The 12-factor app methodology, specifically in Factor XI (Logs), advocates for treating logs as event streams. The application itself should not be concerned with routing or storing its output. Instead, it should write its event stream directly to standard output (stdout). This decouples the application from log management. The execution environment is then responsible for capturing this stream and routing it to appropriate destinations, such as log indexing and analysis systems (e.g., Splunk, ELK stack), for processing and archival. This approach allows for maximum flexibility in handling logs without modifying application code.

B. This describes a specific storage implementation; the 12-factor app itself should be agnostic to how its log stream is stored.
D. This is explicitly what the 12-factor methodology advises against, as it couples the application to the filesystem and complicates log management.
E. This is a log management detail that should be handled by the execution environment, not the application.

1. The 12-Factor App

XI. Logs: "A twelve-factor app never concerns itself with routing or storage of its output stream. It should not attempt to write to or manage logfiles. Instead

each running process writes its event stream

unbuffered

to stdout... The stream can be captured by the execution environment... and routed to one or more final destinations for viewing and long-term archival."

Source: https://12factor.net/logs

2. Heroku Dev Center

Logging Documentation: "Your app should write all logging messages to stdout or stderr. Heroku’s Logplex router is responsible for collating and routing log messages from your app and Heroku components... Log drains are attached to an app to receive all of its log messages. You can send your logs to a syslog server or an HTTPS endpoint [for analysis]."

Source: https://devcenter.heroku.com/articles/logging

"How Logging Works" section.

3. Red Hat OpenShift Documentation

About Cluster Logging: "The recommended practice for applications is to log to stdout and stderr

rather than to a file. The logging stack collects the logs

processes them

and forwards them to a log aggregator

such as Elasticsearch."

Source: https://docs.openshift.com/container-platform/4.14/logging/cluster-logging.html

"About cluster logging" section.