- 19 Mar, 2013 6 commits
-
-
Martin Blix Grydeland authored
-
Martin Blix Grydeland authored
With fix for #1275 we can now do a proper 503 when malformed Vary-headers are received from backend. Error out instead of accepting bogus ':'. Conflicts: bin/varnishd/cache_vary.c
-
Martin Blix Grydeland authored
-
Martin Blix Grydeland authored
(out limit) characters long. Fixes: #1274 Test case by: Dag Haavi Finstad
-
Martin Blix Grydeland authored
Fixes: #1275 Test case by: Dag Haavi Finstad
-
Martin Blix Grydeland authored
headers, and do 503 on parse error. Conflicts: bin/varnishd/cache.h bin/varnishd/cache/cache_req_fsm.c bin/varnishd/cache_vary.c
-
- 12 Feb, 2013 1 commit
-
-
Tollef Fog Heen authored
-
- 06 Feb, 2013 1 commit
-
-
Martin Blix Grydeland authored
This should not have any impact on the "normal" deliveries, as these will as before be done through a single writev operation. For deliveries involving chunked encoding (e.g. ESI or gunzip), this should prevent an unnecessary delay where the Nagle algorithm kicks in on the end-chunk write. This should give higher throughput and better utilization of a single connection, especially for clients not doing pipelining. Conflicts: bin/varnishd/cache_acceptor.c
-
- 05 Feb, 2013 1 commit
-
-
Martin Blix Grydeland authored
Fixes: #1261
-
- 22 Jan, 2013 1 commit
-
-
Martin Blix Grydeland authored
This makes varnishncsa behave the same way as varnishlog, and prevents overly large memory allocations if it should see a bogus fd (shmlog overrun?). XXX: This will not fix varnishncsa for current trunk, as that is using XIDs instead of FDs breaking some functionality in the tools. This will be handled in the API rewrite.
-
- 03 Dec, 2012 1 commit
-
-
Tollef Fog Heen authored
Fixes #1233
-
- 22 Nov, 2012 4 commits
-
-
Lasse Karstensen authored
-
-
Lasse Karstensen authored
-
Lasse Karstensen authored
-
- 19 Nov, 2012 1 commit
-
-
Per Buer authored
-
- 16 Nov, 2012 1 commit
-
-
Lasse Karstensen authored
When looking at varnishlog it looks like it is send_timeout that was reached, but in fact it is idle_send_timeout. Clarify the language so the user understands which parameter to adjust.
-
- 05 Nov, 2012 1 commit
-
-
Martin Blix Grydeland authored
-
- 22 Oct, 2012 1 commit
-
-
Tollef Fog Heen authored
Use VSL_Dispatch instead of VSL_NextSLT in varnishtop, which does this automatically for us. Clean up accumulate() a bit in the process. Fixes #1213
-
- 18 Oct, 2012 1 commit
-
-
Tollef Fog Heen authored
-
- 09 Oct, 2012 1 commit
-
-
Dag Haavi Finstad authored
-
- 20 Sep, 2012 1 commit
-
-
Lasse Karstensen authored
This reverts commit 2587b1b9. 3.0 branch does not build with this commit.
-
- 18 Sep, 2012 1 commit
-
-
Tollef Fog Heen authored
Fixes #1194
-
- 22 Aug, 2012 2 commits
-
-
Dag Haavi Finstad authored
-
Dag Haavi Finstad authored
for ESI processing. Adapted from phk's fix for #1184.
-
- 20 Aug, 2012 2 commits
-
-
Tollef Fog Heen authored
-
Tollef Fog Heen authored
-
- 15 Aug, 2012 1 commit
-
-
Martin Blix Grydeland authored
wrk_herder_thread and wrk_herdtimer_thread causing bogus n_wrk_max stat increments. Fixes: #1183
-
- 08 Aug, 2012 1 commit
-
-
Andreas Plesner Jacobsen authored
-
- 05 Aug, 2012 1 commit
-
-
Andreas Plesner Jacobsen authored
-
- 27 Jul, 2012 1 commit
-
-
Andreas Plesner Jacobsen authored
-
- 20 Jul, 2012 1 commit
-
-
Lasse Karstensen authored
-
- 19 Jul, 2012 3 commits
-
-
Martin Blix Grydeland authored
If the ban lurker finds a busy object, skip the object and defer the checking to a later pass. No test case because it is very timing sensitive (but see the trac bug report for a VTC-file that mostly triggers it on the 3.0 branch only) Fixes: #1162
-
Martin Blix Grydeland authored
There is a short window of opportunity between HSH_Insert() and HSH_Unbusy(), where the locks are released and the expiry thread can try to expire an object that is still busy (for objects with very short TTL). This patch makes the expiry thread check for busy objects and wait if the head of the binheap is busy. No test case as the window is too small. Fixes: #1150
-
Martin Blix Grydeland authored
An attempt to nuke a busy object will end in assertion.
-
- 16 Jul, 2012 1 commit
-
-
- 13 Jul, 2012 1 commit
-
-
Dag Haavi Finstad authored
-
- 12 Jul, 2012 1 commit
-
-
Martin Blix Grydeland authored
objects. This avoids a memory race on the n_sess counter, which can lead to excessive session object allocation. Keeping the counters of in use and released separate allows the acceptor to continue to run lockless. Fixes: #897
-
- 05 Jul, 2012 1 commit
-
-
Dag Haavi Finstad authored
-
- 20 Jun, 2012 1 commit
-
-
Martin Blix Grydeland authored
Patch by: tnt
-