- ripe-atlas - mailman.ripe.net

[atlas]FYI probe problems after power failure
by Jelte Jansen 03 Dec '10

03 Dec '10

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Today I had a power failure, and luckily everything came back up when I fixed it. However, a while later, I wondered if the probe had too, and it had not. The status page said it was up, but it hadn't reported data for about 6 hours (grey graphs). I assume I hit one of the problems in the FAQ: - --- Technical background: the probe tries to get current time indication using NTP upon booting, which sometimes fails. This causes the probe to think the curent date is November 1999, which confuses it a lot; it doesn't register properly or it sends measurement data for the twentieth century which is immediately discarded by the system. We plan to convince the probe to insist on vaild NTP or acquire some other time indication. The fix will be deployed in a firmware update. - --- Since the probe boots a lot faster than my gateway, I guess it never retried NTP. Powercycle fixed it indeed. So the lesson here is (at least until there is an update that retries ntpdate, or something), if the machine that provides your Internet connection takes a while to boot up, and they are both powered on at the same time, you may need to powercycle your probe. Jelte -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkz4HX0ACgkQ4nZCKsdOncVlXgCfZe+3FbLdTWTxCvXYZFrmaXjd TsEAn3vqmfjbUmWfueVcziCi2a3Hs6jF =zqNQ -----END PGP SIGNATURE-----

2 1

[atlas]Inaccurate uptime tracking?
by Richard L. Barnes 02 Dec '10

02 Dec '10

Hey guys, I'm seeing an inconsistency between the reported uptime and the actual measurements from probe #394. On the one hand, the probe info page reports only a little over an hour of uptime: Total Uptime: 0d, 1h, 1m Connect at: Disconnect at: 2010-12-02 02:38:06 UTC (still connected) 2010-12-01 04:22:43 UTC 2010-12-01 04:28:42 UTC But the graphs for the last 8 hours show only a brief down-time; roughly: Disconnect: 2010-12-02 02:27:00 UTC Reconnect: 2010-12-02 02:35:00 UTC This wouldn't bother me much if uptime weren't tied to chances to win an iPad :) --Richard

2 2

[atlas]atlas win!
by Florian Obser 01 Dec '10

01 Dec '10

Hi, we are running two probes. One behind my private dsl line at home, one in our data centre. I noticed that the IPv6 ping times to m.root-servers.net are more then twice as bad from our data centre compared to the dsl probe. As it turns out at home I'm talking to the m-root instance in Paris while in the data centre we were talking to the instance in Tokyo(?) It looks like this: http://eafc8bd05a0435ef14d31bbff29687dd7b71c3ca.de/m.root-servers.net.v6.png Not _that_ important in the great scheme of things but non the less quite nice ;) Thanks! Best regards, Florian

2 1

[atlas]P466 - inconsistent readings?
by Wilfried Woeber, UniVie/ACOnet 30 Nov '10

30 Nov '10

Hi Team, could you pls. have a look at the grpahs of 466 and the list of "perceived" outages for the probe? While I can't prove that there wasn't some problem with my ADSL link at home (while sitting in the office :-) ), the graphs for the pings seem to indicate that the probe was alive and kicking all the time. I may be misreading some stuff, though.... Wilfried

2 3

[atlas]some comments regarding the user interface - re-sizing and outage(s)
by Wilfried Woeber, UniVie/ACOnet 30 Nov '10

30 Nov '10

Just minor cosmetics, and for the to-do-list :-) - the probe and measurements display reacts to a resizing of the window, e.g. from partial screen to full-screen, by redrawing the inforamtion and using all the real estate, but does NOT redraw on a switch back to a smaller size. (IE7, W/XP prof. SP3) - regardig the "Uptime" panel, I'd like to see the most recent outage included by default (the UTC stamp of most recent down transition, the most recent up transition and the duration). Alternatively, I'd be happy to make the selection for "Show last 25 connections sticky, maybe by giving the number of entries to display. Cheers, Wilfried.

2 1

[atlas]how many of us are on this list?
by Wilfried Woeber, UniVie/ACOnet 29 Nov '10

29 Nov '10

Hi folks, just had a look at the graphs and at the # of people "talking" on this list... There are some 130 probes "up" and some 150 "ready" :-) Now I am wondering, how many subscriptions does this list have by now? Have a nice weekend, all, Wilfried.

7 8

[atlas]IPv4 DHCP and varying addresses
by Wilfried Woeber, UniVie/ACOnet 25 Nov '10

25 Nov '10

Having a simple question regarding DHCP and varying addresses within a subnet... While I *do* have access to the config of the DHCP server that delivers addresses to 414, and thus the probe will always get that same IP address, I *do not* have access to configure the DHCP service in my ADSL CPE (yet), which means that the individual address given to the probe may change over time. I guess this is expected and being taken care of by the system, but wanted to double-check. Thanks, Wilfried.

