Support Center > Search Results > SecureKnowledge Details
Cluster member on Gaia OS crashed after changing its state to Active / VRRP Master
Symptoms
  • Cluster member on Gaia OS crashed after changing its state to Active / VRRP Master.

  • Issue occurred on the following machine:

    • Machine with CPU that supports AES-NI (e.g., 13500 / 13800 appliances)
    • Gaia OS in 64-bit mode
    • Enabled IPSec VPN blade
    • VPN tunnel with AES encryption
  • VMCore dump file generated during the crash contained the following stack:

    ID: XXX TASK: ... CPU: N COMMAND: "fw_worker_0"
    #0 [...] machine_kexec at ... 
    #1 [...] crash_kexec at ... 
    #2 [...] kdb_main_loop at ... 
    #3 [...] kdba_main_loop at ... 
    #4 [...] kdb at ... 
    #5 [...] kdba_entry at ... 
    #6 [...] notifier_call_chain at ... 
    #7 [...] __die at ... 
    #8 [...] do_page_fault at ... 
    #9 [...] error_exit at ... 
    #10 [...] _aesni_decrypt3() at ... 
    #11 [...] aesni_cbc_encrypt() at ... 
    #12 [...] rijndael_cbc_decrypt() at ... 
    #13 [...] fwskip_rijndael128() at ... 
    #14 [...] esp_ipsec_crypt() at ... 
    #15 [...] esp_decrypt_ex() at ... 
    #16 [...] fw_ipsec_decrypt_request_ex() at ... 
    #17 [...] vpn_ipsec_decrypt() at ... 
    #18 [...] vpn_decrypt_chain() at ... 
    #19 [...] fwchain_do_ex() at ... 
    #20 [...] fw_filter_ip() at ... 
    #21 [...] fw_filter_locked() at ... 
    #22 [...] fwmultik_process_packet_kernel() at ... 
    #23 [...] fwmultik_process_entry() at ... 
    #24 [...] fwmultik_queue_async_dequeue_cb() at ...
    
Cause

Crashed was caused by a corrupted encryption key (that was probably synchronized from the other cluster member).


Solution
Note: To view this solution you need to Sign In .