✅ Confirmed no rate limiting (API requests are within limits) ✅ Reviewed error logs - no clear pattern ✅ Tested from multiple locations (same issue) ✅ Used recommended SDK versions (node-appwrite v13.0+) ✅ Implemented retry logic (temporary workaround)
Current Status: CRITICAL - Production Down😩
Hey @jalu , thanks for sharing this. I’ll escalate for our team to investigate, in the meanwhile if you have any network logs or steps for reproducing the issues please share here or with me on DMs.
anything from the specific requests that failed, timestamps and error messages would help.
