Hi, Danny at al.,
The problem is not with px4, but instead with pf4 which has gone down (again).
I had forgotten that px4 is directly connected to pf4 instead of the switch.
How many observing days are left? Would it be reasonable to ask Matthys to pay
another visit to the PAPER container?
Dave
On Feb 17, 2015, at 8:14 PM, David MacMahon wrote:
> Hi, Danny,
>
> This is NOT another IERR event. px4 is up, but it is not getting any
> packets. In fact, it looks like none of the 10 GbE interfaces even have a
> link!
>
> davidm@px4:~$ for e in eth{2..5}; do sudo ethtool $e | grep detected; done
> Link detected: no
> Link detected: no
> Link detected: no
> Link detected: no
>
> I have no idea what would have caused this other than cabling problems. I'm
> going to reboot.
>
> Dave
>
> On Feb 17, 2015, at 7:58 PM, David MacMahon wrote:
>
>> Hi, Danny,
>>
>> Looks like px4 had another IERR event. Observing is about to end for the
>> night there (in about 3 minutes from this writing) so I'll fix it up after
>> that and it should be all good to go for tomorrow night.
>>
>> Dave
>>
>>
>> Begin forwarded message:
>>
>>> From: [email protected] (David MacMahon)
>>> Date: February 17, 2015 6:50:01 PM PST
>>> To: [email protected], [email protected]
>>> Subject: PAPER Correlator Status Report
>>>
>>> 2015-02-18T02:50:01+00:00
>>>
>>> Got OUTDUMPS for all X engine instances
>>> Bad OUTDUMPS value for X engine instance px4/0
>>> Bad OUTDUMPS value for X engine instance px4/1
>>> Bad OUTDUMPS value for X engine instance px4/2
>>> Bad OUTDUMPS value for X engine instance px4/3
>>>
>>> Got GPUGBPS for all X engine instances
>>> GPUGBPS for present X engine instances all in range
>>>
>>> Detected 4 errors
>>
>
