]>
granicus.if.org Git - pdns/log
bert hubert [Sun, 28 Feb 2016 19:00:23 +0000 (20:00 +0100)]
Merge pull request #3479 from nlyan/mtdocs
Add some comments to our use of Boost context
bert hubert [Sun, 28 Feb 2016 19:00:14 +0000 (20:00 +0100)]
Merge pull request #3472 from nlyan/mtcleanup
Improve compatibility for old compilers & libs in MTasker
bert hubert [Sun, 28 Feb 2016 18:56:41 +0000 (19:56 +0100)]
don't store empty records in the cache if we did get RRSIGs for a record we did not accept from an authoritative server
Andrew Nelless [Sun, 28 Feb 2016 13:37:23 +0000 (13:37 +0000)]
Add some comments to our use of Boost context
Andrew Nelless [Sun, 28 Feb 2016 12:26:26 +0000 (12:26 +0000)]
Nit for rubenk
bert hubert [Sun, 28 Feb 2016 12:01:38 +0000 (13:01 +0100)]
Merge pull request #3465 from tfarina/fix-3425
add dns_toupper() helper function
Andrew Nelless [Sun, 28 Feb 2016 11:18:16 +0000 (11:18 +0000)]
Add context impls to EXTRA_DIST
Andrew Nelless [Sun, 28 Feb 2016 01:23:54 +0000 (01:23 +0000)]
Fix cast parsing bug with GCC 4.7.
GCC 4.7 doesn't seem to like some_cast<::type>
Andrew Nelless [Sat, 27 Feb 2016 23:20:42 +0000 (23:20 +0000)]
Disable boost.m4's BOOST_CONTEXT tests for Boost <= 1.51
Andrew Nelless [Fri, 26 Feb 2016 14:50:37 +0000 (14:50 +0000)]
Add fallback to ucontext for Boost <= 1.51
bert hubert [Sun, 28 Feb 2016 11:01:56 +0000 (12:01 +0100)]
actually test against the right shipped tld names list if curl can't be found
bert hubert [Sun, 28 Feb 2016 09:57:36 +0000 (10:57 +0100)]
Merge pull request #3473 from nlyan/gcc47
Small fix to get recursor building on GCC 4.7 again
Peter van Dijk [Sun, 28 Feb 2016 07:09:25 +0000 (08:09 +0100)]
nit, thanks @ekohl
Andrew Nelless [Sun, 28 Feb 2016 01:28:02 +0000 (01:28 +0000)]
Fix missing definition UINT16_MAX under GCC 4.7.
std::numeric_limits is used elsewhere, so I went with that.
bert hubert [Sat, 27 Feb 2016 19:44:29 +0000 (20:44 +0100)]
spruce up dnsdist built-in website with cache hitrate percentage (in text & graph)
bert hubert [Sat, 27 Feb 2016 19:38:43 +0000 (20:38 +0100)]
turns out, our delta() output added dates with # comments. Lua uses -- comments. Duh. Plus stop logging empty lines in delta()
bert hubert [Sat, 27 Feb 2016 18:35:26 +0000 (19:35 +0100)]
close #3467: dnsdist cache hits not accounted properly in latency stats. Now count as '0'.
Thiago Farina [Fri, 26 Feb 2016 22:47:21 +0000 (19:47 -0300)]
add dns_toupper() helper function
This patch adds a ASCII version of toupper() function, called
dns_toupper(), and use it in place of toupper().
This fixes issue #3425
bert hubert [Fri, 26 Feb 2016 20:03:39 +0000 (21:03 +0100)]
Merge pull request #3346 from rgacogne/dnsdist-reuseport
dnsdist: Add reuseport support, maxOutstanding -> 10240, document perf tuning
bert hubert [Fri, 26 Feb 2016 20:02:24 +0000 (21:02 +0100)]
Merge pull request #3461 from rgacogne/dnsdist-issue-3460
dnsdist: Fix crash when calling rmServer() with an invalid index
bert hubert [Fri, 26 Feb 2016 17:12:21 +0000 (18:12 +0100)]
in dnsdist maintenance thread we'd lock up trying to lock g_luamutex twice
Remi Gacogne [Fri, 26 Feb 2016 16:16:28 +0000 (17:16 +0100)]
dnsdist: maxOustanding defaults to 10240. Add 'tuning' to README
bert hubert [Fri, 26 Feb 2016 15:14:09 +0000 (16:14 +0100)]
fix up recuweb password popup weirdness, now give nice warning on wrong password. Still open: do we need the api-key password?
