Description of problem: In a neat turnaround from prior behavior, sometimes the browser crashes while npviewer.bin (flash) keeps running. Version-Release number of selected component (if applicable): nspluginwrapper-0.9.91.5-2.fc8 firefox-2.0.0.6-8.fc8 How reproducible: Occasionally. Sometimes everything crashes.
Here's a trace. [Switching to Thread 46912597138704 (LWP 5973)] rpc_method_get_args (connection=0x27c7210) at ../src/rpc.c:756 756 ../src/rpc.c: No such file or directory. in ../src/rpc.c Current language: auto; currently c #0 rpc_method_get_args (connection=0x27c7210) at ../src/rpc.c:756 #1 0x00002aaac95f6178 in npclass_handle_Invalidate (connection=0x27c7210) at ../src/npruntime.c:68 #2 0x00002aaac95f30c3 in _rpc_dispatch (connection=0x27c7210, message=0x7ffff209b3c0) at ../src/rpc.c:1366 #3 0x00002aaac95f4db4 in rpc_method_wait_for_reply (connection=0x27c7210) at ../src/rpc.c:1640 #4 0x00002aaac95f6130 in npclass_invoke_Invalidate (npobj=<value optimized out>) at ../src/npruntime.c:96 #5 0x00002aaac95f61d3 in npclass_handle_Invalidate (connection=<value optimized out>) at ../src/npruntime.c:77 #6 0x00002aaac95f30c3 in _rpc_dispatch (connection=0x27c7210, message=0x7ffff209d6d0) at ../src/rpc.c:1366 #7 0x00002aaac95f4db4 in rpc_method_wait_for_reply (connection=0x27c7210) at ../src/rpc.c:1640 #8 0x00002aaac95f6130 in npclass_invoke_Invalidate (npobj=<value optimized out>) at ../src/npruntime.c:96 #9 0x00002aaac95f61d3 in npclass_handle_Invalidate (connection=<value optimized out>) at ../src/npruntime.c:77 #10 0x00002aaac95f30c3 in _rpc_dispatch (connection=0x27c7210, message=0x7ffff209f9e0) at ../src/rpc.c:1366 #11 0x00002aaac95f4db4 in rpc_method_wait_for_reply (connection=0x27c7210) at ../src/rpc.c:1640 #12 0x00002aaac95f6130 in npclass_invoke_Invalidate (npobj=<value optimized out>) at ../src/npruntime.c:96 #13 0x00002aaac95f61d3 in npclass_handle_Invalidate (connection=<value optimized out>) at ../src/npruntime.c:77 #14 0x00002aaac95f30c3 in _rpc_dispatch (connection=0x27c7210, message=0x7ffff20a1cf0) at ../src/rpc.c:1366 #15 0x00002aaac95f4db4 in rpc_method_wait_for_reply (connection=0x27c7210) at ../src/rpc.c:1640 #16 0x00002aaac95f6130 in npclass_invoke_Invalidate (npobj=<value optimized out>) at ../src/npruntime.c:96 #17 0x00002aaac95f61d3 in npclass_handle_Invalidate (connection=<value optimized out>) at ../src/npruntime.c:77 #18 0x00002aaac95f30c3 in _rpc_dispatch (connection=0x27c7210, message=0x7ffff20a4000) at ../src/rpc.c:1366 #19 0x00002aaac95f4db4 in rpc_method_wait_for_reply (connection=0x27c7210) at ../src/rpc.c:1640 #20 0x00002aaac95f6130 in npclass_invoke_Invalidate (npobj=<value optimized out>) at ../src/npruntime.c:96 ... It appears that nspluginwrapper is recursing somewhere in it's rpc code, and eventually blows its stack. (the same functions continue for about 2000 frames before I got tired of listing the backtrace.)
Should be fixed in nspluginwrapper-0.9.91.5-4.fc8. It uses gtk_main_quit(); so it may not work if there's some hard deadlock...
I've still seen crashes here.
(In reply to comment #3) > I've still seen crashes here. Is the plugin running after that or not?
No, it seems better at least in that regard. Still crashes with some frequency, though.
*** This bug has been marked as a duplicate of 298481 ***