[strongSwan] Crash issue with 1k IPsec tunnels (using 5.0.4 strongswan and load-tester plugin)

Chinmaya Dwibedy ckdwibedy at yahoo.com
Wed Nov 6 14:57:41 CET 2013


Hi,

As a part of debugging session, I turned on --enable-lock-profiler option to know the cumulative time waited in each lock during daemon shutdown during shutdown  with 1000 IPsec Security Associations. But I find, the Charon IKEv2 daemon (at IKE initiator end ) is getting crashed. Here goes the stack trace. Can one please suggest what might be the issue behind this?

Program terminated with signal 6, Aborted.
#0  0x000000556255eda0 in raise () from /lib64/libc.so.6
(gdb) bt
#0  0x000000556255eda0 in raise () from /lib64/libc.so.6
#1  0x000000556256369c in abort () from /lib64/libc.so.6
#2  0x0000000120007d5c in segv_handler (signal=<value optimized out>)
    at charon.c:183
#3  <signal handler called>
#4  0x000000556262af3c in backtrace () from /lib64/libc.so.6
#5  0x0000005562189838 in backtrace_create (skip=2) at utils/backtrace.c:531
#6  0x00000055621817e0 in profiler_init (type=<value optimized out>)
    at threading/lock_profiler.h:76
#7  mutex_create (type=<value optimized out>) at threading/mutex.c:209
#8  0x00000055621e4864 in peer_cfg_create (name=<value optimized out>, 
    ike_cfg=<value optimized out>, cert_policy=<value optimized out>, 
    unique=<value optimized out>, keyingtries=<value optimized out>, 
    rekey_time=<value optimized out>, reauth_time=<value optimized out>, 
    jitter_time=<value optimized out>, over_time=0, mobike=false, 
    aggressive=false, dpd=<value optimized out>, 
    dpd_timeout=<value optimized out>, mediation=false, mediated_by=0x0, 
    peer_id=0x0) at config/peer_cfg.c:653
#9  0x00000055628e1f80 in generate_config (this=0x123ca1f20, num=2)
    at load_tester_config.c:560
#10 0x00000055621e1660 in get_peer_cfg_by_name (this=0x123c86450, 
    name=0x55628e5290 "load-test") at config/backend_manager.c:430
#11 0x00000055628e0fa0 in do_load_test (this=0x123c9ab50)
---Type <return> to continue, or q <return> to quit---
    at load_tester_plugin.c:139
#12 0x000000556217c854 in execute (this=0xfd00000001)
    at processing/jobs/callback_job.c:77
#13 0x000000556217d620 in process_jobs (worker=0x123cb4fe0)
    at processing/processor.c:219
#14 0x0000005562180d3c in thread_main (this=<value optimized out>)
    at threading/thread.c:309
#15 0x0000005562118698 in ?? () from /lib64/libpthread.so.0

Regards,
Chinmaya






More information about the Users mailing list