Hi Wes, Thanks for your response, I also tend to agree with having a more reliable parameter for zone-data publishing than the "load-time". I reviewed other load-time graphs and I believe the regular (green) spikes in this parameter probably don't indicate any potential problem. So in my opinion, future rssac002 reviews/versions could look for more indicative stats than the load-time. [image: image.png] Kind Regards *Hafiz Farooq* [image: https://www.linkedin.com/in/hmfarooq/] <https://www.linkedin.com/in/hmfarooq/> On Thu, Nov 30, 2023 at 2:15 AM Wes Hardaker <hardaker@isi.edu> wrote:
Hi Hafiz,
FYI, I was checking the load time graph for b.root-server but found it broken.
Sorry about the delay in getting back to you. We were rather busy with making sure our recent address change went smoothly.
I've looked into why we stopped publishing load-time's, and indeed we haven't in a while (long story about that section of our data processing pipeline stopping). We have a fix nearly in place and we should back-create as much of the data as we can so the graphs should hopefully reappear shortly.
We'll note that the load-time metric is not one that we believe is all that useful, except when things go really-really wrong (at which point it's probably not the metric that will warn people about the problem). I am curious whether or not anyone actually finds this metric useful, and if so how? In the past we've talked about removing it from RSSAC002 data requirements in the a future version of the specification unless there is a demonstrated need for it (my/our personal opinion, not necessarily the opinion of the other operators). We do intend to support it while it's in the specification though.
Cheers, -- Wes Hardaker USC/ISI
