How to Fix MongoDB WiredTiger Cache Size Issues

In this tutorial, you'll learn about How to Fix MongoDB WiredTiger Cache Size Issues. We cover key concepts, practical examples, and best practices.

The Problem

MongoDB performance degrades, writes slow down, or you see WiredTiger: (cache checkpoint) and cache pressure in logs. The WiredTiger storage engine's internal cache is too small for your working set, causing excessive page evictions and disk I/O. When the cache can't hold the active data set, every read requires a disk access.

Quick Fix

1. Check current cache statistics

mongosh --eval "db.serverStatus().wiredTiger.cache"

Look for these key metrics:

"tracked dirty bytes in the cache": 2147483648,
"pages read into cache": 50000000,
"pages written from cache": 30000000,
"unmodified pages evicted": 15000000

High eviction counts indicate cache pressure.

2. Check the current WiredTiger cache size

mongosh --eval "db.serverStatus().wiredTiger.cache['maximum bytes configured']"

This returns the cache size in bytes. Divide by 1073741824 to get GB.

3. Increase WiredTiger cache size

Edit /etc/mongod.conf:

storage:
  wiredTiger:
    engineConfig:
      cacheSizeGB: 4

Restart MongoDB:

sudo systemctl restart mongod

The default is 50% of RAM minus 1 GB, or 256 MB for systems with less than 2 GB.

4. Calculate cache pressure ratio

mongosh --eval "
  const cache = db.serverStatus().wiredTiger.cache;
  const evicted = cache['pages evicted by application threads'];
  const read = cache['pages read into cache'];
  const pressure = (evicted / read * 100).toFixed(2);
  print('Cache pressure: ' + pressure + '%');
"

If pressure exceeds 20-30%, increase the cache.

5. Monitor evictions in real time

mongostat --noheaders 5 | grep -E "faults|evict"

High faults values mean the cache is too small for the working set.

6. Optimize indexes to reduce cache pressure

# Check index usage
mongosh --eval "
  db.adminCommand({listDatabases:1}).databases.forEach(dbInfo => {
    const db = db.getSiblingDB(dbInfo.name);
    db.getCollectionNames().forEach(coll => {
      db.getCollection(coll).getIndexes().forEach(idx => {
        print(dbInfo.name + '.' + coll + ': ' + idx.name);
      });
    });
  });
"

Remove unused indexes:

db.mycollection.dropIndex('unused_index_name')

Common Causes

Additional Troubleshooting

# Check the error message and stack trace for more context
echo "Review the full error output to identify the root cause"

If the above steps do not resolve the issue, examine the complete error message and stack trace. Often the key detail is in the middle of the traceback rather than the final line. Search for the error message in the project documentation or issue tracker for additional solutions.

Prevention

• Set cacheSizeGB to 50-80% of available RAM, leaving enough for the OS and other processes
• Monitor wiredTiger.cache statistics in your monitoring system
• Ensure your working set fits in the WiredTiger cache
• Add indexes to reduce the number of documents scanned per query