[PLUG] Devise a remote log

Dick Steffens dick at dicksteffens.com
Thu Jul 9 22:36:00 UTC 2020


On 7/9/20 3:24 PM, Mike C. wrote:
>> The problem is that at some point it hangs. The only way to get it to work
>> again is to power it off and then restart it. I'd like to devise a log of
>> some sort that I can run from another machine that will give me some idea
>> of what was going on when it hung. I envision some kind of ping from my
>> main machine to
>> the troublesome one. I'm assuming I could use a script run by CRON (if
>> that's saying it right). Is there something more useful than just ping?
>>
> The good news is there are a lot of logging and monitoring tools. The bad
> news is that if you don't have some idea of what you're looking for or what
> tools give you what information, this can really complicate matters.
>
> You say the machine is up 24/7. For what purpose? Mail server, web server,
> dns cache, etc?

Right now it's my spare machine. It's available if something goes wacky 
with my primary machine and I need to keep working. I don't recall the 
last time that happened, so it's pretty rare. So, nothing in particular 
is running, other than the usual OS stuff. It used to be the machine I 
had in the living room with the HDMI port on the nVidia card connected 
to the TV. I used to use it when we wanted to watch something together. 
Right now I'm using the machine that used to be my spare machine. The 
stuff we watch on it is two Zoom meetings a week, plus anything we both 
want to look at on YouTube, or similar. So, not heavily used.

> How often does it hang? Is there any kind of discernable pattern with an
> app you're using, a background process that's running, etc?

Since it sits more or less idle for long periods, several days in 
length, I don't know when it hangs. Right now, unless I think about 
looking to see if it's still up, I won't know if it's hung or not.

> If it's hanging pretty regularly with short intervals between each hang,
> I'd maybe start with having HTOP or Conky running to watch processes and
> resource usage.
>
> There are a lot of nice syslog viewer apps that have filters you can
> configure that may help expose historical patterns.
>
> I find that anything that can help me see historical patterns or provide
> context is not only helpful but usually necessary because otherwise you're
> left with searching the forums of the great interwebs for some insight on
> what is usually a very cryptic error message that's generally not useful to
> the user.

Kinda hard to find a pattern in system not used often.

Thanks for the thoughts.

-- 
Regards,

Dick Steffens




More information about the PLUG mailing list