Will using more powerful for the project make it more difficult to
ensure that contributor builds are still running in a reasonable time?
As an example of this happening on Travis, contributors currently cannot
run all e2e tests since they timeout, but on apache we have a larger
On 03/09/2019 18:57, Robert Metzger wrote:
I wanted to give a short update on this:
- Arvid, Aljoscha and I have started working on a Gradle PoC, currently
working on making all modules compile and test with Gradle. We've also
identified some problematic areas (shading being the most obvious one)
which we will analyse as part of the PoC.
The goal is to see how much Gradle helps to parallelise our build, and to
avoid duplicate work (incremental builds).
- I am working on setting up a Flink testing infrastructure based on Azure
Pipelines, using more powerful hardware. Alibaba kindly provided me with
two 32 core machines (temporarily), and another company reached out to
privately, looking into options for cheap, fast machines :)
If nobody in the community disagrees, I am going to set up Azure Pipelines
with our apache/flink GitHub as a build infrastructure that exists next to
Flinkbot and flink-ci. I would like to make sure that Azure Pipelines is
equally or even more reliable than Travis, and I want to see what the
required maintenance work is.
On top of that, Azure Pipelines is a very feature-rich tool with a lot of
nice options for us to improve the build experience (statistics about tests
(flaky tests etc.), nice docker support, plenty of free build resources for
open source projects, ...)
On Mon, Aug 19, 2019 at 5:12 PM Robert Metzger <rmetz...@apache.org> wrote:
I have summarized all arguments mentioned so far + some additional
research into a Wiki page here:
I'm happy to hear further comments on my summary! I'm pretty sure we can
find more pro's and con's for the different options.
My opinion after looking at the options:
- Flink relies on an outdated build tool (Maven), while a good
alternative is well-established (gradle), and will likely provide a much
better CI and local build experience through incremental build and cached
Scripting around Maven, or splitting modules / test execution /
repositories won't solve this problem. We should rather spend the effort in
migrating to a modern build tool which will provide us benefits in the long
- Flink relies on a fairly slow build service (Travis CI), while
simply putting more money onto the problem could cut the build time at
least in half.
We should consider using a build service that provides bigger machines
to solve our build time problem.
My opinion is based on many assumptions (gradle is actually as fast as
promised (haven't used it before), we can build Flink with gradle, we find
sponsors for bigger build machines) that we need to test first through PoCs.
On Mon, Aug 19, 2019 at 10:26 AM Aljoscha Krettek <aljos...@apache.org>
I did a quick test: a normal "mvn clean install -DskipTests
-Drat.skip=true -Dmaven.javadoc.skip=true -Punsafe-mapr-repo” on my machine
takes about 14 minutes. After removing all mentions of maven-shade-plugin
the build time goes down to roughly 11.5 minutes. (Obviously the resulting
Flink won’t work, because some expected stuff is not packaged and most of
the end-to-end tests use the shade plugin to package the jars for testing.
On 18. Aug 2019, at 19:52, Robert Metzger <rmetz...@apache.org> wrote:
I wanted to understand the impact of the hardware we are using for
our tests. Each travis worker has 2 virtual cores, and 7.5 gb memory
They are using Google Cloud Compute Engine *n1-standard-2* instances.
Running a full "mvn clean verify" takes *03:32 h* on such a machine
Running the same workload on a 32 virtual cores, 64 gb machine, takes
What is interesting are the per-module build time differences.
Modules which are parallelizing tests well greatly benefit from the
"flink-tests" 36:51 min vs 4:33 min
"flink-runtime" 23:41 min vs 3:47 min
"flink-table-planner" 15:54 min vs 3:13 min
On the other hand, we have modules which are not parallel at all:
"flink-connector-kafka": 16:32 min vs 15:19 min
"flink-connector-kafka-0.11": 9:52 min vs 7:46 min
Also, the checkstyle plugin is not scaling at all.
Chesnay reported some significant speedups by reusing forks.
I don't know how much effort it would be to make the Kafka tests
parallelizable. In total, they currently use 30 minutes on the big
(while 31 CPUs are idling :) )
Let me know what you think about these results. If the community is
generally interested in further investigating into that direction, I
look into software to orchestrate this, as well as sponsors for such an
On Fri, Aug 16, 2019 at 3:27 PM Chesnay Schepler <ches...@apache.org>
@Aljoscha Shading takes a few minutes for a full build; you can see
quite easily by looking at the compile step in the misc profile
<https://api.travis-ci.org/v3/job/572560060/log.txt>; all modules that
longer than a fraction of a section are usually caused by shading lots
of classes. Note that I cannot tell you how much of this is spent on
relocations, and how much on writing the jar.
Personally, I'd very much like us to move all shading to flink-shaded;
this would finally allows us to use newer maven versions without
cumbersome workarounds for flink-dist. However, this isn't a trivial
affair in some cases; IIRC calcite could be difficult to handle.
On another note, this would also simplify switching the main repo to
another build system, since you would no longer had to deal with
relocations, just packaging + merging NOTICE files.
@BowenLi I disagree, flink-shaded does not include any tests, API
compatibility checks, checkstyle, layered shading (e.g., flink-runtime
and flink-dist, where both relocate dependencies and one is bundled by
the other), and, most importantly, CI (and really, without CI being
covered in a PoC there's nothing to discuss).
On 16/08/2019 15:13, Aljoscha Krettek wrote:
Speaking of flink-shaded, do we have any idea what the impact of
is on the build time? We could get rid of shading completely in the
main repository by moving everything that we shade to flink-shaded.
