summaryrefslogtreecommitdiffhomepage
AgeCommit message (Collapse)AuthorFilesLines
2016-02-19SSL: fixed SSL_shutdown() comment.Maxim Dounin1-1/+1
2016-02-18Headers filter: fixed "add_header ... '' always".Ruslan Ermilov1-8/+8
The "always" parameter was ignored if the header value was empty.
2016-02-19Configure: skip building OpenSSL documentation to conserve time.Sergey Kandaurov1-1/+1
The install_sw target first appeared in OpenSSL 0.9.7e and is documented since OpenSSL 1.0.0 as the way to install the OpenSSL software without documentation.
2016-02-18Configure: fixed static nginx build with OpenSSL (ticket #903).Ruslan Ermilov2-7/+9
Before 7142b04337d6, it was possible to build the OpenSSL library along with nginx, and link nginx statically with this library (--with-openssl=DIR --with-ld-opt=-static --with-http_ssl_module). This was broken on Linux by not adding -ldl after -lcrypto. The fix also makes it possible to link nginx statically on Linux with the system OpenSSL library, which never worked before.
2016-02-18Core: added support for more than 64 CPUs in worker_cpu_affinity.Vladimir Homutov6-59/+83
2016-02-16HTTP/2: added debug logging of response headers.Valentin Bartenev1-0/+41
Because of HPACK compression it's hard to see what headers are actually sent by the server.
2016-02-16HTTP/2: use local pointer instead of r->connection.Valentin Bartenev1-2/+2
No functional changes.
2016-02-15Core: improved logging on invalid NGINX variable (ticket #900).Maxim Dounin1-0/+6
2016-02-13Updated OpenSSL used for win32 builds.Maxim Dounin1-1/+1
2016-02-13Win32: simplified and improved handling of MSVC versions.Maxim Dounin10-49/+41
Now we always set NGX_CC_NAME to "msvc", and additionally test compiler version as reported by "cl" in auto/cc/msvc (the same version is also available via the _MSC_VER define). In particular, this approach allows to properly check for C99 variadic macros support, which previously was not used with MSVC versions not explicitly recognized. Now unneeded wildcards in NGX_CC_NAME tests for msvc removed accordingly, as well as unused wildcards for owc and icc.
2016-02-12HTTP/2: fixed undefined behavior in ngx_http_v2_huff_encode().Valentin Bartenev1-2/+6
When the "pending" value is zero, the "buf" will be right shifted by the width of its type, which results in undefined behavior. Found by Coverity (CID 1352150).
2016-02-11HTTP/2: implemented HPACK Huffman encoding for response headers.Valentin Bartenev4-56/+353
This reduces the size of headers by over 30% on average. Based on the patch by Vlad Krasnov: http://mailman.nginx.org/pipermail/nginx-devel/2015-December/007682.html
2016-02-11Dynamic modules: fixed a version mismatch message (ticket #898).Ruslan Ermilov1-1/+1
Based on a patch by Takashi Takizawa.
2016-02-11Stream: initialize variable right before using it.Roman Arutyunyan1-2/+2
2016-02-11Stream: removed useless typedef.Roman Arutyunyan1-3/+0
2016-02-10Dynamic modules: added auth_basic module libs via ngx_module_libs=.Ruslan Ermilov1-2/+1
2016-02-09Core: ngx_module_t compatibility with C++.Piotr Sikora1-1/+1
Changes to NGX_MODULE_V1 and ngx_module_t in 85dea406e18f (1.9.11) broke all modules written in C++, because ISO C++11 does not allow conversion from string literal to char *. Signed-off-by: Piotr Sikora <piotrsikora@google.com>
2016-02-10Version bump.Maxim Dounin1-2/+2
2016-02-09release-1.9.11 tagMaxim Dounin1-0/+1
2016-02-09nginx-1.9.11-RELEASErelease-1.9.11Maxim Dounin1-0/+43
2016-02-08Dynamic modules: NGX_STREAM define removed.Maxim Dounin1-2/+0
It's not used and contradicts the idea of dynamic loading.
2016-02-05Dynamic modules: added dlsym() to configure test.Valentin Bartenev1-1/+1
It appears that AddressSanitizer in GCC 5 intercepts dlopen() and dlclose() but not dlsym().
2016-02-05Dynamic modules: fixed a copy-n-paste error.Ruslan Ermilov1-1/+1
2016-02-04Dynamic modules.Maxim Dounin21-34/+1136
The auto/module script is extended to understand ngx_module_link=DYNAMIC. When set, it links the module as a shared object rather than statically into nginx binary. The module can later be loaded using the "load_module" directive. New auto/module parameter ngx_module_order allows to define module loading order in complex cases. By default the order is set based on ngx_module_type. 3rd party modules can be compiled dynamically using the --add-dynamic-module configure option, which will preset ngx_module_link to "DYNAMIC" before calling the module config script. Win32 support is rudimentary, and only works when using MinGW gcc (which is able to handle exports/imports automatically). In collaboration with Ruslan Ermilov.
2016-02-04Dynamic modules: auto/module script.Maxim Dounin4-558/+848
This script simplifies configuration of additional modules, including 3rd party ones. The script is extensible, and will be used to introduce dynamic linking of modules in upcoming changes. 3rd party module config scripts are called with ngx_module_link preset to "ADDON" - this allows config scripts to call auto/module without ngx_module_link explicitly defined, as well as testing if new interface is in place if compatibility with older nginx versions is desired. In collaboration with Ruslan Ermilov.
