MHO-175 Poweroff

[dj0Nz]

Hi Mates,

deployed four MHO-175 lately at a customers site. Wanted to power down the appliances after day one because they were running in an empty office room instead of the data center. Next day, the http daemon on one of the appliances did not start any more and we suspected that a corrupted file or file system could be the reason.

We then found out that the MHO appliances won't shut down with any of the known shutdown commands (halt -p, poweroff, shutdown -h now ...). Instead, if you issue one of these commands, it will do everything a "normal" shutdown procedure does and reboots afterwards.

Version is R81.10SP 338 with lates JHF.

I did not find anything in the official documentation so I am wondering if I'm just using the wrong shutdown procedure or if this is a bug and I have to contact TAC.

Thanks in advance!

Michael

[Chris_Atkinson]

MHO is not ACPI-compliant like other appliances. When you issue poweroff, it properly shuts down services and OS. Once it starts making a loud noise, you can safely unplug it.

[dj0Nz]

Thanks for answering, But that's not exactly workable. I would actually expect a stopped system to stay stopped (and indicate it using some LEDs like the 6x000/4x000 chassis) and not reboot after 10 seconds. 

Will open a case.

[Danny]

The system should stop if you execute the halt command as described here.
You can then power it off by unplugging the power cords or start it again by clicking the reset button.

[dj0Nz]

Thanks Danny but we already tried that. A "halt" (no matter, what "flavour" we use) indeed causes a system halt. But after a short period (10 sec approx), the system reboots. Case opening is pending... 😉

[dj0Nz]

Addendum. Found in sk174202 (How to replace a Quantum Maestro Orchestrator): "[...] Important - Do not halt (shut down) the Orchestrator that failed. It might reboot, and the Orchestrator services will start again. [...]"

Maybe there is some kind of overlay / read only file system in use to prevent corruption?