update nginx #2
Merged
update nginx #2
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ve 8 bytes for every key-value pair on 64-bit systems, for example.
Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…ly call malloc_trim(1) every N requests when malloc_trim() is available. By default, "lua_malloc_trim 1000" is configured. This should fix the glibc oddity of holding too much freed memory when it fails to use brk() to allocate memory in the data segment.
…m was first introduced in ngx_lua 0.10.7.
…, send_timeout, read_timeout). Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
… their own shm-based data structures for the Lua land usage (that is, to create custom siblings to lua_shared_dict). thanks helloyi and doujiang for the patches. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…mizations in shdict storage layout.
… test modes. now we use MOCKEAGAIN_WRITE_TIMEOUT_PATTERN feature instead of sending mega bytes of data. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…ow only allowed in the http {} context.
Use of these session hooks in the server {} scope did not make much
sense since server name dispatch happens *after* ssl session resumption.
Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
… requests to avoid segfaults. When setting a multi-value header during bad requests, the value isn't added into the header's array. Despite failing to be set, the size of the array is still incremented. Later, if Nginx attempts to iterate over the array, a segfault will occur. e21d9b5 established a pattern of silently ignoring the header being set. I've moved the check for a bad request into `ngx_http_lua_set_input_header` so that setting a header value is ignored for both types of headers. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
… unsafe URIs (#905) Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…worker_by_lua* contexts. thanks kraml for the report in #876. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…ementation of lua-resty-core.
…ct when lua_code_cache is off. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…iang Zhu. also removed the copyright notice in this module since it is already granted.
…opyright is already granted.
…c with msec precision through the decimal part) for each error log entry. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…rrent request with arbitrary HTTP response status codes via ngx.exit(). Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
now it is now consistent with what Google Chrome's JavaScript API function encodeURIComponent() does. See: https://tools.ietf.org/html/rfc2396. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…listen on the same port or unix domain socket file path *and* some of them are using ssl_certificate_by_lua* configurations while some are not. thanks petrovich-ua for the report and original patch in #1055.
…l host name mismatch.
…n linux i386 and x64.
…ons fail. as discussed in #1098. Signed-off-by: skyever <[email protected]> Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
… in the "check leak" test mode since there is a known memory leak in this case.
…d. thanks Kyra Zimmer for the original patch.
…ssage "semaphore gc wait queue is not empty" might be logged. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…ccessible on the Lua land. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…lua_socket_read_handler.
…ns after the downstream cosocket object is closed. thanks Dejiang Zhu for the report.
Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
ngx.escape_uri should also escape RFC 3986 section 2.2 Reserved Characters. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…1124" Commit f170505 breaks the compatibility with RFC 3986. Here is two reasons: 1. Quote from RFC 3986 Section 2.2: > A subset of the reserved characters (gen-delims) is used as delimiters of the generic URI components described in Section 3 Note that RFC 3986 says 'a subset of the reserved characters (gen-delims)', not all the reserved characters. The characters escaped in that commit are 'sub-delims'. They are not required to be escaped according to Section 2.2. 2. Refer to RFC 3986 "Appendix A. Collected ABNF for URI", sub-delims could be used as part of query and other components. This use case shows that sub-delims are valid in some component of URI. Therefore, it would be better if we don't escape them for URI component. Signed-off-by: Yichun Zhang (agentzh) <[email protected]>
…eaered in 0.10.9.
splitice
pushed a commit
that referenced
this pull request
Apr 1, 2022
==openresty==70603==ERROR: AddressSanitizer: memcpy-param-overlap: memory ranges [0x621000001500,0x621000002181) and [0x62100000187f, 0x621000002500) overlap
#0 0x7f3db1899ffe (/lib64/libasan.so.5+0x99ffe)
#1 0x9da926 (/usr/local/openresty-debug/nginx/sbin/nginx+0x9da926)
#2 0x9dd1a1 (/usr/local/openresty-debug/nginx/sbin/nginx+0x9dd1a1)
openresty#3 0x4c89c6 (/usr/local/openresty-debug/nginx/sbin/nginx+0x4c89c6)
openresty#4 0x5d1e4e (/usr/local/openresty-debug/nginx/sbin/nginx+0x5d1e4e)
openresty#5 0x4c89c6 (/usr/local/openresty-debug/nginx/sbin/nginx+0x4c89c6)
openresty#6 0x5b8583 (/usr/local/openresty-debug/nginx/sbin/nginx+0x5b8583)
openresty#7 0x4c89c6 (/usr/local/openresty-debug/nginx/sbin/nginx+0x4c89c6)
openresty#8 0x4b4419 (/usr/local/openresty-debug/nginx/sbin/nginx+0x4b4419)
openresty#9 0x427f16 (/usr/local/openresty-debug/nginx/sbin/nginx+0x427f16)
openresty#10 0x7f3daff27554 (/lib64/libc.so.6+0x22554)
openresty#11 0x42d537 (/usr/local/openresty-debug/nginx/sbin/nginx+0x42d537)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.