On 16. Aug 2019, at 14:58, Bowen Li <bowenl...@gmail.com> wrote:
+1 to Till's points on #2 and #5, especially the potential
gradual migration approach if we decide to go that route.
To add on, I want to point it out that we can actually start with
flink-shaded project  which is a perfect candidate for PoC. It's
smaller size, totally isolated from and not interfered with flink
, and it actually covers most of our practical feature
a build tool - all making it an ideal experimental field.
On Fri, Aug 16, 2019 at 4:52 AM Till Rohrmann <trohrm...@apache.org>
For the sake of keeping the discussion focused and not cluttering
discussion thread I would suggest to split the detailed reporting
reusing JVMs to a separate thread and cross linking it from here.
On Fri, Aug 16, 2019 at 1:36 PM Chesnay Schepler <
TL;DR: table-planner is a good candidate for enabling fork reuse
away, while flink-tests has the potential for huge savings, but we
to figure out some issues first.
Build link: https://travis-ci.org/zentol/flink/builds/572659220
4/8 profiles failed.
No speedup in libraries, python, blink_planner, 7 minutes saved in
The kafka and connectors profiles both fail in kafka tests due to
producer leaks, and no speed up could be confirmed so far:
java.lang.AssertionError: Detected producer leak. Thread name:
kafka-producer-network-thread | producer-239
The tests profile failed due to various errors in migration tests:
junit.framework.AssertionFailedError: Did not see the expected
results within time limit.
*However*, a normal tests run takes 40 minutes, while this one
failed after 19 minutes and is only missing the migration tests
currently need 6-7 minutes). So we could save somewhere between 15
Finally, the misc profiles fails in YARN:
No significant speedup could be observed in other modules; for
flink-yarn-tests we can maybe get a minute or 2 out of it.
On 16/08/2019 10:43, Chesnay Schepler wrote:
There appears to be a general agreement that 1) should be looked
I've setup a branch with fork reuse being enabled for all tests;
report back the results.
On 15/08/2019 09:38, Chesnay Schepler wrote:
improving our build times is a hot topic at the moment so let's
discuss the different ways how they could be reduced.
First up, let's look at some numbers:
1 full build currently consumes 5h of build time total ("total
time"), and in the ideal case takes about 1h20m ("run time") to
complete from start to finish. The run time may fluctuate of
depending on the current Travis load. This applies both to
the Apache and flink-ci Travis.
At the time of writing, the current queue time for PR jobs
running on flink-ci) is about 30 minutes (which basically means
we are processing builds at the rate that they come in), however
are in an admittedly quiet period right now.
2 weeks ago the queue times on flink-ci peaked at around 5-6h as
everyone was scrambling to get their changes merged in time for
(Note: Recently optimizations where added to ci-bot where pending
builds are canceled if a new commit was pushed to the PR or the
was closed, which should prove especially useful during the rush
hours we see before feature-freezes.)
Over the years we have done rather few things to improve this
situation (hence our current predicament).
Beyond the sporadic speedup of some tests, the only notable
in total build times was the introduction of cron jobs, which
consolidated the per-commit matrix from 4 configurations
scala/hadoop versions) to 1.
The separation into multiple build profiles was only a
for the 50m limit on Travis. Running tests in parallel has the
obvious potential of reducing run time, but we're currently
hard limit since a few modules (flink-tests, flink-runtime,
flink-table-planner-blink) are so loaded with tests that they
consume an entire profile by themselves (and thus no further
splitting is possible).
The rework that introduced stages, at the time of introduction,
also not provide a speed up, although this changed slightly once
profiles were added and some optimizations to the caching have
Very recently we modified the surefire-plugin configuration for
flink-table-planner-blink to reuse JVM forks for IT cases,
a significant speedup (18 minutes!). So far we have not seen any
This is a list of /all /suggestions for reducing run/total times
I have seen recently (in other words, they aren't necessarily
nor may I agree with all of them).
1. Enable JVM reuse for IT cases in more modules.
* We've seen significant speedups in the blink planner, and
should be applicable for all modules. However, I presume
a reason why we disabled JVM reuse (information on this
2. Custom differential build scripts
* Setup custom scripts for determining which modules might be
affected by change, and manipulate the splits accordingly.
approach is conceptually quite straight-forward, but has
since it has to be pessimistic; i.e. a change in flink-core
_must_ result in testing all modules.
3. Only run smoke tests when PR is opened, run heavy tests on
* With the introduction of the ci-bot we now have
more options on how to handle PR builds. One option could
only run basic tests when the PR is created (which may be
modified modules, or all unit tests, or another low-cost
scheme), and then have a committer trigger other builds
test run, e2e tests, etc...) on demand.
4. Move more tests into cron builds
* The budget version of 3); move certain tests that are
expensive (like some runtime tests that take minutes) or in
rarely modified modules (like gelly) into cron jobs.
* Gradle was brought up a few times for it's built-in support
differential builds; basically providing 2) without the
of maintaining additional scripts.
* To date no PoC was provided that shows it working in our CI
environment (i.e., handling splits & caching etc).
* This is the most disruptive change by a fair margin, as it
affect the entire project, developers and potentially users
they build from source).
6. CI service
* Our current artifact caching setup on Travis is basically a
hack; we're basically abusing the Travis cache, which is
for long-term caching, to ship build artifacts across jobs.
brittle at times due to timing/visibility issues and on
the cleanup processes can interfere with running builds. It
also not as effective as it could be.
* There are CI services that provide build artifact caching
the box, which could be useful for us.
* To date, no PoC for using another CI service has been