CS student here with probably a dumb question: why does the loop stop after 90 seconds? Is there some BGP safety mechanism or did Contabo manually fix it?
Also, what is BCP? I saw kenji3 mention "their BCP kicked in."
CS student here with probably a dumb question: why does the loop stop after 90 seconds? Is there some BGP safety mechanism or did Contabo manually fix it?
Also, what is BCP? I saw kenji3 mention "their BCP kicked in."
BCP = Business Continuity Plan, or in this context probably just their runbook for "stop the bleeding." Could be automated, could be a NOC human pulling a lever. 90 seconds suggests automated.
I have the full pcap if anyone really wants it but it's 400MB of looped SYNs. Not very exciting.
Update: Contabo just replied to my ticket. They confirm "a routing policy template was incorrectly applied to clean-path advertisements from the Singapore scrubbing center." Fixed now, no recurrence expected. No mention of credits.
I got the same copy-paste reply 20 minutes ago. "Fixed now, no recurrence expected." They didn't even change the template between customers.
Still no credit. I asked explicitly.
Kenji3, you should post this to LowEndBox forum also. More people need to know Contabo Singapore scrubbing got this problem. I help you write summary if you want.
Has anyone tested since their "fix"? I just activated scrubbing on my Singapore VPS 30 minutes ago and the tunnel came up fine, no loop. But that's one data point, might be luck.
Also re: OVHcloud comparison, their Singapore scrubbing is €15 more per month for similar bandwidth. Not nothing if you're price-sensitive.