Dear Gunther, Thank you for your response.

money since one can use the original robot in combination with a test-object or something like that.

Using that method would however generate tickets for the NCC Internet Protocol Resource Analysts who would have to then manually check if they were test tickets and then close them. -- Best Regards Timothy Lowe X-Organization: RIPE Network Coordination Centre X-Phone: +31 20 535 4444 X-Fax: +31 20 535 4445 On Thu, Oct 20, 2005 at 04:34:06PM +0200, gstammw@gmx.net wrote:

Hello,

I think that this would be a nice feature in general but also a waste of money since one can use the original robot in combination with a test-object or something like that. Better save the money.

Gunther

On 20 okt 2005, at 19.01, Gert Doering wrote:

Hi,

On Thu, Oct 20, 2005 at 02:45:08PM +0200, Timothy Lowe wrote:

...

One of our members has asked the RIPE NCC to provide a test facility for the RIPE NCC ticket robot.

Very useful thing (if it can be done without spending a year's wage on programmers time :-) ).

I agree. I think this would be a useful system and should be a fairly straight forward copy of the existing system. - kurtis -

Hello Carsten, Thank you for that information. To answer your question we don't know how much extra work using the production robot would cause our resource analysts as we don't know how many tickets would be created by testing automated resource managment systems, however we do know that each ticket so created would have to be manually processed by them. Please note that creating a ticket robot test facility would require approximately two days for one programmer. -- Best Regards Timothy Lowe X-Organization: RIPE Network Coordination Centre X-Phone: +31 20 535 4444 X-Fax: +31 20 535 4445 On Thu, Oct 20, 2005 at 11:03:04PM +0200, Carsten Schiefner wrote:

Hi Timothy,

Timothy Lowe wrote:

...

One of our members has asked the RIPE NCC to provide a test facility for the RIPE NCC ticket robot.

I don't have any particular view here as I don't have to deal with the robot these days.

Having said that, DENIC, the german ccTLD registry, has a test registry system ready for its members for more than x; x > 3 years now. And in the absence of any negative feedback I consider it (happily) welcome by them.

The real difference I'd see is that DENIC's robot is really an automat with no human intervention - when the NCC's ticket robot by its very nature requires that (aka. manpower). But I might be wrong here, see first para.

So for the time being some of the comments here hold somewhat true for me - is there an estimation of the additional workload on the IP Resource Analysts to process these test tickets? Even by only alternately approving and rejecting every single one of them? But maybe even this human intervention can be automated in the test system on a per LIR basis then?

Cheers,

Carsten

Dear Colleagues, Thank you all for your input. We will go ahead with this project starting Monday. Have a nice weekend everyone. -- Best Regards Timothy Lowe X-Organization: RIPE Network Coordination Centre X-Phone: +31 20 535 4444 X-Fax: +31 20 535 4445 On Fri, Oct 21, 2005 at 12:29:03PM +0200, Jeroen Massar wrote:

On Fri, 2005-10-21 at 10:36 +0200, Timothy Lowe wrote:

...

Please note that creating a ticket robot test facility would require approximately two days for one programmer.

Go for it and set it up. Having test tickets in the production system and having the analysts mark them as test gives more overhead in the long run.

my 1 cent of your local currency ;)

Greets, Jeroen