Postgres Pro
Downloads
Home   >   mailing lists

Re: server crash on raspberry pi for large queries - Mailing list pgsql-bugs

From Matthew Clark
Subject Re: server crash on raspberry pi for large queries
Date
Msg-id 90650442.10108278.1724181522516@mail.yahoo.com
Whole thread Raw
In response to Re: server crash on raspberry pi for large queries  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: server crash on raspberry pi for large queries
List pgsql-bugs
Tree view
OS:  Linux pi 6.6.31+rpt-rpi-v6 #1 Raspbian 1:6.6.31-1+rpt1 (2024-05-29) armv6l GNU/Linux

I can try to make a test case, essentially a large table, then "select count(*)" from table.  The select works for smaller tables.

I'm using the build of postgresql packaged with the OS version above; from the apt repository and installed with "apt install postgresql".  repository - 

deb [ arch=armhf ] http://raspbian.raspberrypi.com/raspbian/ bookworm main contrib non-free rpi


postgres --version

postgres (PostgreSQL) 15.8 (Raspbian 15.8-0+deb12u1)






On Tuesday, August 20, 2024 at 09:16:16 AM EDT, Tom Lane <tgl@sss.pgh.pa.us> wrote:


Matthew Clark <mclark@drmatthewclark.com> writes:

> issue: when issuing a large query ,  e.g.   "Select count(*) from temperature" on a table with > 150M rows the system crashes.
>   example error:   LOG:  server process (PID 20037) was terminated by signal 4: Illegal instruction  no more diagnostic information is generated.
> Workaround : this stops happening when jit_above_cost is set to -1 in postgresql.conf   After changing this setting the problem stops.


What this sounds like is a memory leak in the JIT stuff.  We fixed
one such issue last year, but perhaps there's more.  Can you provide
a self-contained test case?  Also, please be more specific about
which Linux version you are using, and which LLVM version.

            regards, tom lane

pgsql-bugs by date:

Previous
From: Alvaro Herrera from 2ndQuadrant
Date:
Subject: Re: BUG #18559: Crash after detaching a partition concurrently from another session
Next
From: Tomas Vondra
Date:
Subject: Re: FDW INSERT batching can change behavior