After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 632135 - (Evolution miner) memory leaking
(Evolution miner) memory leaking
Status: RESOLVED FIXED
Product: tracker
Classification: Core
Component: Miners
git master
Other Linux
: Normal critical
: ---
Assigned To: tracker-general
Jamie McCracken
Depends on:
Blocks:
 
 
Reported: 2010-10-14 11:37 UTC by Brian J. Murrell
Modified: 2010-12-09 11:34 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Brian J. Murrell 2010-10-14 11:37:15 UTC
As reported in Launchpad bug 654429 (https://bugs.launchpad.net/ubuntu/+source/evolution/+bug/654429) there seems to be a pretty severe memory leak in 2.30.3.  Attached to that LP bug is some valgrind diagnostics.  I wonder if somebody here could take a look and see if some of the leaks have been plugged, and if so, which commits would have taken care of them.

Much thanx!
Comment 1 Brian J. Murrell 2010-10-14 11:39:11 UTC
If desired, I could attach the valgrind logs from the LP bug here, but I'm not sure I see the point.  They can be downloaded from LP at:

https://bugs.launchpad.net/evolution/+bug/654429/+attachment/1670880/+files/valgrind.tar.gz

Cheers.
Comment 2 Milan Crha 2010-10-26 11:45:33 UTC
Moving this to tracker, because it leads the table:
==5887== 92,125,440 bytes in 71,973 blocks are possibly lost in loss record 62,863 of 62,864
==5887==    at 0x4C2815C: malloc (vg_replace_malloc.c:236)
==5887==    by 0x66E6C2E: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66C9A18: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66C9B17: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66CB74A: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66E8B8A: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66D3B33: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66ED1F3: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66ED377: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66EFBB3: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66EFC13: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x66EFDA6: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x6737852: ??? (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x672056F: sqlite3_step (in /usr/lib/libsqlite3.so.0.8.6)
==5887==    by 0x27F7A94D: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F7AF49: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F78609: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0xCD2173E: ??? (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0xCD1F7E3: ??? (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0x8D73970: start_thread (pthread_create.c:304)
==5887==    by 0xD07E92C: clone (clone.S:112)
==5887== 
==5887== 909,248,614 bytes in 240,306 blocks are possibly lost in loss record 62,864 of 62,864
==5887==    at 0x4C2815C: malloc (vg_replace_malloc.c:236)
==5887==    by 0xCCFF304: g_malloc (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0xCD16D5D: g_strdup (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0x27F79D7D: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F7ADDF: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F7AEEC: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F7AF49: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0x27F78609: ??? (in /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)
==5887==    by 0xCD2173E: ??? (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0xCD1F7E3: ??? (in /lib/libglib-2.0.so.0.2600.0)
==5887==    by 0x8D73970: start_thread (pthread_create.c:304)
==5887==    by 0xD07E92C: clone (clone.S:112)
Comment 3 Milan Crha 2010-10-26 11:46:27 UTC
Disabling the plugin meanwhile should help. Please set the proper tracker version.
Comment 4 Aleksander Morgado 2010-10-26 12:00:48 UTC
(In reply to comment #2)
> Moving this to tracker, because it leads the table:
> ==5887== 92,125,440 bytes in 71,973 blocks are possibly lost in loss record
> 62,863 of 62,864
> ==5887==    at 0x4C2815C: malloc (vg_replace_malloc.c:236)
> ==5887==    by 0x66E6C2E: ??? (in /usr/lib/libsqlite3.so.0.8.6)

This previous one may be normal, not real memory leaks (marked as possibly lost due to the interior pointers found by valgrind on sqlite memory)

> ==5887== 
> ==5887== 909,248,614 bytes in 240,306 blocks are possibly lost in loss record
> 62,864 of 62,864
> ==5887==    at 0x4C2815C: malloc (vg_replace_malloc.c:236)
> ==5887==    by 0xCCFF304: g_malloc (in /lib/libglib-2.0.so.0.2600.0)
> ==5887==    by 0xCD16D5D: g_strdup (in /lib/libglib-2.0.so.0.2600.0)
> ==5887==    by 0x27F79D7D: ??? (in
> /usr/lib/evolution/2.30/plugins/liborg-freedesktop-Tracker-evolution-plugin.so)

This one definitely seems a real memory leak. Not sure if we actually use interior pointers ourselves in Tracker, so don't really know why it's marked as possibly lost.
Comment 5 Martyn Russell 2010-12-08 10:29:04 UTC
Setting version to "trunk" from "unspecified", sorry for the spam.
Comment 6 Philip Van Hoof 2010-12-09 11:34:44 UTC
The evolution plugin has been reworked since this bug, closing