Remi Gacogne [Mon, 8 Feb 2016 11:38:59 +0000 (12:38 +0100)]
dnsdist: Add an optional reuseport param to {add,set}Local()
If set to true, this parameter sets SO_REUSEPORT on platforms
supporting for this option, allowing multiple servers to
bind to the same port.
The same parameter is also added to addDNSCryptBind().
bert hubert [Fri, 26 Feb 2016 10:36:29 +0000 (11:36 +0100)]
Merge pull request #3427 from nlyan/mtcleanup
Massage System V context switching out of MTasker
Pieter Lexis [Fri, 26 Feb 2016 10:33:45 +0000 (11:33 +0100)]
Change a <span> to a <div> for better rendering
Remi Gacogne [Fri, 26 Feb 2016 10:17:29 +0000 (11:17 +0100)]
dnsdist: Fix crash when calling rmServer() with an invalid index
Fix #3460.
bert hubert [Fri, 26 Feb 2016 09:32:34 +0000 (10:32 +0100)]
make calidns send out precisely measured cache miss rates
Remi Gacogne [Fri, 26 Feb 2016 08:06:40 +0000 (09:06 +0100)]
Merge pull request #3457 from rgacogne/dnsdist-string-only-server
dnsdist: Fix str-only server not being added to the default pool
bert hubert [Thu, 25 Feb 2016 21:23:44 +0000 (22:23 +0100)]
Merge pull request #3450 from rgacogne/dnsdist-lock-cache
dnsdist: Prevent the cache ptr from being altered under our feet
bert hubert [Thu, 25 Feb 2016 21:23:19 +0000 (22:23 +0100)]
Merge pull request #3458 from rgacogne/dnsdist-cache-tests
dnsdist: Fix cache tests. Clean tests backends.
bert hubert [Thu, 25 Feb 2016 18:52:56 +0000 (19:52 +0100)]
Merge pull request #3451 from rgacogne/rec-replace-to-back
recursor: Move replaced cached entries to the back
Remi Gacogne [Thu, 25 Feb 2016 17:58:46 +0000 (18:58 +0100)]
dnsdist: Fix str-only server not being added to the default pool
As reported by @pieterlexis, server defined with the "string-only"
syntax were not correctly added to the default pool. This should
fix #3456.
In addition to that, this commit adds some Lua bindings for server
objects:
- member functions `getName()` and `getNameWithAddr()`
- member data `name`
bert hubert [Thu, 25 Feb 2016 17:34:39 +0000 (18:34 +0100)]
Merge pull request #3449 from rgacogne/dnsdist-pool-terminal
dnsdist: Make PoolAction() stop the rule processing again
bert hubert [Thu, 25 Feb 2016 17:34:24 +0000 (18:34 +0100)]
Merge pull request #3455 from ahupowerdns/comfun
DNSName escaped parsing speedups, comfun bulk zone/ns analysis tool, tiny dnsdist fixup
Remi Gacogne [Thu, 25 Feb 2016 16:51:24 +0000 (17:51 +0100)]
dnsdist: Fix cache tests. Clean tests backends.
Several issues:
- the cache tests used a vey small cache, not large enough for the
number of responses we expected to cache during the test suite
- this was hidden by the default response from the backends
matching what the test expected
- and by the backends not counting properly what looked like a
health check but wasn't.
Pieter Lexis [Thu, 25 Feb 2016 16:16:29 +0000 (17:16 +0100)]
Be a little more explicit on the addition of the SEP on CSKs
Pieter Lexis [Thu, 25 Feb 2016 14:50:02 +0000 (15:50 +0100)]
Merge pull request #3454 from pieterlexis/auth-4.0.0-alpha2-docs
Documentation updates for the upcoming alpha2 releases
Remi Gacogne [Thu, 25 Feb 2016 14:46:22 +0000 (15:46 +0100)]
dnsdist: Prevent the cache ptr from being altered under our feet
Make sure we hold the Lua mutex before getting the packet cache
shared_ptr, so that we don't have a thread reading it at the
exact same time it is altered by another.
