I have observed the following error while looking at unrelated things in the logs of ats-be:
Jul 13 09:04:09 cp1081 traffic_manager[1133]: [Jul 13 09:04:09.249] [ET_NET 63] ERROR: SSL connection failed for 'planet.wikimedia.org': error:1416F086:SSL routines:tls_process_server_certificate:certificate verify failed
Indeed the public hostname "planet.wikimedia.org" is missing from the certificate subjectAltName:
DNS:planet.discovery.wmnet, DNS:planet.svc.eqiad.wmnet, DNS:planet.svc.codfw.wmnet, DNS:planet1001.eqiad.wmnet, DNS:planet2001.codfw.wmnet, DNS:*.planet.wikimedia.org
Note that *.planet.wm.org is there, but planet.wm.org is missing.
We should:
- fix the certificate
- make sure these kind of errors end up in logstash