[cfarm-users] cfarm439 crashes and downtime

[Jacob Bachmeyer]

On 6/7/26 07:50, Luke Yasuda wrote:
> On 2026-06-06 11:16, Jacob Bachmeyer wrote:
>> On 6/5/26 05:03, Luke Yasuda via cfarm-users wrote:
>>> Hi cfarm439 users,
>>>
>>> Sorry, this email was sitting in my "Draft" folder for too long. As 
>>> you know cfarm439 is a FreeBSD powerpc64le system. These days, 
>>> cfarm439 keeps crashing everyday and I have basically no clue what's 
>>> going on. Every time it crashes, it drops to FreeBSD kernel 
>>> debugger, I have to restart it manually, so much it became one of my 
>>> morning/evening routines. But when I'm traveling (like now), it's a 
>>> pain to do it over a high latency 4G LTE mobile hotspot ssh 
>>> connection, besides some small security concerns. Please forgive my 
>>> "mild to severe" delay in dealing with cfarm439. 
> [...]
>
>> What is the backtrace when it drops to the kernel debugger, i.e. 
>> where is it crashing?
>
> Ah. I forgot to mention the backtrace. I couldn't catch one because 
> the crash happened somewhat randomly, I need to connect to the virtio 
> serial port of cfarm439 at all times to catch it. But the last time I 
> tried to catch it, I left the serial console connected for a few days 
> and nothing happened and I felt I was being played, somehow. I should 
> try again to get the backtrace, maybe by redirecting stdout to a file 
> or something, but this is not high priority for now (something of 
> higher priority: sometimes could not recv/send emails).

Could you leave a session on the VM host under screen(1) connected to 
the virtio serial port with the session log enabled? Detach screen and 
leave it running on the host, and you should get the crash messages in 
screen's session log.


-- Jacob