Retrace server is too slow

[mcatanzaro]

Every backtrace I've seen ABRT upload to bug reports in the past couple weeks has ended:

Timeout exceeded: 240 seconds, killing /usr/libexec/gdb.
Looks like gdb hung while generating backtrace.
This may be a bug in gdb. Consider submitting a bug report to gdb developers.
Please attach coredump from this crash to the bug report if you do.

Example. I doubt there is any gdb bug here, because it seems to be happening for every bug report, and I haven't noticed any recent problems using gdb locally. I think the server is just too slow.