We could have used atomic_load/atomic_store but libstdc++ uses
a pool of mutex for that anyway.
This might fix #3396.
Pieter Lexis [Thu, 25 Feb 2016 13:46:28 +0000 (14:46 +0100)]
Merge pull request #3453 from pieterlexis/silence-biowrite-warning
Silence a BIO_write warning
Pieter Lexis [Thu, 25 Feb 2016 12:26:04 +0000 (13:26 +0100)]
Recursor 4.0.0-alpha2 changelog
Pieter Lexis [Thu, 25 Feb 2016 11:30:57 +0000 (12:30 +0100)]
Documentation on Auth 4.0.0 alpha2
Pieter Lexis [Thu, 25 Feb 2016 11:48:48 +0000 (12:48 +0100)]
Silence a BIO_write warning
Pieter Lexis [Thu, 25 Feb 2016 12:27:52 +0000 (13:27 +0100)]
Merge pull request #3452 from pieterlexis/detect-network-libs
Detect several network libraries properly
bert hubert [Thu, 25 Feb 2016 12:03:29 +0000 (13:03 +0100)]
this commit uglifies DNSName escaped representation parsing for tremendous speedup (2x) during bulk zone loading from disk or database.
Part of the uglification is that we now special case unescaped names, which are the vast majority of cases.
Simultaneously, this moves us back to DNSName boost::container::string on non-Apple platforms, which delivered another 15% speedup on general operations
Finally, an additional unit test is added.
Pieter Lexis [Wed, 17 Feb 2016 19:25:57 +0000 (20:25 +0100)]
Detect several network libraries properly
Closes #3369
bert hubert [Wed, 24 Feb 2016 11:54:38 +0000 (12:54 +0100)]
set RD=1 in calidns for now, more right than what we do now..
bert hubert [Mon, 22 Feb 2016 14:27:42 +0000 (15:27 +0100)]
don't add NSEC/NSEC3 unless the packet asked for DNSSEC records
bert hubert [Mon, 15 Feb 2016 22:01:56 +0000 (23:01 +0100)]
gitignore comfun
bert hubert [Mon, 15 Feb 2016 20:20:13 +0000 (21:20 +0100)]
scanning zonefiles & polling nameservers
Pieter Lexis [Thu, 25 Feb 2016 10:09:35 +0000 (11:09 +0100)]
Merge pull request #3037 from pieterlexis/issue-3024-ENT-cleanups
Remove superfluous gsql queries and stop relying on schema defaults
bert hubert [Thu, 25 Feb 2016 09:51:20 +0000 (10:51 +0100)]
Merge pull request #3448 from zeha/recuconfig
Remove edns-subnet-whitelist whitelist pointing to powerdns.com
Pieter Lexis [Thu, 25 Feb 2016 09:08:21 +0000 (10:08 +0100)]
Merge pull request #3230 from zeha/websec
Auth/Recursor: drop JSONP, add web security headers
Pieter Lexis [Thu, 25 Feb 2016 09:08:15 +0000 (10:08 +0100)]
Merge pull request #3279 from zeha/bio-api-correctness
Handle return codes from OpenSSL BIO_*
Christian Hofstaedtler [Wed, 24 Feb 2016 22:02:25 +0000 (23:02 +0100)]
Remove edns-subnet-whitelist whitelist pointing to powerdns.com
Christian Hofstaedtler [Wed, 24 Feb 2016 21:06:29 +0000 (22:06 +0100)]
Handle return codes from OpenSSL BIO_*
bert hubert [Wed, 24 Feb 2016 20:41:34 +0000 (21:41 +0100)]
Merge pull request #3443 from ahupowerdns/speedups
Speedups! Closes some big impediments for performance & even correctness
bert hubert [Wed, 24 Feb 2016 20:40:41 +0000 (21:40 +0100)]
Merge pull request #3442 from pieterlexis/recursor-ixfr-catch-exception
recursor: catch exception during IXFR of RPZ
bert hubert [Wed, 24 Feb 2016 20:40:25 +0000 (21:40 +0100)]
Merge pull request #3444 from pieterlexis/recursor-ixfr-tsig-algo-names
recursor: set the TSIG algoname on IXFR correctly
bert hubert [Wed, 24 Feb 2016 20:39:50 +0000 (21:39 +0100)]
Merge pull request #3445 from pieterlexis/recursor-rpz-docs
Document recursor RPZ options
Pieter Lexis [Wed, 24 Feb 2016 17:48:43 +0000 (18:48 +0100)]
Document recursor RPZ options
Pieter Lexis [Wed, 24 Feb 2016 17:14:31 +0000 (18:14 +0100)]
recursor: set the TSIG algoname on IXFR correctly
Pieter Lexis [Wed, 24 Feb 2016 16:35:22 +0000 (17:35 +0100)]
recursor: catch exception during IXFR of RPZ
This would lead to a SIGABRT before when the IXFR failed.
