io_error could not read block N in file “…”: read only 0 of N bytes — 58030

Symptoms

PostgreSQL tried to read a data block but the file was shorter than expected (it read 0 of the requested bytes). This usually signals truncation or storage failure and raises SQLSTATE 58030 (io_error).

• The block PostgreSQL expects is beyond the end of the file.
• Often a sign of filesystem damage, truncation, or hardware faults.
• Reads of the affected relation fail.

What the server log shows

ERROR:  could not read block 8423 in file "base/16384/24576": read only 0 of 8192 bytes

Why PostgreSQL raises this — what the manual says

As Section 30.2 Data Checksums explains:

A read returned fewer bytes than the expected block size (zero of the requested bytes), indicating the relation file is truncated or the underlying storage failed to return the page; this points to filesystem or hardware-level corruption.

PostgreSQL expects each relation file to contain whole 8 kB blocks. When a requested block lies past the actual end of the file (read returns 0 bytes), it cannot proceed and reports 58030, indicating storage-level damage or truncation.

Common causes

• Filesystem corruption or a truncated relation file.
• Failing disk, bad sectors, or storage-layer faults.
• An interrupted restore or a partially copied data directory.

How to fix it

1. Stop and take a full backup of the data directory before anything else.
2. Check OS/disk logs (SMART, dmesg) and the underlying storage health.
3. Restore the affected relation/table from a known-good backup; consider PITR.

Related & next steps

• SQLSTATE 58030 — I/O Error (full reference)
• Message: invalid page in block of relation
• Glossary: Data block

Reference: PostgreSQL 18 Section 30.2 “Reliability”.