varnishlog output column meaning

Tung Nguyen tnguyen at
Sun Mar 22 12:11:40 CET 2009

Ic, Poul-Henning thank you!
Hey guys,

Im still trying to get a handle how to debug these vcl rules best here..

Ive got some simple tests set up, mainly based on

With app #1 caching happens fine, but with app #2 with the same very general
vcl rules, I cant seem to get caching to work :(.

There's a few questions at the end of the gist:
1. Why isnt app #2 picking up the cache?  I think maybe the rails app
returns a slighly
different header that could be causing a problem but I dont see anything
different in the
varnishlog output :(.
2. Is there a way to check what is in the cache from the "telnet localhost
6082" or anything
else, if I have the key, "/articles;feed?tag_id=16" in this case.  Maybe
thats whats going on,
the data isnt being written into the cache...
3. I also tried adding some C code (commented out) to maybe puts some output
out or writing to
a file (maybe through a system call could work too) to see which part of the
vcl chain Im
hitting.  Whats a good way of doing that?  This would be great to know how
to do so I can
check the vcl rules.

Thanks again and any responses are appreciated.

On Sun, Mar 22, 2009 at 2:08 AM, Poul-Henning Kamp <phk at>wrote:

> In message <ca6364050903220201t1f50f65btd5b4904f5703f735 at>,
> Tung
> Nguyen writes:
> >Hey guys,
> >
> >In the the varnishlog output
> >
> >
> >
> >Where can I find out more info on what the columns mean,
> >So what does the 7 vs 10 mean in the first column?
> >What does the b vs c mean in the second column?
> The first column is a magic number for each transaction.
> This is what allows you collect all the messages that
> belong to the same transaction.
> For the time being, it is also the file descriptor number
> of the socket, but this is not guaranteed to always be the
> case.
> The 'b' means "backend transaction, and 'c' client side
> transaction.
> > 7 VCL_return   c pass
> >10 BackendOpen  b default 57141 3000
> > 7 Backend      c 10 default default
> >10 TxRequest    b HEAD
> --
> Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
> phk at FreeBSD.ORG         | TCP/IP since RFC 956
> FreeBSD committer       | BSD since 4.3-tahoe
> Never attribute to malice what can adequately be explained by incompetence.

Tung Nguyen, Lead Developer
Bleacher Report, The Open Source Sports Network
(510) 928-0475
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the varnish-misc mailing list