And while I was not specifically thinking about "benchmarking different ntp.conf setups and measuring comparative offsets and jitter", I was thinking that it is generally true that the unknown unknowns start turning into known unknowns when actual reality gets to have a seat at the table. I saw that there was evidence that there was quite a lot "we didn't know that we didn't know", and also some evidence that there were "things we knew that weren't so".
..m On Mon, May 23, 2016 at 2:03 PM Mark Atwood <fallenpega...@gmail.com> wrote: > Please set an account up on thyrsus.com for me. > > > You can grab SSH keys like so: > curl -O https://github.com/<username>.keys > > > So thus get mine at https://github.com/fallenpegasus.keys > > > > ..m > > On Mon, May 23, 2016 at 1:54 PM Eric S. Raymond <e...@thyrsus.com> wrote: > >> Really good plans have happy consequences one didn't foresee. But also >> complicating ones. >> >> Mark encouraged me to work on the Microserver HOWTO as a recruitment >> device for future time-service experts and, implicitly, so I would >> learn more about how NTP works from the outside. Those were good >> reasons. >> >> Now it's growing another use that Mark may not have foreseen - >> benchmarking different ntp.conf setups and measuring comparative >> offsets and jitter. For these purposes, having several co-located >> machines with identical hardware/software configurations and the >> same Internet connectivity is ideal. >> >> However, this also creates an implicit bottleneck. Hal wants me to >> set up a machine with both a HAT and a GR701W and use it to measure >> offsets. It's a an excellent idea, but I can't do it in any >> predictable amount of time - I'm working as hard as I can just keeping >> the project-related backlog in my mailbox from overwhelming me. More >> generally, Eric as the only person who can use the test farm is not >> going to scale well. >> >> Accordingly, I'll make logins on the thyrsus.com bastion host >> available to project developers. The test farm is accessible from >> there. I'll make myself available to plug and unplug hardware >> on request. >> >> Current test farm inventory: >> >> au.local - Pi 2 with blue-wired SKU 424254 >> cu.local - Pi 2 with Uputronics HAT >> fe.local - Pi 3 with Adafruit HAT >> >> All three have skyview good enough that they pretty much always have >> lock except just after power-cycling. GR601-W or GR-701 USB GPSes >> can be plugged into any of these on request. >> >> Soon to arrive: >> >> * an Odroid C2 (ordered) >> >> * 8-port Netgear switch because my existing one is run out of ports - >> going to dedicate this one to the lab. POE capable because that's >> part of the plan for the BeagleBone variant (ordered) >> >> * an Anker 10-port powered USB hub, because Mark turned out to be >> unsurpringly right that el cheapo unpowered hubs aren't stable >> enough (ordered) >> >> * two more Pi 3/Adafruit HAT combinations and a HAT for the Odroid >> (not yet ordered, but I know I have to for the ntp.conf comparisons) >> >> Currently planning for a maximum of 10 test machines and an actual >> complement of 7 - 2 Pi2s, 3 Pi3s, an Odroid C2, and a Beaglebone Black. >> >> Sigh. I've spent years avoiding learning enough about network >> management to run a server farm this size. No longer, it seems. >> >> Funniest possibility: If I need a rack for these systems, >> >> https://www.youtube.com/watch?v=Jq5nrHz9I94 >> >> A functional rack made of legos. That's worth some geekery points >> right there. >> -- >> >>esr>> >> _______________________________________________ >> devel mailing list >> devel@ntpsec.org >> http://lists.ntpsec.org/mailman/listinfo/devel >> >
_______________________________________________ devel mailing list devel@ntpsec.org http://lists.ntpsec.org/mailman/listinfo/devel
