List:General Discussion« Previous MessageNext Message »
From:Jeremy Zawodny Date:March 20 2002 12:02am
Subject:4.0.x memory bug on FreeBSD...?
View as plain text  
Running the latest build from bitkeeper (as of Saturday or so) on my
4.0.x slave (FreeBSD 4.3), after several hours, I start getting these
messages when apps try to connect to the server:

  ERROR 1041: Out of memory;  Check if mysqld or some other process
  uses all available memory. If not you may have to use 'ulimit' to
  allow mysqld to use more memory or you can add more swap space

However, I've verified that the ulimit is unlimited.  There is over
1.4GB of available memory on the machine, and there is a lot of disk
space.

Here's the relevant entry from the error log, repeated over and over:

  020319 15:57:33  /home/mysql-4-bk/libexec/mysqld: Out of memory at line 100,
'net_serv.cc'
  020319 15:57:33  /home/mysql-4-bk/libexec/mysqld: needed 16391 byte (17k), memory in
use: 535192717 bytes (522650k)

It now has the CPU at 100% but isn't doing any real work.  The binary
log is no longer updating.  I can kill mysqld and restart and it'll
run fine for a day or so.  I know because this happened yesterday too.

What can I do to debug this further?  Give it a "kill -6" and
backtrace on the core?

Other ideas?

FYI, this was build with gcc 2.95.3 using the standard
compile-pentium-debug script.

Thanks,

Jeremy
-- 
Jeremy D. Zawodny, <jzawodn@stripped>
Technical Yahoo - Yahoo Finance
Desk: (408) 349-7878   Fax: (408) 349-5454   Cell: (408) 685-5936

MySQL 3.23.47-max: up 40 days, processed 1,192,195,107 queries (341/sec. avg)
Thread
4.0.x memory bug on FreeBSD...?Jeremy Zawodny20 Mar
  • Backtrace for (Re: 4.0.x memory bug on FreeBSD...?)Jeremy Zawodny20 Mar
    • Backtrace for (Re: 4.0.x memory bug on FreeBSD...?)Michael Widenius22 Mar
      • Re: Backtrace for (Re: 4.0.x memory bug on FreeBSD...?)Jeremy Zawodny22 Mar
  • Re: 4.0.x memory bug on FreeBSD...?Dan Nelson20 Mar
    • Re: 4.0.x memory bug on FreeBSD...?Jeremy Zawodny20 Mar