07-22-2014 06:29 AM
We have Brocade 48000 in fabric. Due to kernel panic, hafailover happened and after that active CP not pinging and not able to login to active CP using neither CP IP nor chassis IP.Below are the errdump logs. Please let me know if anyone faced this issue or any solution on the proble.
2014/07/22-02:24:12, [KSWD-1002], 722463, SLOT 6 | FFDC | CHASSIS, WARNING, IBM_2109_M48, Detected termination of process icpd:1297
2014/07/22-02:24:12, [RAS-1001], 722464, SLOT 6 | CHASSIS, INFO, IBM_2109_M48, First failure data capture (FFDC) event occurred.
2014/07/22-02:24:12, [FSSM-1003], 722465, SLOT 5 | CHASSIS, WARNING, IBM_2109_M48, HA State out of sync.
2014/07/22-02:24:13, [HAM-1014], 722466, SLOT 6 | CHASSIS, CRITICAL, IBM_2109_M48, Non restartable component (icpd (pid=1297)) died.
2014/07/22-02:24:16, [SNMP-1008], 722467, SLOT 6 | FID 128, INFO, DC1SSWPE1, The last device change happened at : Tue Jul 22 02:24:12 2014
2014/07/22-02:28:42, [HAM-1004], 722468, SLOT 6 | CHASSIS, INFO, IBM_2109_M48, Processor rebooted - Software Fault:Kernel Panic
2014/07/22-02:28:53, [TRCE-1001], 722469, SLOT 6 | CHASSIS, WARNING, IBM_2109_M48, Trace dump available (Slot 6)! (reason: PANIC)
2014/07/22-02:28:53, [TRCE-1004], 722470, SLOT 6 | CHASSIS, WARNING, IBM_2109_M48, Trace dump (Slot 6) was not transferred because trace auto-FTP disabled.
2014/07/22-02:27:37, [FSSM-1002], 722471, SLOT 6 | CHASSIS, INFO, IBM_2109_M48, HA State is in sync.
07-22-2014 11:01 PM
Usually that issue is caused by a ethernet infrastructure issue:
- does the new Active CP (former Standby CP) has a link on the ethernet switch side?
- Are you able to ping the standby CP (former active CP)? If you can, try to ping the Active CP from the Standby one. You can ping the backend management IP to see if it responds.