DuckDB runs out of memory

DevOps and infra4 posts from 4 people4d active+00 posts in the last 7 days, 2 the 7 days before (fading)

Posts per day4 posts, Sep 21 to Oct 21

The posts behind it

4, newest first

GitHub issuesOct 14 SIGSEGV in ~GroupedAggregateHashTable after an Out of Memory error: FinalizePinState on a partition whose append state was never initialised### What happens? A grouped aggregate whose aggregates have destructors (`string_agg(... ORDER BY ...)`, `quantile_cont`, `arg_min` on a hashed key) runs out of memory under `SET memory_limit`, and instead of returning the `Out of Memory Error` the process dies with SIGSEGV in...duckdb/duckdb · h4nsmullerOpen ↗GitHub issuesOct 10 INSERT ... ON CONFLICT inside a transaction holds ~32 KB per column per inserted row until commit (OOM on ~1,000-row transactions)### What happens? Inside an explicit transaction, every row inserted by `INSERT ... ON CONFLICT (id) DO UPDATE` holds about 32 KB **per column** of transaction-local storage (`duckdb_memory()` tag `IN_MEMORY_TABLE`) until commit. A plain `INSERT` of the same row holds about 0.2...duckdb/duckdb · dtubbOpen ↗GitHub issuesOct 4 Out-of-memory error in a hash aggregate with ordered list states aborts the process (exception thrown in ~GroupedAggregateHashTable)### What happens? A query that runs past `memory_limit` ends the process with `std::terminate` instead of returning an Out of Memory error. The backtrace shows an exception thrown inside a destructor while the failed query is cleaned up: In v1.5.1,...duckdb/duckdb · sounkou-bioinfoOpen ↗GitHub issuesSep 21 Consider a dedicated allocator for Windows embedded DuckDB workloads### What happens? ## Background I recently investigated a Windows performance problem in DuckDB when it is embedded in an R process hosted by Positron/RStudio. The investigation started from a substantial loss of DuckDB parallel performance after repeated workloads, and the...duckdb/duckdb · koenniemOpen ↗

Companies and products named

About this problem

Evidence

4 posts from 4 people in 1 place, about 1 a week over 24 days. Mostly on duckdb/duckdb. Tools named alongside: Positron, Segment.

Frustration Frustration 0 of 3· Seen on GitHub issues

How it was grouped

Posts that state a pain and match the "oom" rule about DuckDB. First post Sep 21, 2026, latest Oct 14, 2026. Corroborated: 3 or more posts from 2 or more people or places. Method

The same problem elsewhere

Elsewhere9 postsGemini runs out of memory Elsewhere9 postsJobs and services run out of memory Elsewhere5 postsGitHub runs out of memory Elsewhere4 postsNVIDIA runs out of memory Elsewhere4 postsApache Iceberg runs out of memory Elsewhere3 postsPolars runs out of memory Elsewhere2 postsvLLM runs out of memory Elsewhere2 postsOllama runs out of memory Elsewhere2 postsllama.cpp runs out of memory Elsewhere2 postsApache Airflow runs out of memory Elsewhere2 postsMistral runs out of memory

Other DuckDB problems

Same product15 postsDuckDB upgrades break working setups Same product14 postsDuckDB users keep filing bugs Same product9 postsTrouble with DuckDB column Same product8 postsTrouble with DuckDB overflow Same product6 postsDuckDB error messages are cryptic Same product6 postsTrouble with DuckDB internal Same product5 postsTrouble with DuckDB null Same product5 postsTrouble with DuckDB values Same product5 postsTrouble with DuckDB cast Same product5 postsTrouble with DuckDB result Same product3 postsDuckDB costs outgrow the value it gives Same product2 postsDuckDB permissions are confusing and error-prone

History

MomentumOct 21, 2026steady to fading

CorroborationOct 15, 2026unverified to corroborated

MomentumOct 15, 2026fading to steady

MomentumOct 11, 2026steady to fading

AddedOct 7, 2026DuckDB runs out of memory (2 posts)