Open odivlad opened 10 months ago
Hi, not reproduced using the inputs of this ticket on upstream. is it possible to get the back trace from GDB ?
Hi, thanks for the response!
(gdb) bt
#0 zvrf_id (zvrf=0x1bdf8a0) at zebra/zebra_vrf.h:207
#1 rib_process (rn=0x1bdd040) at zebra/zebra_rib.c:1127
#2 process_subq_route (qindex=2 '\002', lnode=0x1bdd160) at zebra/zebra_rib.c:2386
#3 process_subq (qindex=<optimized out>, subq=0x1a00760) at zebra/zebra_rib.c:2428
#4 meta_queue_process (dummy=<optimized out>, data=0x1a00e20) at zebra/zebra_rib.c:2461
#5 0x00007f9a6cad2630 in work_queue_run (thread=0x7ffcb4558aa0) at lib/workqueue.c:291
#6 0x00007f9a6cac853b in thread_call (thread=thread@entry=0x7ffcb4558aa0) at lib/thread.c:1825
#7 0x00007f9a6ca888f0 in frr_run (master=0x184d9e0) at lib/libfrr.c:1155
#8 0x000000000041e868 in main (argc=9, argv=0x7ffcb4558e88) at zebra/main.c:470
Please, note, this backtrace is collected within frr frr-8.0.1-02.el7.x86_64.rpm, 'cause my environment uses it.
@pguibert6WIND, hi! Do you need any additional information?
This issue is stale because it has been open 180 days with no activity. Comment or remove the autoclose
label in order to avoid having this issue closed.
This issue will be automatically closed in the specified period unless there is further activity.
The problem persists.
This issue will no longer be automatically closed.
I run FRR with next
/etc/frr/daemons
config:Sometimes when adding or deleting netns-based VRF zebra daemon crashes with SIGSEGV.
When create with these commands:
sometimes daemon crashes with next trace.
When delete with these commands:
sometimes I get this trace:
Strace output right before and during crash:
The issue was first observed with FRR 8.0 and confirmed with 9.1. OS: CentOS 7.8 Kernel: 3.10.0-1127.19.1.el7.x86_64
I'm ready to provide additional info if needed for debug.