2016-02-04Dynamic modules: introduced HTTP_INIT_FILTER_MODULES.Maxim Dounin1-8/+10
Additionally, HTTP_HEADERS_FILTER_MODULE now added to HTTP_FILTER_MODULES. This avoids explict use of modules at the later stages, now only module lists are used. This will be needed in later patches.
2016-02-04Dynamic modules: dlopen() support.Maxim Dounin13-4/+134
2016-02-04Dynamic modules: changed ngx_modules to cycle->modules.Maxim Dounin16-140/+148
2016-02-04Dynamic modules: moved module-related stuff to separate files.Maxim Dounin11-95/+151
2016-02-04HTTP/2: fixed possible buffer overrun (ticket #893).Valentin Bartenev1-2/+2
Due to greater priority of the unary plus operator over the ternary operator the expression didn't work as expected. That might result in one byte less allocation than needed for the HEADERS frame buffer.
2016-02-02HTTP/2: fixed padding handling in HEADERS frame with CONTINUATION.Valentin Bartenev1-1/+11
2016-02-02HTTP/2: fixed request length accounting.Valentin Bartenev1-1/+10
Now it includes not only the received body size, but the size of headers block as well.
2016-02-02HTTP/2: fixed excessive memory allocation for pool cleanup.Valentin Bartenev1-1/+1
2016-02-02HTTP/2: removed unused field from ngx_http_v2_stream_t.Valentin Bartenev1-1/+0
2015-12-17Resolver: style.Ruslan Ermilov1-9/+8
Use the original query name in error and debug messages when processing PTR responses.
2015-12-17Resolver: improved PTR response processing.Ruslan Ermilov1-28/+74
The previous code only parsed the first answer, without checking its type, and required a compressed RR name. The new code checks the RR type, supports responses with multiple answers, and doesn't require the RR name to be compressed. This has a side effect in limited support of CNAME. If a response includes both CNAME and PTR RRs, like when recursion is enabled on the server, PTR RR is handled. Full CNAME support in PTR response is not implemented in this change.
2015-12-17Resolver: style.Ruslan Ermilov1-13/+13
Renamed argument in ngx_resolver_process_a() for consistency.
2016-02-02Resolver: free TCP buffers on resolver cleanup.Roman Arutyunyan1-0/+10
2016-02-02Resolver: fixed possible resource leak introduced in 5a16d40c63de.Ruslan Ermilov1-0/+2
Found by Coverity (CID 1351175).
2016-01-28Resolver: TCP support.Roman Arutyunyan2-51/+549
Resend DNS query over TCP once UDP response came truncated.
2016-01-28Resolver: per-request DNS server balancer.Roman Arutyunyan2-5/+17
Previously, a global server balancer was used to assign the next DNS server to send a query to. That could lead to a non-uniform distribution of servers per request. A request could be assigned to the same dead server several times in a row and wait longer for a valid server or even time out without being processed. Now each query is sent to all servers sequentially in a circle until a response is received or timeout expires. Initial server for each request is still globally balanced.
2016-01-28Resolver: renamed UDP-specific structures, fields and variables.Roman Arutyunyan2-63/+63
They will be used for TCP connections as well.
2016-01-28Resolver: removed unused field from ngx_resolver_ctx_t.Roman Arutyunyan1-3/+1
2016-01-28Version bump.Roman Arutyunyan1-2/+2
2016-01-26release-1.9.10 tagMaxim Dounin1-0/+1
2016-01-26nginx-1.9.10-RELEASErelease-1.9.10Maxim Dounin1-0/+103
2016-01-26Resolver: limited CNAME recursion.Ruslan Ermilov1-6/+22
Previously, the recursion was only limited for cached responses.
2016-01-26Resolver: fixed use-after-free memory accesses with CNAME.Roman Arutyunyan2-39/+35
When several requests were waiting for a response, then after getting a CNAME response only the last request's context had the name updated. Contexts of other requests had the wrong name. This name was used by ngx_resolve_name_done() to find the node to remove the request context from. When the name was wrong, the request could not be properly cancelled, its context was freed but stayed linked to the node's waiting list. This happened e.g. when the first request was aborted or timed out before the resolving completed. When it completed, this triggered a use-after-free memory access by calling ctx->handler of already freed request context. The bug manifests itself by "could not cancel <name> resolving" alerts in error_log. When a request was responded with a CNAME, the request context kept the pointer to the original node's rn->u.cname. If the original node expired before the resolving timed out or completed with an error, this would trigger a use-after-free memory access via ctx->name in ctx->handler(). The fix is to keep ctx->name unmodified. The name from context is no longer used by ngx_resolve_name_done(). Instead, we now keep the pointer to resolver node to which this request is linked. Keeping the original name intact also improves logging.
2016-01-26Resolver: changed the ngx_resolver_create_*_query() arguments.Roman Arutyunyan1-30/+27
No functional changes. This is needed by the following change.
2016-01-26Resolver: fixed CNAME processing for several requests.Ruslan Ermilov1-6/+15
When several requests were waiting for a response, then after getting a CNAME response only the last request was properly processed, while others were left waiting.
generated by cgit (git 2.34.1) at 2025-12-30 12:45:00 +0000