help understanding backend_health varnishlog line

[David Birdsong]

I'm trying to understand why varnish's health checks marks my backend
down at random intervals.  Could somebody explain the columns in these
lines?

varnishlog  -i backend_health  | grep -i sick
    0 Backend_health - mogfe Went sick 4--X--- 2 3 3 0.000000 0.000387

This is a healthy line:
  0 Backend_health - mogfe Still healthy 4--X-RH 3 3 3 0.000267
0.000315 HTTP/1.1 200 OK


My backend is nginx on the same machine.  Varnish connects over
127.0.0.1.  The check stanza in nginx simply returns a 1x1 gif that
nginx keeps around in memory so I'm really confused what could be
causing the bad response.

I captured one of the failed checks in a pcap file and searched on TCP
RST flag set and only found one session which I dont know whether it
corresponds to the check or not.  It went:

varnish: SYN time: 15.55156 seq=0
nginx: ACK   time: 15.551577 seq=1 ack=3034931510
varnish: RST time: 15.551589 seq=3034931510

...no http check after the RST obviously.