pgvector HNSW Build Out of Memory or Shared Memory Error

How to fix it

// try these in order, the most common fix is first

1. FIX 1Docker shared memory

   1. 01Set shm_size in docker-compose to at least maintenance_work_mem
   Did this fix it? Yes No

2. FIX 2Reduce settings

   Copy steps
   1. 01Lower max_parallel_maintenance_workers
   2. 02Lower maintenance_work_mem
   Did this fix it? Yes No

3. FIX 3Check for OOM kills

   Copy steps
   1. 01Run dmesg and look for OOM killer lines
   2. 02Move to a larger instance
   Did this fix it? Yes No

Symptoms

Build aborts with could not resize shared memory segment or the process is OOM killed.

Why it happens

• C-01Docker /dev/shm default of 64 MB
• C-02maintenance_work_mem larger than available RAM
• C-03Too many parallel workers

// this guide also covers

• pgvector could not resize shared memory segment
• pgvector out of memory
• hnsw index build fails

The same problem in other apps

All export, render & quality loss guides →

• Chroma pip install "Failed to build hnswlib" Error (Chroma)

Other pgvector errors

• pgvector Error: extension "vector" is not available
• pgvector HNSW Index Build Slow
• pgvector "operator does not exist: vector <=>" Error
• pgvector Dimension Limit: Cannot Index More Than 2000 Dimensions
• pgvector "different vector dimensions" Error
• pgvector Query Returns Fewer Results Than LIMIT With Filter
• pgvector Index Not Being Used
• How to Enable pgvector on Supabase