2 1

[atlas]The probes and IPv6
by Wilfried Woeber, UniVie/ACOnet 25 Nov '10

25 Nov '10

Just had a look again at the info page for my 414, in the "Configuration" panel. While this thing does have IPv6 connectivity and performs measurements just OK, the section "IPv6", next to the MAC Address, is empty. Bug or "reserved for future use"? While we did the EUI-64 math locally to generate a DNS AAAA entry derived from the MAC, some folks may be better off to see the auto-conf'd address. As an aside, for the "ideas stack", it would be hip to have a means of educating the probe about an IPv6-transport-capable DNS server. I know, there are some issues with this idea :-/ All the best, Wilfried. PS: Looking forward to see probe #2, 466, becoming visible as living in my village, after making pole position for #1 already :-)

2 2

[atlas]Some issues today
by Robert Kisteleki 24 Nov '10

24 Nov '10

Dear All, As you already know, RIPE Atlas is in a prototype phase. A good illustration for it is an issue we had today. We're still investigating the cause, but in the meantime we had to ignore some data points for a subset of the probes -- please don't be surprised to see missing parts of your probe graphs for today. Thanks for your understanding, Robert

1 0

[atlas]Graphs gone...
by Wolfgang Tremmel 24 Nov '10

24 Nov '10

Hello, Probe is up (and shows up) but all the graphs show empty since yesterday. best regards, Wolfgang

4 3

[atlas]warning pop-up in IE for the various atlas-related pages
by Wilfried Woeber, UniVie/ACOnet 24 Nov '10

24 Nov '10

Folks, I noticed that I have to ack a warning pop-up (served by IE7 ;-) ) for each page that is realted to the atlas stuff. IE complains about a mix of secured and unsecured components on the page. In principle I'd prefer continuing with httpS, but the warnings are a real nuisance. Anyone else seeing this? Btw, the Labs stuff shows the same behaviour. Wilfried.

4 5

[atlas]internal timing constraints?
by Wilfried Woeber, UniVie/ACOnet 24 Nov '10

24 Nov '10

Dear Atlas folks, do you have a rough estimate or data about the internal timing of the probes, like what sort of "delay" is added by the probe itself when measuring e.g. ping RTT? The background is that I get different readings or ping RTTs from the probe, compared to the "same" thing done from a regular PC running Linux and sitting on the very same subnet, connected to the very same (small) switch. The absolute difference I see is small (sub-millisecond), but it seems to be systematic. The ping target is a "local" instance of i.root, btw. Thanks, Wilfried.

2 1

[atlas]probe appears down on atlas.ripe.net
by Nuno Vieira - nfsi telecom 24 Nov '10

24 Nov '10

Hello, I have deployed one probe and it has been up since a few hours ago. I checked this morning and it was down. The thing is, i can ping properly the probe on IPv4 and IPv6, and i see traffic comming in and out. We did not changed anything on the network/firewall to prevent traffic. What could be the problem ? regards, --nvieira

2 1

[atlas]labeling the axes in the graphs
by Wilfried Woeber, UniVie/ACOnet 23 Nov '10

23 Nov '10

Dear probe hosts and "herders", a couple of first experiences and thoughts... While I know that RTT is measured in "time" rather than distance, I was slightly detracted in the beginning by he use of "m" as the abbreviation for milli seconds. I guess there would be enough room in the graphs to use "ms"? Maybe it is just me :-) Wilfried.

1 0

[atlas]Upgrades today
by Robert Kisteleki 10 Nov '10

10 Nov '10

Hi, We're doing some major upgrades today on the RIPE Atlas network, both on the central architecture and on the probes. It's expected that during this time probes will disconnect. Regards, Robert

4 7

[atlas]is this correct
by Bert (IETF) Wijnen 10 Nov '10

10 Nov '10

looking at "Last 25 connections", I vaguely suspect that this is somehow not correct (at least not what I expected) Bert *Current status:* Up since 2010-11-09 19:23:06 UTC *Total Uptime:* 20d, 17h, 7m Show last 25 connections Hide last 25 connections *Connect at:* *Disconnect at:* 2010-11-09 19:23:06 UTC (still connected) 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC 2010-11-05 09:53:28 UTC 2010-11-05 21:53:31 UTC

2 4

[atlas]test
by Tom Sylwestrowicz 14 Oct '10

14 Oct '10

test mail to check mailing list works

1 0