[Linux-HA] CVS HEAD ccm dumps core

Simon Rowe srowe at cambridgebroadband.com
Wed Oct 5 06:25:34 MDT 2005


I've had to ditch using 2.0.2 because certain fixes weren't present. Using 
HEAD last week worked fine but this week I can't get my two node cluster to 
run at all.

The first problem I can see is that ccm dumps core continually on one node. 
The stack trace is

#0  0x001ce402 in __kernel_vsyscall ()
(gdb) where
#0  0x001ce402 in __kernel_vsyscall ()
#1  0x003d3118 in raise () from /lib/libc.so.6
#2  0x003d4888 in abort () from /lib/libc.so.6
#3  0x003cc2f5 in __assert_fail () from /lib/libc.so.6
#4  0x0804ddc8 in report_mbrs (info=Variable "info" is not available.
) at ccm_statemachine.c:422
#5  0x08052266 in ccm_state_new_node_wait_for_mem_list 
(ccm_msg_type=CCM_TYPE_MEM_LIST, reply=0x97c1d08, hb=0x97b4258, 
info=0x97b7310) at ccm_statemachine.c:3766
#6  0x08049ddd in ccm_control_process (info=0x97b7310, hb=0x97b4258) at 
ccm.c:183
#7  0x0804eaba in ccm_take_control (data=0x0) at ccm_statemachine.c:3138
#8  0x00d68eb2 in G_CH_dispatch (source=0x97b6088, callback=0, user_data=0x0) 
at GSource.c:539
#9  0x005594ce in g_main_context_dispatch () from /usr/lib/libglib-2.0.so.0
#10 0x0055c4d6 in g_main_context_check () from /usr/lib/libglib-2.0.so.0
#11 0x0055c7c3 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#12 0x0804c4d9 in main (argc=1, argv=0xbf8a4d74) at ccmmain.c:288

The services don't come up on the other node but I thought this needed 
addressing first.



More information about the Linux-HA mailing list