Currently, benchmark code is not exercised at all in CI. This adds a
minimal test for net benchmarks. If this is deemed acceptable, similar
minimal tests for other benchmarks can be written. Additionally, as
issues and edge cases are uncovered, checks for them can be added.
PR-URL: https://github.com/nodejs/node/pull/11979
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Currently when node is build --without-ssl and the test are run,
there are a number of failing test due to tests expecting crypto
support to be available. This commit fixes fixes the failure and
instead skips the tests that expect crypto to be available.
PR-URL: https://github.com/nodejs/node/pull/11631
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
The intention of test case is to make sure that `timeout` property is honored
and the code in context terminates and throws correct exception. However in
test case, the code inside context would complete before `timeout` for windows
and would sometimes fail. Updated the code so it guarantee to not complete
execution until timeout is triggered.
Fixes: https://github.com/nodejs/node/issues/11261
PR-URL: https://github.com/nodejs/node/pull/11530
Reviewed-By: James M Snell <jasnell.gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Josh Gavant <josh.gavant@outlook.com>
Even after being moved to `sequential` in
1ce05ad540, `test-regress-GH-897` still
was occasionally flaky on Raspberry Pi devices on CI.
The test is especially sensitive to resource constraints. It failed
reliably on my laptop if I moved it to `parallel` and ran 32 competing
node test processes. Even for a flaky test, that's unusually low. I
typically don't see problems, even for flaky tests, until I get up to
around four times that number.
On a Raspberry Pi, of course, that sensitivity to resource constraints
will manifest much sooner.
This change checks the order of timers firing, rather than the duration
before a timer is fired. This eliminates the sensitivity to resource
constraints. The test can now be moved back to `parallel`. I am able to
run many copies of the test simultaneously without seeing test failures.
PR-URL: https://github.com/nodejs/node/pull/10903
Fixes: https://github.com/nodejs/node/issues/10073
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
test-http-client-timeout-with-data has failed here and there in CI on
FreeBSD and OS X. The test has a socket timeout set to 50ms and a timer
set for 100ms. However, they are not necessarily set in the same tick of
the event loop and their ordering is therefore not guaranteed.
Instead of using a timer, this change listens for an event on the
listener to know when the socket timeout has occurred and then runs the
code originally in the timer.
Additional refactoring: Replaced `process.on('exit', ...)` checks with
`common.mustCall()` and replaced usage of `assert.equal()` with
`assert.strictEqual()`.
PR-URL: https://github.com/nodejs/node/pull/10431
Reviewed-By: James M Snell <jasnell@gmail.com>
* use const and let instead of var
* use common.mustCall to control functions executions
* use assert.strictEqual instead of assert.equal
* use assert.ifError to handle errors
* use arrow functions
* remove console.log and process.stdout.write
PR-URL: https://github.com/nodejs/node/pull/10452
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Brian White <mscdex@mscdex.net>
test-timers-same-timeout-wrong-list-deleted was flaky under load because
there is no guarantee that a timer will fire within a given period of
time. It had an exit handler that checked that the process was finishing
in less than twice as much as a timer was set for. Under load, the
timer could take over 200ms to fire even if it was set for 100ms, so
this was causing the test to be flaky on CI from time to time.
However, that timing check is unnecessary to identify the regression
that the test was written for. When run with a version of Node.js that
does not contain the fix that accompanied the test in its initial
commit, an assertion indicating that there were still timers in the
active timer list fired. So, this commit removes the exit handler timing
check and relies on the existing robust active timers list length check.
This allows us to move the test back to parallel because it does not
seem to fail under load anymore.
The test was refactored slightly, removing duplicated code to a
function, using `assert.strictEqual()` instead of `assert.equal()`,
changing a 10ms timer to 1ms, and improving the messages provided by
assertions.
Fixes: https://github.com/nodejs/node/issues/8459
PR-URL: https://github.com/nodejs/node/pull/10362
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
test-buffer-creation-regression is flaky on some SmartOS hosts in CI,
timing out. Move to sequential so it does not compete with other tests
for resources. Reduce three test cases to just the one needed to
identify the regression.
PR-URL: https://github.com/nodejs/node/pull/10161
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Italo A. Casas <me@italoacasas.com>
`test-regress-GH-897` is dependent on a timer firing within a period of
time. Especially on some of the FreeBSD hosts on CI, we have seen tests
like that fail when run in parallel. (This may have nothing to do with
FreeBSD and may just mean that the hosts are resource-constrained.) Move
this test to sequential as we have done with several other
timer-dependent tests recently.
The test has also been refactored and documented via comments.
PR-URL: https://github.com/nodejs/node/pull/9487
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Minwoo Jung <jmwsoft@gmail.com>
Timer-dependent tests fail frequently on certain platforms in CI when
run in parallel with other tests, likely due to competition for
resources. Move test-repl-timeout-throw to sequential to avoid this
problem. Also did some minor refactoring (var->const and more use of
assert.strictEqual of looser assertions).
PR-URL: https://github.com/nodejs/node/pull/9431
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Johan Bergstrom <bugs@bergstroem.nu>
checkExecSyncError() creates error objects for execSync() and
execFileSync(). If the child process created stderr output, then
it is attached to the end of the error message. However, stderr
can be an empty Buffer object, which always passes the truthy
check, leading to an extra newline in the error message. This
commit adds a length check, which will work with both strings and
Buffers.
PR-URL: https://github.com/nodejs/node/pull/9343
Reviewed-By: Wyatt Preul <wpreul@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Remove unneeded timers from some tests and move others from parallel
testing to sequential testing.
This is to resolve test failures on freebsd10-64 on CI. The failures
are all due to timers firing later than expected. Timers firing later
than they are set for can happen on resource-constrained hosts and is
not a bug.
In general, it may be wise to put tests that depend on timing into
sequential testing rather than parallel testing, as the timing can
be affected by other simultaneously-running test processes.
Fixes: https://github.com/nodejs/node/issues/8041
Fixes: https://github.com/nodejs/node/issues/9227
PR-URL: https://github.com/nodejs/node/pull/9317
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Julien Gilli <jgilli@nodejs.org>
Reviewed-By: Johan Bergstrom <bugs@bergstroem.nu>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Michaël Zasso