temporary file size exceeds temp_file_limit (N kB)

SQLSTATE 53400 condition configuration_limit_exceeded class 53 — Insufficient Resources severity ERROR

Reproduced & verified on PostgreSQL 14.23, 15.18, 16.14, 17.10 and 18.4 — identical message on every version.

Last reviewed 30 May 2025 · Reproduced live with the SQL on this page.

Symptoms

A query’s temporary files exceeded the configured temp_file_limit. PostgreSQL aborts the query with SQLSTATE 53400 (configuration_limit_exceeded).

A spill-to-disk operation grew past temp_file_limit.
Common with large sorts, hashes, or CTEs.
The limit is a safety valve against runaway temp usage.

What the server log shows

ERROR:  temporary file size exceeds temp_file_limit (5242880 kB)

Why PostgreSQL raises this — what the manual says

Section 19.4.2 Disk (temp_file_limit):

“A transaction attempting to exceed this limit will be canceled.”

When an operation cannot fit in work_mem, PostgreSQL spills to temporary files. If their total size exceeds temp_file_limit, the query is cancelled with 53400 to protect disk space.

Common causes

A huge sort/hash/aggregate spilling massive temp data.
temp_file_limit set conservatively.
A query processing far more data than expected (bad plan or missing filter).

Relevant GUC parameters

Parameter	Default	Effect
`temp_file_limit`	-1	Per-process cap on temp file space; -1 means unlimited.
`work_mem`	4MB	Higher values reduce spilling, lowering temp file usage.

How to fix it

Optimize the query to process less data (better filters/indexes/joins).
Raise temp_file_limit if the workload legitimately needs more temp space.
Increase work_mem for the session to reduce spilling.

Related & next steps

Reference: PostgreSQL 18 Section 20.4 “Resource Consumption”.