Skip to content

RDS Aurora Postgres Instance Memory Usage Fluctuating

0

Running a RDS Aurora Postgres cluster with two r6g.2xlarge instances that started fluctuating in memory usage around 6:50pm EST yesterday. Typically we are running with ~16gb headroom but it dips to ~5gb at ~10m intervals. At some points it dipped below the threshold that it needed to restart the instance. We rolled back any code changes that had happened during the day

How can I see which process/query is using this memory?

Noticed this corresponded very closely to the announcement of performance insights.

1 Answer
0

Have you thought about any of these to help narrow your options and identify the potential root cause?

Leverage Performance Insights: If Performance Insights was recently activated, it could be consuming resources itself or highlighting queries with high memory usage. You can use this to identify long-running or resource-intensive queries.

Monitor Active Sessions with pg_stat_activity: By querying pg_stat_activity, you can see which processes or queries are active, their durations, and whether they’re consuming substantial resources. This can help pinpoint troublesome queries or user sessions.

Analyze Memory and Temp Usage with pg_stat_database: Temporary file writes or excessive reads can signal inefficient queries. Running queries on pg_stat_database can reveal patterns of resource usage across your databases.

Use Enhanced Monitoring: AWS Enhanced Monitoring provides a more detailed view of OS-level metrics, such as freeable memory and page cache usage. Spikes or correlations in these metrics can often reveal inefficient workloads.

answered a year ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.