Understanding Heroku Key-Value Store Log Statements and Common Errors
Last updated October 11, 2024
Table of Contents
Where possible, we changed noninclusive terms to align with our company value of Equality. We retained noninclusive terms to document a third-party system, but we encourage the developer community to embrace more inclusive language. We’ll update the term when it’s no longer required for technical accuracy.
Heroku Key-Value Store logs to the logplex which collates and publishes your application’s log-stream. You can isolate Heroku Key-Value Store events with the heroku logs command by filtering by process type -p flag and indicating your add-on name:
$ heroku logs -p redis-triangular-12345 -t
Besides seeing system-level Redis activity, these logs are also useful for understanding your application’s use of Redis and for diagnosing common errors. This article lists common log statements, their purpose, and any action that must be taken.
Heroku Key-Value Store metrics logs are emitted with heroku-redis as the process type:
$ heroku logs -p heroku-redis -t
REPLICA OF …
2022-08-18T17:29:39Z app[redis-triangular-12345]: * REPLICAOF would result into synchronization with the master we are already connected with. No operation performed.
This log statement indicates that Heroku Key-Value Store is ensuring your HA standby is following your primary Key-Value Store instances.
CONFIG REWRITE
2022-08-18T17:29:39Z app[redis-triangular-12345]: # CONFIG REWRITE executed with success.
This log statement indicates that Heroku Key-Value Store has proceeded to rewrite your Key-Value Store instance configuration. This statement happens whenever we must change your instance configuration or when you trigger a configuration change.
Database Memory Close to the Limit
2022-09-12T07:27:33+00:00 app[heroku-redis]: source=REDIS_URL sample#used_memory_over_limit=10435336.0bytes message=Database memory close to the limit. Please upgrade your database plan or change your maxmemory policy.
This log is seen when a Key-Value Store instance is using more than 90% of its plan memory limit. Since you can see this log before you actually exceed the limit, you can monitor this log to make an alert.