<HTML>
<HEAD>
<TITLE>Server crash in src/io/flow/flowproto-bmi-trove/flowproto-multiqueue.c</TITLE>
</HEAD>
<BODY>
<FONT FACE="Courier, Courier New"><SPAN STYLE='font-size:11pt'>I occasionally get a server crash in what appears to be src/io/flow/flowproto-bmi-trove/flowproto-multiqueue.c. The backtrace is useless. I’m running off the head branch code that I compiled on 7/3.<BR>
<BR>
[E 07/14 18:06] PVFS2 server: signal 11, faulty address is (nil), from (nil)<BR>
[E 07/14 18:06] [bt] [(nil)]<BR>
[D 07/15 08:19] PVFS2 Server version 2.8.1pre1-2009-07-03-123548 starting.<BR>
<BR>
I added a few extra gossip_err statements in the handle_io_error routine and narrowed it down to the following few lines:<BR>
<BR>
else if (src == TROVE_ENDPOINT && dest == BMI_ENDPOINT)<BR>
{<BR>
ret = cancel_pending_trove(&flow_data->src_list, flow_data->parent->src.u.trove.coll_id);<BR>
flow_data->cleanup_pending_count += ret;<BR>
<BR>
Any ideas?<BR>
<BR>
Thanks,<BR>
Randy</SPAN></FONT>
</BODY>
</HTML>