Thanks Peter for your fast answer !
I was thinking about watchdog stack itself for the software part. I had
no idea that it was able to manage the HW watchdog of the RPI4 :)
the watchdog stack is a little confusing because the documentation is so
small...especially when using HW module already loaded !
Installation and configuration done, I will have to monitor now for
unexpected reboot ( instead of hanged nodes :) )
On 1/8/23 02:54, Peter Robinson wrote:
> Hi Pierre-Francois,
>> I am running 6 RPI4s with fedora 37. K3S is powering this cluster and it
>> is working well :)
>> But from time to time, 1 RPI is randomly hanging.
>> I am thinking about implementing a watchdog :
>> - software based, using embeded linux kernel
> If the RPi itself is locking up I'm not sure how a SW based approach
> would work, how would it trigger if the kernel isn't working? Also not
> sure what you mean here but "embeded linux kernel", care to explain
> some more?
>> - hardware based such as https://www.omzlo.com/articles/the-piwatcher
> The RPI4 already has a HW watchdog onboard, the driver is bcm2835_wdt,
> maybe you could start by investigating the use of the Linux watchfog
> stack with that. systemd has support for HW WDTs and we enable all the
> pieces in Fedora already, I suspect that will be a good start for you.
>> Do you have any experience on one of theses two solutions ? Do you have
>> alternatives ?
>> By the way your job is fantastic and it is a great pleasure to be able
>> to run F37 on aarch64 so easily !
>> Thanks a lot
>> arm mailing list -- firstname.lastname@example.org
>> To unsubscribe send an email to email@example.com
>> Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
>> List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>> List Archives: https://firstname.lastname@example.org
>> Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
arm mailing list -- email@example.com
To unsubscribe send an email to firstname.lastname@example.org
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://email@example.com
Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue