- 15 Jun, 2017 4 commits
-
-
Martin Blix Grydeland authored
-
Poul-Henning Kamp authored
object creation. Fixes #1764 Conflicts: bin/varnishd/cache/cache_fetch.c bin/varnishd/storage/stevedore.c bin/varnishd/storage/storage.h bin/varnishd/storage/storage_lru.c bin/varnishd/storage/storage_persistent.c bin/varnishd/storage/storage_simple.c
-
Geoff Simmons authored
-
Geoff Simmons authored
Conflicts: bin/varnishd/cache/cache.h bin/varnishd/cache/cache_ws.c
-
- 12 Jun, 2017 10 commits
-
-
Federico G. Schwindt authored
-
Dridi Boukelmoune authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
The parsing still requires some TLC but this does not make it any worse so let's get this in first. Prompted by and ideas from github::xcir via #2077.
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
Change some tests to use the new parameter and more coverage.
-
Poul-Henning Kamp authored
be found are options to "shell". Deprecate err_shell. Structure output to always be prefixed by "shell" since we're running at the top level. Also prefix commands with "exec 2>&1 ;" to join stdout+stderr by default. This still doesn't cover the shells stderr in case of bad shell syntax, which is annoying, but not worth fighting for right now.
-
Federico G. Schwindt authored
-
Federico G. Schwindt authored
-
- 06 Jun, 2017 3 commits
-
-
Federico G. Schwindt authored
Otherwise the backend might see multiple Connection entries. Fixes #2337.
-
Akshay Ranganath authored
In the section related to make test, the sentences were a bit hard incorrect. I've corrected it.
-
Dridi Boukelmoune authored
Fixes #2329
-
- 21 May, 2017 3 commits
-
-
Dridi Boukelmoune authored
-
Enerdhil authored
-
Dridi Boukelmoune authored
This is a cherry-pick of 3271c51e that fixed the issue for Varnish 3.0 since it wasn't a problem in the 4.1 series then. This is only a back-port of the regression test since the referenced commits already fixed the regression. Refs 5e4d6c84 Refs 9f422620 Refs #2069
-
- 26 Apr, 2017 3 commits
-
-
Pål Hermunn Johansen authored
-
Denes Matetelki authored
-
Denes Matetelki authored
Fixes: #2313 This is a back port of f1741d99 from master.
-
- 25 Apr, 2017 1 commit
-
-
Jonathan Huot authored
Conflicts: lib/libvmod_std/vmod.vcc lib/libvmod_std/vmod_std.c
-
- 24 Apr, 2017 2 commits
-
-
Dridi Boukelmoune authored
This should help diagnose #2253.
-
Poul-Henning Kamp authored
already failed. Fixes #2301
-
- 10 Apr, 2017 1 commit
-
-
Dridi Boukelmoune authored
Conflicts: configure.ac varnish.m4
-
- 07 Apr, 2017 1 commit
-
-
Pål Hermunn Johansen authored
-
- 06 Apr, 2017 11 commits
-
-
Martin Blix Grydeland authored
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
-
Dridi Boukelmoune authored
It shares a lot of parsing infrastructure with log records, but doesn't support the taglist syntax. Only a verbatim `vxid` token can be used as the LHS, and only integer operators are supported.
-
Pål Hermunn Johansen authored
There was a missing bracket that I fixed but forgot to amend the commit. Fixing with this one.
-
Reza Naghibi authored
By rounding Age down, we make sure Age < max-age while the object is fresh. Otherwise, we can prematurely get Age == max-age and Varnish will calculate that as a 0s TTL and create a pass scenario. Conflicts: bin/varnishd/cache/cache_req_fsm.c
-
Dag Haavi Finstad authored
This lets you set environment variables in a test case. Usage is setenv "FOO" "bar" which will set FOO=bar if FOO does not already exist. There is also an optional '-force' argument, which will overwrite the current variable if already set: setenv -force "FOO" "baz" The force argument corresponds to passing a non-zero third argument to setenv(3).
-
Dridi Boukelmoune authored
It uses `explicit_bzero` when available and falls back to `memset` otherwise. Conflicts: configure.ac
-
- 05 Apr, 2017 1 commit
-
-
Martin Blix Grydeland authored
VBE_Poll would not advance to test the next backend on the cooling list when the backend's n_conn is non-zero. This would create a spinlock effect, causing delays that could make the master kill the child because of CLI timeout. Fixes: #2295
-