Remi Gacogne [Wed, 24 Feb 2016 16:13:14 +0000 (17:13 +0100)]
recursor: Move replaced cached entries to the back
When we replace an existing entry, it keeps its existing place in
the expunge queue, while new entries are inserted to the back and
hits are moved to the back.
Moving replaced entries to the back of the queue is more fair and
so probably more efficient, as it would increase the likelihood of
expunging unused expired entries from the cache.
Remi Gacogne [Wed, 24 Feb 2016 15:46:14 +0000 (16:46 +0100)]
dnsdist: Make PoolAction() stop the rule processing again
It could clearly be confusing, and can easily be done by
ordering rules correctly anyway.
Thanks @rygl for the very valuable feedback.
bert hubert [Wed, 24 Feb 2016 15:36:44 +0000 (16:36 +0100)]
we were inconsistent in comparing ComboAddresses with sin_family==0. Removed possibility for inconsistency, plus surrounded this bug with unit tests. Added similar test for DNSName.
Andrew Nelless [Wed, 24 Feb 2016 15:10:22 +0000 (15:10 +0000)]
Switch from std::function to boost::function
GNUs implementation of std::function can't eat 3 pointers without allocating,
(24 bytes) whereas both LLVM libc++ and boost::function can. boost::function
has been move enabled since 1.52 (Nov 2012).
Remi Gacogne [Wed, 24 Feb 2016 15:01:55 +0000 (16:01 +0100)]
Merge pull request #3432 from rgacogne/dnsdist-split-tests
dnsdist: Split Caching, Routing and Spoofing from "Advanced" tests
Pieter Lexis [Wed, 17 Feb 2016 15:24:41 +0000 (16:24 +0100)]
Stop relying on SQL schema defaults
Pieter Lexis [Wed, 10 Feb 2016 14:35:10 +0000 (15:35 +0100)]
Deprecate the insert-slave-query
Pieter Lexis [Wed, 10 Feb 2016 11:20:56 +0000 (12:20 +0100)]
Deprecate the insert-record-order-query
Pieter Lexis [Mon, 14 Dec 2015 11:34:43 +0000 (12:34 +0100)]
Remove superfluous gsql query for ENTs
This fixes part of #3024
Pieter Lexis [Wed, 24 Feb 2016 14:49:05 +0000 (15:49 +0100)]
Merge pull request #3429 from mind04/failed
make sure we don't forget passing test marked as failing
bert hubert [Wed, 24 Feb 2016 13:59:30 +0000 (14:59 +0100)]
Merge pull request #3435 from Habbie/lua-dnsname
add two methods to DNSName for recursor Lua scripting
bert hubert [Wed, 24 Feb 2016 13:59:07 +0000 (14:59 +0100)]
Merge pull request #3433 from rgacogne/dnsdist-check-response
dnsdist: Check that the answer matches the initial query over UDP
bert hubert [Wed, 24 Feb 2016 13:58:22 +0000 (14:58 +0100)]
Merge pull request #3437 from Habbie/lua-pdnsunixtime
add pdns.now timeval struct to recursor lua
bert hubert [Wed, 24 Feb 2016 13:58:10 +0000 (14:58 +0100)]
Merge pull request #3439 from Habbie/lua15minutes
link to learn lua in 15 minutes
bert hubert [Wed, 24 Feb 2016 10:55:36 +0000 (11:55 +0100)]
lots of speedups local to zoneparser-tng, mostly reducing malloc load and locale use
bert hubert [Wed, 24 Feb 2016 11:55:38 +0000 (12:55 +0100)]
add "reuseport" to recursor as a switch, off by default
