It was actually the output of fortune.
On 23 Oct 2014 21:47, "Winston Kodogo" <kodogo@gmail.com> wrote:

> Now I'm even more confused than normal. "cpu: can't dial: plan9.lanl.gov:
> The operation completed successfully."
>
> This is a Windows error message?
>
> On 23 October 2014 09:04, Quintile <steve@quintile.net> wrote:
>
>> I fear a gnu style recursive definition coming on...
>>
>> -Steve
>>
>>
>>
>>
>>
>> On 22 Oct 2014, at 19:14, Skip Tavakkolian <skip.tavakkolian@gmail.com>
>> wrote:
>>
>> i think this situation is more fortune-worthy than the fortune that
>> caused it.
>>
>>
>> On Wed, Oct 22, 2014 at 8:19 AM, Mats Olsson <plan9.meo@gmail.com> wrote:
>>
>>> I kind of had a feeling it was that way because when installing again
>>> on another card, I got another message with this: If you think out
>>> loud you're about to get a lot of ememies; as the bottom line (don't
>>> remember the exact words).
>>>
>>> 2014-10-22 17:12 GMT+02:00, Kurt H Maier <khm@sciops.net>:
>>> > Quoting Charles Forsyth <charles.forsyth@gmail.com>:
>>> >
>>> >> On 22 October 2014 15:34, Kurt H Maier <khm@sciops.net> wrote:
>>> >>
>>> >>> Quoting Mats Olsson <plan9.meo@gmail.com>:
>>> >>>
>>> >>>
>>> >>>  cpu: can't dial: plan9.lanl.gov: The operation completed
>>> successfully.
>>> >>>>
>>> >>>
>>> >>> this exact error message is in the fortunes file.
>>> >>
>>> >>
>>> >> oh well, that explains that: obviously the rio start-up on the pi runs
>>> >> fortunes, to aid debugging.
>>> >
>>> > That is precisely what is happening.  The startup script Steve Simon
>>> sent
>>> > runs his logwin script, which looks like this:
>>> >
>>> > #!/bin/rc
>>> >
>>> > fortune
>>> > calendar -y
>>> > news
>>> > echo
>>> >
>>> > exec rc -i
>>> >
>>> > ...now compare Mats' output:
>>> >
>>> > cpu: can't dial: plan9.lanl.gov: The operation completed successfully.
>>> > calendar: can't open /usr/glenda/lib/calendar:
>>> > '/usr/glenda/lib/calendar' does not exist
>>> >
>>> > khm
>>> >
>>> >
>>> >
>>>
>>>
>>
>