Re: [mpls-linux-general] ldp crash..??
Status: Beta
Brought to you by:
jleu
|
From: James R. L. <jl...@mi...> - 2002-12-14 17:21:16
|
When you experience a crash that results in a core file. Start up gdb with that core file (gdb mplsd core) and the do a 'where'. That will print the stack at the time of the failure. Either send me the stack dump or go over to sf.net/projects/mpls-linux/ and submit a bug report with the stack trace and the info you have gathered below. Thanks On Sat, Dec 14, 2002 at 03:15:43PM +0900, Kwangwoo Lee wrote: > [MPLS network info] > zebra version : 0.93b > ldpportable version : 0.305 > mpls-linux version : 1.172 >=20 > +--------+ > eth1| |eth0 =20 > +-----------------+Router E+--------------+ > / | | \ eth2 > +--------+ / +--------+ \ +--------+ > | +/ \+ |= eth1 > ----+Router A| |Router D+= ---- =20 > | +\ /+ | > +--------+ \ +--------+ +--------+ / +--------+ > \ | | | | / eth0 > +-------+Router B+------+Router C+--------+ > | | | |eth1 > +--------+ +--------+ >=20 > hello... > this is our mpls network topology. > if mplsd run on each router, a certain mpls deamon is killed. >=20 > mpls deamons of Router B and Router C and router E run well. > but,=20 > mpls deamon of "Router D" run for a second=20 > then,=20 > it is killed soon. > (In this network, we disabled "router A".=20 > so don't care of it. and At route D, eth1 was "not enabled" to use mpls.) > why??? >=20 > we appended gdb excution result. >=20 > yesterday, error message > ------------------------ > mplsd: ldp_init.c:76: MPLS_ASSERT(0)=20 > --->maybe it is because of "ldp_init.c 's line 70 #if MPLS_USE_LSR" ??? > =20 > ------------------------ > today, two type error messages > ------------------------ > mplsd: ldp_attr.c:306: ldp_attr_delete: Assertion `a->in_tree =3D=3D MPLS= _BOOL_FALSE' failed. > Program received signal SIGABRT, Aborted. > 0x400f8781 in kill () from /lib/libc.so.6 =20 > ------------------------ > 224 thread_call(&thread); > (gdb) n > ldp_get_session_by_next_hop: NH_OTHER 0x80c8788 > ldp_get_next_hop_session_for_fec: 0x80c8788 > _ldp_global_add_outlabel > ldp_label_mapping_with_xc: enter > mpls_mpls_xconnect_add: 10000 <-> 10007 > ldp_label_mapping_with_xc: exit > ldp_mpls_in2out_del: 10000 <-> 10007 > session delete > 225 } > (gdb) n > 224 thread_call(&thread); > (gdb) n > ldp_label_mapping_with_xc: enter >=20 > Program received signal SIGPIPE, Broken pipe. > 0x401923e4 in write () from /lib/libc.so.6 > ---------------------------- >=20 > what happen?? > i don't know why.... >=20 > while gdb is running,=20 > we find these messages > adj delete > session delete > if adj means adjacency, adjacent router don't die.=20 > =FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF= =FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=D3=86+,=7F=F9=DE=B5=E9= =9A=8AX=AC=B2=9A'=B2=8A=DEu=BC=FFZ+a=1A=B7=9A=B4=FA0z=BF=C2=A2g=AC=1A=B7=9A= =B5=17=AC=A6=89=EC=89=B8=A5=8A=DC=8By=AA=E7=B6=8B=AC{*.=AE=9A0z=B6=AD9 =CD= =FE=C1=E2=82=13=DE=AD=FA+=99=A9=DCx*&=A6=EBb=9E=00=A1jy=DE=96=1Bm=A7=FF=FF= =86=97=FFu=EB=DC=85=A9=E7z_=E8=AE=0F=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF= =FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=FF=F9=A9=96= =CF=E5=8A{=B1=FE=07=A7z=B6=A5=99=A8=A5=8Ax%=8A=CBf=A6[=FF=96)=EE=C7=F8=1E= =9D=EA=DA=97=F9b=B2=DB=FF=B2=8B=ABq=E7=E8=AE=07=FF=9D=EBa=B6=DAl=FF=FF=E5= =8A=CBl=FE=CA.=AD=C7=9F=A2=B8=1E=FEw=AD=FEX=AC=B6=CF=E5=8A=CBb=9D=FA?=9A=99= l=FEX=A7=BB=1F=E0zw=AB --=20 James R. Leu |