12-07-2020 04:53 AM
Hi,
during the blade firmware upgrade on Windows OS, the server gets rebooted after the user acknoledgement, the OS has a hard reboot. the error in the event id is 41. the server was rebooted unexpectedly. is there any way that the OS can have soft reboot instead of hard reboot during blade firmware upgrade.
12-07-2020 05:07 AM - edited 12-07-2020 05:09 AM
So most 'reset/reboot' type actions taken by the UCSM needing to power cycle a blade/rack server involve acpi commands sent to the server OS, and then after a period (i think 120 seconds) if the UCSM doesn't detect any movement by the OS, it will do hard power reset.
There is an option in the user ACK/Maintenance Policies
for 'pending next reboot', which means the actual ucsm tasks, won't actually trigger the power cycle, but will wait for you to do the actually reboot via the OS.
Sounds like if the UCSM is having a hard time getting the OS to gracefully reboot, that the 'on next boot' option is your best option for a graceful acknowledgement.
Kirk...
12-07-2020 06:09 AM
can we shutdown the server manually, change the firmware policy and acknowledge reboot.
will this work?
12-07-2020 11:37 AM - edited 12-07-2020 11:38 AM
It's probably more scalable to get comfortable with the 'on next reboot' methodology, as you can line up/apply multiple servers sharing the same host firmware policy, and then just reboot individually from the OS.
Kirk...
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide