Uh oh!
There was an error while loading. Please reload this page.
http: make sure end is always emitted with keepalive connections - #29263
http: make sure end is always emitted with keepalive connections #29263mcollina wants to merge 2 commits into
Conversation
This reverts commit 779a05d.
This test covers a regression where 'end' was not emitted in the case of keepalive requests without parsing the full body.
mcollina
commented
Aug 22, 2019
cc @indutny |
nodejs-github-bot
commented
Aug 22, 2019
lpinca
commented
Aug 22, 2019
#9668 should be reopened if/when this is merged. |
nodejs-github-bot
commented
Aug 23, 2019
Trott
commented
Aug 24, 2019
Landed in d5c3837...b3172f8 |
This test covers a regression where 'end' was not emitted in the case of keepalive requests without parsing the full body. PR-URL: #29263 Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
This is a bit strange to me. In |
addaleax
commented
Aug 24, 2019
@ronag The issue is that in (Aside, if there’s a regression with one of my PRs, feel free to @addaleax me because I’ve configured my inbox to give explicit mentions of my handledhigher priority over other messages.) |
This resolves a memory leak for keep-alive connections and does not regress in the way that 779a05d did by waiting for the incoming request to be finished before releasing the `parser.incoming` object. Refs: nodejs#28646 Refs: nodejs#29263Fixes: nodejs#9668
This test covers a regression where 'end' was not emitted in the case of keepalive requests without parsing the full body. PR-URL: #29263 Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Notable changes: This release fixes two regressions in the http module: * Fixes an event listener leak in the HTTP client. This resulted in lots of warnings during npm/yarn installs. #29245 * Fixes a regression preventing the `'end'` event from being emitted for keepalive requests in case the full body was not parsed. #29263 PR-URL: #29321
This resolves a memory leak for keep-alive connections and does not regress in the way that 779a05d did by waiting for the incoming request to be finished before releasing the `parser.incoming` object. Refs: #28646 Refs: #29263Fixes: #9668 PR-URL: #29297 Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
This resolves a memory leak for keep-alive connections and does not regress in the way that 779a05d did by waiting for the incoming request to be finished before releasing the `parser.incoming` object. Refs: #28646 Refs: #29263Fixes: #9668 PR-URL: #29297 Reviewed-By: Luigi Pinca <luigipinca@gmail.com> Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Notable changes: This release fixes two regressions in the http module: * Fixes an event listener leak in the HTTP client. This resulted in lots of warnings during npm/yarn installs. #29245 * Fixes a regression preventing the `'end'` event from being emitted for keepalive requests in case the full body was not parsed. #29263 PR-URL: #29321
Notable changes: This release fixes two regressions in the http module: * Fixes an event listener leak in the HTTP client. This resulted in lots of warnings during npm/yarn installs. #29245 * Fixes a regression preventing the `'end'` event from being emitted for keepalive requests in case the full body was not parsed. #29263 PR-URL: #29321
Unfortunately 779a05d introduced a regression in 12.8.0, making the autocannon test failing on Node 12. This PR reverts 779a05d and adds a test to ensure that we do not regress again.
Checklist
make -j4 test(UNIX), orvcbuild test(Windows) passes