Bug #19340 valgrind: small leak in ndbd
Submitted: 25 Apr 2006 16:59 Modified: 26 Apr 2006 11:03
Reporter: Stewart Smith Email Updates:
Status: Closed Impact on me:
Category:MySQL Cluster: Cluster (NDB) storage engine Severity:S1 (Critical)
Version:5.0.21 OS:
Assigned to: Stewart Smith CPU Architecture:Any

[25 Apr 2006 16:59] Stewart Smith
==19667== 12 bytes in 1 blocks are definitely lost in loss record 2 of 15
==19667==    at 0x401B7F6: operator new(unsigned) (vg_replace_malloc.c:164)
==19667==    by 0x8267966: Configuration::calcSizeAlt(ConfigValues*) (Configuration.cpp:552)
==19667==    by 0x8268662: Configuration::setupConfiguration() (Configuration.cpp:457)
==19667==    by 0x80A09C5: main (main.cpp:405)
==19667== 15 bytes in 1 blocks are definitely lost in loss record 3 of 15
==19667==    at 0x401B45A: malloc (vg_replace_malloc.c:149)
==19667==    by 0x420EE6F: strdup (in /lib/tls/i686/cmov/libc-2.3.5.so)
==19667==    by 0x826719A: Configuration::init(int, char**) (Configuration.cpp:151)
==19667==    by 0x80A08E6: main (main.cpp:247)

I don't think anybody is ever going to hit this... it looks like something that's just not cleaned up on exit. If not, not in any frequently called code.

How to repeat:
run ndbd under valgrind

Suggested fix:
don't leak. (patch coming)
[25 Apr 2006 17:03] Bugs System
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

[25 Apr 2006 18:30] Stewart Smith
pushed to 5.0.21.

will be in 5.1 when somebody merges 5.0 to 5.1
[26 Apr 2006 11:03] Jon Stephens
Based on Stewart's comments, there's nothing here to document for end users. Closed.