[atlas] Huge credit debit

On 2013.10.11. 10:05, Robert Kisteleki wrote:
> On 2013.10.10. 23:10, Stephane Bortzmeyer wrote:
>> On Thu, Oct 10, 2013 at 02:54:26PM -0400,
>>  Mike. <the.lists at mgm51.com> wrote 
>>  a message of 20 lines which said:
>>
>>> I don't know where or why the 764596 samples came from, usually only
>>> 90 samples are performed.
>>>
>>>
>>> Was there a proble in the measurement or credit system recently?
>>
>> Someone reported exactly the same problem on Twitter:
>>
>> https://twitter.com/bastien_durel/status/388407290262736897
> 
> I can confirm that the billing cycle yesterday afternoon went out of whack:
> we see that in some cases it overcharged (and in some other cases it
> undercharged) for measurement results. We've suspended billing for the time
> being and are investigating the cause for the issue.
> 
> Of course we'll compensate the affected users for the overcharge.
> 
> We'll get back with more details later.

Update: so far we haven't identified the root cause of the problem. However,
in order to prevent the issue form reoccuring, we deployed an emergency
update of the billing code that includes sanity checks about the charges and
refuses to bill unreasonable amounts. (This piece of code was scheduled for
deployment after the RIPE76 meeting.)

We also identified overcharged measurements and provided compensation for
these.