GNOME Bugzilla – Bug 544115
Exchange storage crashed at start up
Last modified: 2009-09-02 04:55:28 UTC
Evolution 2.23.5 (after Cameldb commit) Gdb traces of exchange-storage-process Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0xb60886e0 (LWP 3626)] 0x0805f296 in mse_get_folder_online_sync_updates (key=0x4, value=0x2, user_data=0x8183428) at mail-stub-exchange.c:582 582 if (mmsg->seq != seq) { (gdb) t a a bt
+ Trace 203465
Thread 1 (Thread 0xb60886e0 (LWP 3626))
Crash is happening 3 out of 5 times. Valgrind of exchange-storage process ==5552== ==5552== Use of uninitialised value of size 4 ==5552== at 0x805F296: mse_get_folder_online_sync_updates (mail-stub-exchange.c:582) ==5552== by 0x5C5EF55: g_hash_table_foreach (ghash.c:1076) ==5552== by 0x80625EA: get_folder_contents_online (mail-stub-exchange.c:805) ==5552== by 0x80627E0: get_folder_online (mail-stub-exchange.c:899) ==5552== by 0x8062DF1: get_folder (mail-stub-exchange.c:1007) ==5552== by 0x80651C1: connection_handler (mail-stub.c:127) ==5552== by 0x5CA3D7C: g_io_unix_dispatch (giounix.c:162) ==5552== by 0x5C6D177: g_main_context_dispatch (gmain.c:2068) ==5552== by 0x5C70812: g_main_context_iterate (gmain.c:2701) ==5552== by 0x5C70D31: g_main_loop_run (gmain.c:2924) ==5552== by 0x56BE0A2: bonobo_main (bonobo-main.c:311) ==5552== by 0x805B968: main (main.c:278) ==5552== ==5552== Invalid read of size 4 ==5552== at 0x805F296: mse_get_folder_online_sync_updates (mail-stub-exchange.c:582) ==5552== by 0x5C5EF55: g_hash_table_foreach (ghash.c:1076) ==5552== by 0x80625EA: get_folder_contents_online (mail-stub-exchange.c:805) ==5552== by 0x80627E0: get_folder_online (mail-stub-exchange.c:899) ==5552== by 0x8062DF1: get_folder (mail-stub-exchange.c:1007) ==5552== by 0x80651C1: connection_handler (mail-stub.c:127) ==5552== by 0x5CA3D7C: g_io_unix_dispatch (giounix.c:162) ==5552== by 0x5C6D177: g_main_context_dispatch (gmain.c:2068) ==5552== by 0x5C70812: g_main_context_iterate (gmain.c:2701) ==5552== by 0x5C70D31: g_main_loop_run (gmain.c:2924) ==5552== by 0x56BE0A2: bonobo_main (bonobo-main.c:311) ==5552== by 0x805B968: main (main.c:278) ==5552== Address 0x8 is not stack'd, malloc'd or (recently) free'd ==5552== ==5552== Process terminating with default action of signal 11 (SIGSEGV) ==5552== Access not within mapped region at address 0x8 ==5552== at 0x805F296: mse_get_folder_online_sync_updates (mail-stub-exchange.c:582) ==5552== by 0x5C5EF55: g_hash_table_foreach (ghash.c:1076) ==5552== by 0x80625EA: get_folder_contents_online (mail-stub-exchange.c:805) ==5552== by 0x80627E0: get_folder_online (mail-stub-exchange.c:899) ==5552== by 0x8062DF1: get_folder (mail-stub-exchange.c:1007) ==5552== by 0x80651C1: connection_handler (mail-stub.c:127) ==5552== by 0x5CA3D7C: g_io_unix_dispatch (giounix.c:162) ==5552== by 0x5C6D177: g_main_context_dispatch (gmain.c:2068) ==5552== by 0x5C70812: g_main_context_iterate (gmain.c:2701) ==5552== by 0x5C70D31: g_main_loop_run (gmain.c:2924) ==5552== by 0x56BE0A2: bonobo_main (bonobo-main.c:311) ==5552== by 0x805B968: main (main.c:278) ==5552==
cann't use evolution with exchange, it crashes like hell :(
Committed to Trunk as r1697 http://svn.gnome.org/viewvc/evolution-exchange?view=revision&revision=1697 This patch would check for the index value over-running the size of the messages array. Since all messages not being fetched at start up we face this crash. Now it should fail unnoticingly. Users won't notice any change but the root problem needs to be fixed.
Reducing the priority as per comment #3
This is fixed on trunk with the href being read properly from the db. Closing.
*** Bug 560590 has been marked as a duplicate of this bug. ***
*** Bug 557180 has been marked as a duplicate of this bug. ***
*** Bug 567610 has been marked as a duplicate of this bug. ***
*** Bug 567621 has been marked as a duplicate of this bug. ***
*** Bug 545460 has been marked as a duplicate of this bug. ***
*** Bug 593838 has been marked as a duplicate of this bug. ***