bert hubert [Mon, 22 Feb 2016 14:26:55 +0000 (15:26 +0100)]
we used toString() to compare incoming packets - showed up in profile at 3% or so
bert hubert [Wed, 24 Feb 2016 10:54:22 +0000 (11:54 +0100)]
we don't need set close on exec in pdns_recursor since it won't (can't) exec anything
bert hubert [Wed, 24 Feb 2016 11:41:08 +0000 (12:41 +0100)]
because unset/empty netmasks did not compare as equal, we would fill the cache with tens of thousands of duplicate entries per name
Peter van Dijk [Tue, 23 Feb 2016 16:18:52 +0000 (17:18 +0100)]
add pdns.now timeval struct to recursor lua
Peter van Dijk [Wed, 24 Feb 2016 11:32:41 +0000 (12:32 +0100)]
link to learn lua in 15 minutes
bert hubert [Wed, 24 Feb 2016 10:24:59 +0000 (11:24 +0100)]
Merge pull request #3434 from Habbie/remove-old-lua-recursor
remove v3 lua recursor files (dead code)
Peter van Dijk [Wed, 24 Feb 2016 09:46:16 +0000 (10:46 +0100)]
Merge pull request #3428 from zeha/apirrsets
API: Fix zone/records design mistake
Andrew Nelless [Tue, 23 Feb 2016 15:15:56 +0000 (15:15 +0000)]
Hack in compatibility for Boost <= 1.55
Peter van Dijk [Tue, 23 Feb 2016 15:31:29 +0000 (16:31 +0100)]
add two methods to DNSName for recursor Lua scripting
Peter van Dijk [Tue, 23 Feb 2016 15:03:14 +0000 (16:03 +0100)]
remove v3 lua recursor files (dead code)
Remi Gacogne [Tue, 23 Feb 2016 08:02:50 +0000 (09:02 +0100)]
dnsdist: Split Caching, Routing and Spoofing from "Advanced" tests
Andrew Nelless [Mon, 22 Feb 2016 16:49:24 +0000 (16:49 +0000)]
Initial stab at Boost context backend for MTasker
bert hubert [Mon, 22 Feb 2016 18:51:49 +0000 (19:51 +0100)]
Merge pull request #3423 from Habbie/tinydns-yak
tinydns testing yaks
bert hubert [Mon, 22 Feb 2016 18:31:58 +0000 (19:31 +0100)]
Merge pull request #3422 from ahupowerdns/multispoof
Multispoof: make the dnsdist spoofing actions accept multiple IPv4 and IPv6 addresses, retaining old syntax + updated regression tests
Remi Gacogne [Mon, 22 Feb 2016 18:22:55 +0000 (19:22 +0100)]
dnsdist: Check that the answer matches the initial query over UDP
If we wrap around our maxOutstanding counter too fast, we need
to check that the answer we get is for the right query.
In order to do that, we now parse the question section in the
response and compare it to the one we expect (type, class and
name).
Peter van Dijk [Mon, 22 Feb 2016 11:05:58 +0000 (12:05 +0100)]
update incbin to upstream 3d4aa9, fixing osx builds
Kees Monshouwer [Mon, 22 Feb 2016 08:11:53 +0000 (09:11 +0100)]
s/failing/fail
Kees Monshouwer [Sun, 21 Feb 2016 23:14:57 +0000 (00:14 +0100)]
remove 'failing' flag for passing oracle tests
Kees Monshouwer [Sun, 21 Feb 2016 22:32:49 +0000 (23:32 +0100)]
remove 'failing' flag for passing bind tests
Kees Monshouwer [Sun, 21 Feb 2016 22:30:49 +0000 (23:30 +0100)]
remove 'failing' flag for passing ldap tests
Kees Monshouwer [Sun, 21 Feb 2016 09:46:58 +0000 (10:46 +0100)]
add 'fail.*' option to the regression-test
A failing test must be brokem.