Investigate flaky test-net-error-twice #4057
Comments
|
Another one: |
Trott
added a commit
to Trott/io.js
that referenced
this issue
Nov 28, 2015
The test was not reliably creating the error event on Windows 2012. This makes the test more robust by delaying the start of the server write until nextTick, allowing the connection to close and cause the error more reliably on Windows 2012. Fixes: nodejs#4057
|
PR to hopefully fix this: #4062 |
Trott
added a commit
to Trott/io.js
that referenced
this issue
Nov 29, 2015
The test was not reliably creating the error event on Windows 2012. This makes the test more robust by using setImmediate() to delay the server writing to the socket to give time for socket.destroy() to take effect. Fixes: nodejs#4057
This was referenced Dec 5, 2015
|
This failure reproduces every time when using a fast machine with a single logical processor, only in Windows 2012 R2. This test was introduced in v0.10.26 and did not fail like this then. The problem started when v0.10 merged with v0.11, introducing this test there, meaning that v0.11 had the cause of this problem before. Using the test on previous revisions, the first bad commit is 2b6e078 . This does not help much, but is coherent with what @Trott observed in #4062 (comment) . Still investigating. |
mscdex
added a commit
to mscdex/io.js
that referenced
this issue
Dec 18, 2015
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: nodejs#4057
Fishrock123
pushed a commit
to Fishrock123/node
that referenced
this issue
Dec 22, 2015
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: nodejs#4057 PR-URL: nodejs#4342 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: João Reis <reis@janeasystems.com>
Fishrock123
pushed a commit
to Fishrock123/node
that referenced
this issue
Jan 6, 2016
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: nodejs#4057 PR-URL: nodejs#4342 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: João Reis <reis@janeasystems.com>
mscdex
added a commit
that referenced
this issue
Jan 7, 2016
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: #4057 PR-URL: #4342 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: João Reis <reis@janeasystems.com>
MylesBorins
pushed a commit
that referenced
this issue
Jan 19, 2016
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: #4057 PR-URL: #4342 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: João Reis <reis@janeasystems.com>
scovetta
pushed a commit
to scovetta/node
that referenced
this issue
Apr 2, 2016
On Windows there can exist some race condition where the notification of the client's `socket.destroy()` isn't received before the server writes to the socket. This race condition was more evident/reproducible on a single core system. This commit fixes the flakiness by waiting until the server's connection event handler has been called to destroy the client socket and perform the server socket write. Fixes: nodejs#4057 PR-URL: nodejs#4342 Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl> Reviewed-By: João Reis <reis@janeasystems.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Example failure:
The text was updated successfully, but these errors were encountered: