Controller kicking out probe with no clear reason?
Hi there, I run a probe with ID, 1001577. I had to restart it few days back and now it's just not connecting to the controller. It cycles through the list of controllers, Finds 1 controller it can connect to but that controller kicks it out. I haven't been able to figure out why this is happening. I have tried clearing out all the data related to that probe, restarting it and updating the probe key on the dashboard but nothing seems to be working. The probe dashboard is receiving stats data, Current IP address and all the other information but it's marked offline. I have attached the logs below. ssh -i /var/atlas-probe/etc/probe_key -p 443 atlas@193.0.19.247 INIT Got good controller info check cached controller info from previous registeration NO cached controller info. NO REMOTE port info Do a controller INIT Controller init -p 443 atlas@ctr-fsn01.atlas.ripe.net INIT initiating KEEP connection to -R 53662 -p 443 ctr-fsn01.atlas.ripe.net { "name": "lo", "bytes_recv": 0, "pkt_recv": 0, "errors_recv": 0, "dropped_recv": 0, "fifo_recv": 0, "framing_recv": 0, "compressed_recv": 0, "multicast_recv": 0, "bytes_sent": 0, "pkt_sent": 0, "errors_sent": 0, "dropped_sent": 0, "fifo_sent": 0, "collisions_sent": 0, "carr_lost_sent": 0, "compressed_sent": 0 }, { "name": "eth0", "bytes_recv": 12352, "pkt_recv": 52, "errors_recv": 0, "dropped_recv": 0, "fifo_recv": 0, "framing_recv": 0, "compressed_recv": 0, "multicast_recv": 1, "bytes_sent": 12456, "pkt_sent": 76, "errors_sent": 0, "dropped_sent": 0, "fifo_sent": 0, "collisions_sent": 0, "carr_lost_sent": 0, "compressed_sent": 0 }condmv: not moving, destination '/var/atlas-probe/data/out/v6addr.txt' exists condmv: not moving, destination '/var/atlas-probe/data/out/simpleping' exists /usr/local/bin/atlas: 95: kill: No such process no ssh client matching . cleanup state files. for next restart Controller kicked us out. KEEP result is nonzero kill the session and forced reregister /usr/local/bin/atlas: 88: kill: No such process RESULT 9003 done 1632454514 8ead83275bd8 REREGISTER Controller kicked us out from KEEP Can someone here tell me what do I need to do fix this? I have another probe that's connected to atlas@reg03.atlas.ripe.net and it's working perfectly. Is ctr-fsn01.atlas.ripe.net operating normally? Regards Ishan Jain
Hi Ishan, I see that you probe is currently connected and functioning as expected. The problem was that the probe was trying to reconnect to controller too often. For security reason if a probe is trying to connect to controller more than 10 times per hour the throttling kicks in. In such a case I would just stop the probe for an hour and try again. Best regards /vty On 9/24/21 5:44 AM, Ishan Jain wrote:
Hi there,
I run a probe with ID, 1001577. I had to restart it few days back and now it's just not connecting to the controller. It cycles through the list of controllers, Finds 1 controller it can connect to but that controller kicks it out. I haven't been able to figure out why this is happening. I have tried clearing out all the data related to that probe, restarting it and updating the probe key on the dashboard but nothing seems to be working.
participants (2)
-
Ishan Jain
-
Viktor Naumov