Heroku Log Drain Support for Sentry Logging

[AbhiPrasad]

You can now send structured logs to Sentry! See the docs here.

This issue tracks support log drains from Heroku. If you're interested in Heroku Log Drain support, please react to this issue and leave a comment about your use case. That will help us prioritize this accordingly.

Another ask for this (related to errors): #28270

[joemartinezsentry]

Request via Zendesk #91727

[yelworc]

This would probably be super useful for us! ❤️

We're already using Sentry for application-level error reporting (have been for years), but occasionally something goes wrong on the OS level in a Heroku dyno, and the issue goes entirely unreported until somebody notices some resulting breakage.

A typical case would be an out-of-memory issue, where the dyno OS just SIGKILLs our process without warning, so there's no way to report anything from "inside" the app. This is what Heroku is logging in such a situation:

2025-09-10T15:04:28.877 heroku[scheduler.9593]: State changed from starting to up
2025-09-10T15:04:48.318 heroku[scheduler.9593]: Process running mem=791M(154.6%)
2025-09-10T15:04:48.320 heroku[scheduler.9593]: Error R14 (Memory quota exceeded)
2025-09-10T15:05:10.229 heroku[scheduler.9593]: Process running mem=1865M(364.3%)
2025-09-10T15:05:10.231 heroku[scheduler.9593]: Error R15 (Memory quota vastly exceeded)
2025-09-10T15:05:10.232 heroku[scheduler.9593]: Stopping process with SIGKILL
2025-09-10T15:05:13.430 heroku[scheduler.9593]: Process exited with status 137
2025-09-10T15:05:13.451 heroku[scheduler.9593]: State changed from up to complete

A Sentry log drain that can report specific messages (such as the above Error R15) to our existing Sentry projects would be marvelous!

Like the description in #28270 says, we'd need some filtering abilities around this (e.g. no reporting for benign messages like "State changed…" or many others, ignoring specific errors like R14 which may be logged several times before the fatal R15, etc.).