OCI: The Time Stamp Recorded in the Vmcore and Reboot Log Messages Does not Match

Applies To:

Oracle Cloud Infrastructure - Version N/A and later 

Linux x86-64

Symptoms:

The sys command in crash utilities displays the reboot time as "DATE: Sun Oct 27 10:06:43 UTC 2024," whereas the actual reboot occurred around "Oct 27 06:05:15"

crash> sys
KERNEL: /share/linuxrpm/vmlinux_repo/64/5.4.17-2136.322.6.3.el8uek.x86_64/vmlinux [TAINTED]
DUMPFILE: <hostname>.vmcore [PARTIAL DUMP]
CPUS: 96
DATE: Sun Oct 27 10:06:43 UTC 2024
UPTIME: 91 days, 15:36:28
LOAD AVERAGE: 1054.39, 1048.87, 1035.63
TASKS: 9662
—skip
crash>

The /var/log/messages log shows the following:

Oct 27 06:05:15 <hostname> root[139264]: INFO <HOSTNAME> - Begin nscd monitor on <hostname>
Oct 27 06:10:55 <hostname> kernel: Linux version 5.4.17-2136.322.6.3.el8uek.x86_64 (mockbuild@host-100-100-224-40) (gcc version 8.3.1 20190507 (Red Hat 8.3.1-4.5.0.8) (GCC)) #2 SMP Thu Oct 12 20:45:26 PDT 2023
Oct 27 06:10:55 <hostname> kernel: Kernel is locked down from EFI Secure Boot; see man kernel_lockdown.7
Oct 27 06:10:55 <hostname> kernel: Command line: BOOT_IMAGE=/vmlinuz-5.4.17-2136.322.6.3.el8uek.x86_64 root=LABEL=DBSYS bootarea=dbsys bootfrom=BOOT ro loglevel=7 panic=60 log_buf_len=1m nmi_watchdog=0 rd_NO_PLYMOUTH transparent_hugepage=never audit=1 console=tty1 console=ttyS0,115200n8 crashkernel=548M numa=on pci=noaer ifnames_skip=100 biosdevname=0 net.ifnames=0 sched_uek=wakeidle