Re: [atlas] Probe 2285 is disconnected (Randy/Bainbridge Comcast)
i keep getting these
Your probe 2285 has been disconnected from the RIPE Atlas infrastructure since 2020-03-18 06:43:44 UTC.
status shows it down but i can ping6 the probe, and from the local gateway, i can ping4 it's natted address psg.com:/usr/home/randy> ping6 2601:600:c680:2fe:220:4aff:fee0:246d PING6(56=40+8+8 bytes) 2001:418:1::62 --> 2601:600:c680:2fe:220:4aff:fee0:246d 16 bytes from 2601:600:c680:2fe:220:4aff:fee0:246d, icmp_seq=0 hlim=56 time=13.533 ms 16 bytes from 2601:600:c680:2fe:220:4aff:fee0:246d, icmp_seq=1 hlim=56 time=12.370 ms root@wrt-biwa:~# ping 192.168.0.27 PING 192.168.0.27 (192.168.0.27): 56 data bytes 64 bytes from 192.168.0.27: seq=0 ttl=64 time=0.921 ms 64 bytes from 192.168.0.27: seq=1 ttl=64 time=0.701 ms network tab shows a long history of very short disconnects there is a strat 1 sitting on the same lan, and a lot of chimers out in the world think the strat 1 is preferred. no, that is not a definitive test in any way, but it's all i can think of at my current caffeine level. comcast? modem? v1 probe? how to better diagnose? randy
Hi Randy Not sure if it helps but once I had a similar issue where the probe was getting disconnected after every 4-5 days. I could see probe was reachable on the LAN and even had a smokeping instance pointed to it which was all clean. My own internet behind the same network was working fine. Eventually found that issue was with the Ubnt edgerouter which used tweaked VyOS and in their latest version 2.0.8 it was breaking the secondary routing tables running on the device after every 4-5 days of uptime. So except the main table, rest all tables were in config but just didn't show up in "show routing table X" output. Eventually downgraded the OS version and waiting for an update in next release notes of the firmware update. On Wed, Mar 18, 2020 at 6:23 PM Randy Bush <randy@psg.com> wrote:
i keep getting these
Your probe 2285 has been disconnected from the RIPE Atlas infrastructure since 2020-03-18 06:43:44 UTC.
status shows it down
but i can ping6 the probe, and from the local gateway, i can ping4 it's natted address
psg.com:/usr/home/randy> ping6 2601:600:c680:2fe:220:4aff:fee0:246d PING6(56=40+8+8 bytes) 2001:418:1::62 --> 2601:600:c680:2fe:220:4aff:fee0:246d 16 bytes from 2601:600:c680:2fe:220:4aff:fee0:246d, icmp_seq=0 hlim=56 time=13.533 ms 16 bytes from 2601:600:c680:2fe:220:4aff:fee0:246d, icmp_seq=1 hlim=56 time=12.370 ms
root@wrt-biwa:~# ping 192.168.0.27 PING 192.168.0.27 (192.168.0.27): 56 data bytes 64 bytes from 192.168.0.27: seq=0 ttl=64 time=0.921 ms 64 bytes from 192.168.0.27: seq=1 ttl=64 time=0.701 ms
network tab shows a long history of very short disconnects
there is a strat 1 sitting on the same lan, and a lot of chimers out in the world think the strat 1 is preferred. no, that is not a definitive test in any way, but it's all i can think of at my current caffeine level.
comcast? modem? v1 probe?
how to better diagnose?
randy
-- Anurag Bhatia anuragbhatia.com
hi anurag
Eventually found that issue was with the Ubnt edgerouter which used tweaked VyOS and in their latest version 2.0.8 it was breaking the secondary routing tables running on the device after every 4-5 days of uptime.
lovely in this case, it is a pretty stable (seming) openwrt on a netgear 3800. but, when i am next on prem, i will power and plug everything. randy
participants (2)
-
Anurag Bhatia
-
Randy Bush