Per-Entity Metadata Working Group

[Nick Roy <>]

> On Aug 11, 2016, at 11:04 AM, Chris Phillips 
> <>
>  wrote:
> 
> We use DNSMadeEasy with site monitoring between Toronto Ontario and
> Vancouver BC where we have our mirrors and haven't looked back:
> 
> https://www.dnsmadeeasy.com/services/dnsfailover/
> 
> 
> If you have an IP address endpoint that is consistent (load balanced to
> nodes within your environment/datacenter/cloudprovider) you apply a health
> check URL to verify reachability and alive status. If failure occurs DNS
> is changed real-time.
> 
> Service quality has been high and costs are economical.

The documentation says that the checks happen every 2-4 minutes.  That’s not 
good enough for what Scott is asking, it seems.

Nick

> 
> C
> 
> 
> On 2016-08-11, 12:51 PM, 
> "
>  on behalf of
> Nick Roy" 
> <
>  on behalf of
> >
>  wrote:
> 
>> And if so, could we do that in a way that still leverages the geo IP
>> capabilities of the multiple CDNs?  Straight DNS round-robin will be
>> problematic in a number of ways:
>> 
>> 1) No easy way to detect down CDN nodes and take them out of the mix
>> (probably could be scripted, the script itself would probably be a
>> significant investment and possibly a point of failure)
>> 
>> 2) Might not be compatible with geo IP (I don¹t know - does anyone here
>> know?)
>> 
>> Thanks,
>> 
>> Nick
>> 
>>> On Aug 11, 2016, at 10:47 AM, Scott Koranda 
>>> <>
>>>  wrote:
>>> 
>>>> On Wed, Aug 10, 2016 at 5:12 PM, Cantor, Scott 
>>>> <>
>>>> wrote:
>>>> 
>>>>> Of course, usually one extended outage is much, much better than
>>>>> constant short ones. I'm more interested in monthly or weekly than
>>>>> annual.
>>>> 
>>>> Cloud harmony monitors most providers and give a dashboard for weekly
>>>> or the last month.
>>>> Here is the CDN dashboard: https://cloudharmony.com/status-of-cdn
>>>> 
>>> 
>>> Thanks Patrick.
>>> 
>>> Naive question: could an organization like InCommon publish
>>> across multiple, different CDN providers in order to decrease
>>> still further the likelihood of going completely dark?
>>> 
>>> Scott K
>>> 
>>> 
>> 
>