Varnish 503ing on ~1/100 POSTs

[Tollef Fog Heen]

]] Ronan Mullally 

| I'm seeing intermittant 503s on POSTs to a fairly busy VBulletin website.
| The current load is light (up to a couple of thousand active sessions,
| peak is around five thousand).  Varnish has a fairly simple config with
| a director consisting of two Apache backends:

This looks a bit odd:

|  backend backend1 {
|          .host = "1.2.3.4";
|          .port = "80";
|          .connect_timeout = 5s;
|          .first_byte_timeout = 90s;
|          .between_bytes_timeout = 90s;
| A typical request is below.  The first attempt fails with:
| 
|   33 FetchError   c http first read error: -1 0 (Success)

This just means the backend closed the connection on us.

| there is presumably a restart and the second attempt (sometimes to
| backend1, sometimes backend2) fails with:
| 
|   33 FetchError   c backend write error: 11 (Resource temporarily unavailable)

This is a timeout, however:

|    33 ReqEnd       c 657185708 1299604110.559967279 1299604113.447372913 0.000037670 2.887368441 0.000037193

That 2.89s backend response time doesn't add up with your timeouts.  Can
you see if you can get a tcpdump of what's going on?

Regards,
-- 
Tollef Fog Heen
Varnish Software
t: +47 21 98 92 64