Issue

• In vmcore crash utilities 'sys' command shows reboot time as 'DATE: Sat Apr 6 18:55:22 2013' but the actual reboot as per logs took place at 'Apr 6 16:15:23'.
• system time is in PDT. How does kdump take the time ? it doesn't look Timezone ?:

    crash> sys
          KERNEL: /cores/20130408020143/work/vmlinux
        DUMPFILE: /cores/20130408020143/work/RHcase00817055.<hostname>.jasriram  [PARTIAL DUMP]
            CPUS: 4
            DATE: Sat Apr  6 18:55:22 2013
          UPTIME: 21 days, 19:58:15
    LOAD AVERAGE: 8.29, 3.29, 2.08
           TASKS: 633
        NODENAME: <hostname>
         RELEASE: 2.6.18-238.9.1.el5
         VERSION: #1 SMP Fri Mar 18 12:42:39 EDT 2011
         MACHINE: x86_64  (3325 Mhz)
          MEMORY: 15.8 GB
           PANIC: "SysRq : Trigger a crashdump"


    Apr  6 12:43:03 <hostname> kernel: nfs: server <SERVERNAME_1> not responding, still trying
    Apr  6 12:43:03 <hostname> kernel: nfs: server <SERVERNAME_1> not responding, still trying
    Apr  6 12:43:33 <hostname> kernel: nfs: server <SERVERNAME_2> OK
    Apr  6 12:43:03 <hostname> kernel: nfs: server <SERVERNAME_1> not responding, still trying
    Apr  6 12:43:33 <hostname> kernel: nfs: server <SERVERNAME_2> OK
    Apr  6 12:43:33 <hostname> kernel: nfs: server <SERVERNAME_2> OK
    Apr  6 12:43:34 <hostname> kernel: nfs: server <SERVERNAME_1> OK
    Apr  6 12:43:33 <hostname> kernel: nfs: server <SERVERNAME_2> OK
    Apr  6 12:43:34 <hostname> kernel: nfs: server <SERVERNAME_1> OK
    Apr  6 16:15:23 <hostname> syslogd 1.4.1: restart.
    Apr  6 16:15:23 <hostname> kernel: klogd 1.4.1, log source = /proc/kmsg started.
    Apr  6 16:15:23 <hostname> kernel: klogd 1.4.1, log source = /proc/kmsg started.

• System reboot time ans 'sys' command shown output time should be matching.
• 'ntp' is one possibility but 'ntp' never correct such high drifts ~ 2 hours.
• As per the 'ntp' logs there are no time corrections.