<div dir="ltr">Hi Geoff, <div><br></div><div> Thanks for your input. But I remembered that I had an issue like this a while back. The problem was that the connect timeouts on my backend definitions was set too low. There is a bit of latency between the varnish servers and the web servers. So I raised the connect timeout from 30s to 60s for each of the back ends. That seems to do the trick! And now it seems that the whole site is caching.</div><div><br></div><div>Thanks!</div><div><br></div><div>Tim</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Sun, Jul 5, 2015 at 5:35 AM, Geoff Simmons <span dir="ltr"><<a href="mailto:geoff@uplex.de" target="_blank">geoff@uplex.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">-----BEGIN PGP SIGNED MESSAGE-----<br>
Hash: SHA256<br>
<span class=""><br>
On 7/5/15 2:54 AM, Tim Dunphy wrote:<br>
><br>
> But my current problem is that I've been able to get the main page<br>
> to the site cached, and a few others. But when I click to like the<br>
> 4th or 5th page on the wiki that I want to cache through varnish,<br>
> everything comes to a halt. I get a 503 - service unavailable<br>
> error, instead of varnish going to the web server to retrieve the<br>
> page I had clicked on.<br>
<br>
</span>503 errors are in most cases an indication that there's a problem with<br>
your backends, and Varnish cannot retrieve responses. When that<br>
happens, you'll often get some info from the log in the FetchError<br>
line, as you did here:<br>
<span class=""><br>
> 10 FetchError c no backend connection<br>
<br>
</span>No backend connection usually means that health checks are failing --<br>
in your case, they're failing for both backends that are assigned to<br>
your director. When health checks fail, Varnish doesn't attempt a<br>
fetch at all and sends a 503 response immediately.<br>
<br>
When you see that, check the Backend_health entries in the log. They<br>
describe the current health state (you'll probably see "Still sick"),<br>
and at the end of the log entry they either show the response status<br>
that the checks are getting; or if they show no response status, then<br>
the checks aren't getting any response. That should help you diagnose<br>
the problem.<br>
<br>
<br>
HTH,<br>
Geoff<br>
- --<br>
UPLEX Systemoptimierung<br>
Scheffelstraße 32<br>
22301 Hamburg<br>
<a href="http://uplex.de/" rel="noreferrer" target="_blank">http://uplex.de/</a><br>
Mob: <a href="tel:%2B49-176-63690917" value="+4917663690917">+49-176-63690917</a><br>
-----BEGIN PGP SIGNATURE-----<br>
Version: GnuPG/MacGPG2 v2.0.14 (Darwin)<br>
<br>
iQIcBAEBCAAGBQJVmPpwAAoJEOUwvh9pJNUREXUP/1EMnxHq8BmANhCL+8qwDpO4<br>
sTOdspRdHcpidtyR2MzRpjsZSBmvwjCze4ShMT0sop7Jp9qytlB6jh+DEOMNa/wd<br>
PF+1LGkWNZiNya7BBB/RSGmpXL/t9wQQWxtu4UpyV2Nqra2ktzEburxCPytV2Rwx<br>
ph4AAMiSMzI0mIimRbihmdNDL+/LTTXFIFPLJ8LLjMy7/ynB7wLlaRb6999d7PVq<br>
0RIglbDCjvIKz87vtcjG6gAuSHIs7BpODNjZ04UWTMfFqAtYk+UwLj1ALS32/1mO<br>
K5AeDYU6x8UqR9ja4j/41rYeCpE7SwPeuZoVB7GmuJTSVZ8O56dZwne/veVEzrup<br>
uyl3oXuzYFqPu9Aw1aBZc7oVJJiw+97Xf7q6tPW83a1iWbJlg3pMy1+vtNOIrbKu<br>
Z9Anm1Jr8nB1Esb7X1xS3+YPjK4klqnnwR0Mod1MMwhcWkg/iqF3X91hPXzU8YId<br>
KoET7lSpGlHBnr1Czx4IuV/yjfn/iJJta67vljqeFMY6zZId+KYYhQUQgl9J+mDq<br>
q8xxAeybYHwA3t9L9OZ78eJBhMTi31V7OI+A7aTvx2L198bGTZIIQP4V/bGctSm2<br>
CQ/g8FHVqovgtFfULVj0gkfr6+T0a5XWrpRYd1HcdtaaU2lptTcOIY33GBun3nHU<br>
MOnwzewxqXJueu1H1MWZ<br>
=4iaZ<br>
-----END PGP SIGNATURE-----<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">GPG me!!<br><br>gpg --keyserver <a href="http://pool.sks-keyservers.net" target="_blank">pool.sks-keyservers.net</a> --recv-keys F186197B<br><br></div>
</div>