forked from apache/trafficserver
-
Notifications
You must be signed in to change notification settings - Fork 2
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Hostdb restructure merged with master #32
Merged
SolidWallOfCode
merged 92 commits into
SolidWallOfCode:master
from
bneradt:hostdb-restructure-merged-with-master
Apr 11, 2022
Merged
Hostdb restructure merged with master #32
SolidWallOfCode
merged 92 commits into
SolidWallOfCode:master
from
bneradt:hostdb-restructure-merged-with-master
Apr 11, 2022
Conversation
This file contains 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
Without this, large production strategies.yaml files with thousands of remaps take hours to load, because each remap independently loads and parses the strategies.yaml file. This makes the plugin reuse the parsed object if multiple remaps use the same file. On reload, files are loaded again.
Reverts the behavior of getOriginSession() for a timed out session to the previous behavior, where it returns a null session pointer.
Co-authored-by: Chris McFarlen <cmcfarlen@apple.com>
ats_free already does null checks
…f none (apache#8655) * money_trace: add global support, custom header name, create if none. This enhances the money_trace to allow it to run globally, with the ability to override the global with a remap plugin. Adds in better logging support for the current transaction's trace header. Allows passthru mode which passes along the client trace header without modification making ATS transparent. * money_trace: change logic for how an invalid incoming header is handled * money_trace autest: remove commented code
…interpret (apache#8742) the YAML elements in the host protocol in both core and the parent_select plugin.
…che#8710) Co-authored-by: Walt Karas <wkaras@yahooinc.com>
Sphinx has fixed the following issue that was blocking us from moving to Sphinx 4.x: sphinx-doc/sphinx#9322 With that fixed, we can unpin from 3.x, which this patch does. Fixes: apache#7941
also remove link to TS timeouts in ocsp settings; unrelated to TS timeouts
This reverts commit 38bf663.
Proxy Verifier version 2.3.1 actively closes the connection when the proxy (ATS) returns a response with a "Connection: close" header. This improves the stability of the cache-request-method AuTest, wherein ATS, dependent upon packet timing, would sometimes return a "Connection: close" and follow it with a TCP FIN. The FIN would then be processed by the Verifier client after request headers were sent, which would interrupt the traffic flow and cause the test to fail. With Verifier version 2.3.1, this is avoided by the client actively closing the connection when ATS returns the intermittent "Connection: close".
This is a second attempt to upgrade to Sphinx 4.x. The first was apache#8750. It was noticed, however, that the new Sphinx didn't automatically deliver the theme.css file. This updates the conf.py file to make the addition of theme.css explicit.
These buffer size computation macros are meant to be used for callers of ats_base64_encode/decode. They were not taking into account a null terminator, which is always written by those functions. This causes a buffer overflow when encode or decode are called on buffers of certain sizes, e.g. decode on an input buffer of size 4. This change makes these buffer size calculations match the functions. It also updates unit tests for the access_control plugin, which uses these functions. Since this is a macro change, plugins which use encode/decode will need to be recompiled.
The AuTests using httpbin all started failing today when new AuTest pipenv environments started using the 2.1.0 version of Werkzeug. This new Werkzeug release removes its BaseResponse API because it has been deprecated for a few releases. httpbin has a dependency upon BaseResponse, however, and thus it currently fails with this latest Werkzeug release. This patch temporarily pins Werkzeug until httpbin is updated so that our httpbin AuTests can run.
Fix a use-after-free error when loading SSL cert or key fails.
Co-authored-by: Chris McFarlen <cmcfarlen@apple.com>
The latest jinja2 breaks Sphinx 3.x builds. For now, pinning jinja2 to an earlier release for doc builds.
This reverts commit 4dbf914. In the previous attempt to use Sphinx 4.x, I tried to fix our CSS 4.x rendering by explicitly adding theme.css to the css_files html_context. That did indeed add the needed theme.css rendering to the various docs, but there are still some artifacts that don't render correctly (such as ts:cv parameters). Reverting 4.x for now since 3.x works fine and moving to 4.x is turning out to be a bigger project than I expected.
This fix adds reference counting for the Lua plugin remap instance handles. The reference counting allows us to eliminate an existing memory leak of the instance handles. In addition, this means that the old Lua memory allocated by LuaJIT may also be freed via LuaJIT garbage collection. This fix also adds the '--ljgc' remap instance plugin parameter to the Lua plugin. This paramter enables on-demand LuaJIT garbage collection while the remap instances are created and deleted. This is useful when operating close to the LuaJIT memory limit, which is currently 2GB on Linux using LuaJIT v2.1.0-beta3 from 2017. Fixes apache#8728
Co-authored-by: Chris McFarlen <cmcfarlen@apple.com>
Co-authored-by: Fei Deng <feid@yahooinc.com>
Co-authored-by: Serris Lew <lserris@apple.com>
…ections (apache#8784) * Propagate proxy.config.net.sock_option_flag_in to newly accepted connections * For client connections, disable setting SO_LINGER by default This was originally approved under apache#8463 and was reverted in apache#8622
Conflicts: iocore/hostdb/HostDB.cc iocore/hostdb/I_HostDBProcessor.h plugins/lua/ts_lua_misc.c proxy/http/HttpSM.cc proxy/http/HttpTransact.cc proxy/http/remap/unit-tests/nexthop_test_stubs.cc tests/gold_tests/next_hop/strategies_ch2/strategies_ch2.test.py tests/gold_tests/tls/tls_verify_override_base.test.py
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.
This contains a merge commit from latest master onto the top of SolidWallOfCode:hostdb-restructure. This preserves the merge metadata as well as the comits, but let me know whether you prefer this merge commit to be a squashed commit on top of your branch.