This is a quick fix to attempt to resolve most of the travis failures by rsafonseca · Pull Request #482 · apache/cloudstack

rsafonseca · 2015-06-18T20:08:29Z

Most of the failures have been due to transient network failures, that are causing dependency artifact downloads to fail
Maven does not have a way to retry this without restarting the whole build, so the mvn dependency plugin is the best bet
Unfortunately, running a dependency:resolve on the project returns yet to be compiled dependencies, and causes it to fail...
There is an option to excludeGroupIds and excludeArtifactIds in the docs for this goal, but unfortunately they don't seem to work
This drafts a dummy pom in a quick and dirty way, just to download all the deps in one go, while retrying for RETRY_COUNT times if it fails

UPDATE: this is no longer a quick fix, as it fixes several issues with travis ;)

rsafonseca · 2015-06-18T20:49:37Z

Travis failure on 2c0d07e was due to second phase not running after i moved the script to tools/travis
xerxesImpl is downloaded on that phase
Issue is already fixed is in d2faa90

asfbot · 2015-06-18T21:50:24Z

cloudstack-pull-requests #491 SUCCESS
This pull request looks good

asfbot · 2015-06-18T23:29:24Z

cloudstack-pull-requests #492 ABORTED

rsafonseca · 2015-06-18T23:58:37Z

Travis is still not looking very well :( will resume soon

rsafonseca · 2015-06-19T11:49:37Z

7 successes in a row is looking better :)
Will do a couple more minor changes to force some more runs.. will rebase at the end when it feels like travis is running stable

DaanHoogland · 2015-06-19T13:55:31Z

Are you done with this @rsafonseca ? As in; should I merge?

rsafonseca · 2015-06-19T14:36:43Z

@DaanHoogland LGTM, altough travis has been very stable today :)
Lastest commit is also useful, since the after_failure.sh script was doing nothing at all.. this will allow some insight into failures such as these:

https://travis-ci.org/apache/cloudstack/jobs/67514846

rsafonseca · 2015-06-19T17:05:50Z

Perfect, last travis failure was just what i was referring to in the last comment, after_failure.sh printed full marvin logs, we can properly troubleshoot that issue now :)
Will look into that later or during the weekend if no one else beats me to it hehe :)

rsafonseca · 2015-06-19T17:44:34Z

It seems like this test sometimes fails because the SSVM takes a while to startup.. the test/integration/smoke/test_secondary_storage.py:159 asserts that "Starting" != "Running" and just exits there.
This needs to allow some time for the SSVM to start if state="Starting" instead of immediately throwing an AssertionError.
File hasn't been touched for over a year, so i guess this is a pretty old issue on travis, also explains why it's intermittent.

asfbot · 2015-06-19T18:42:27Z

cloudstack-pull-requests #493 SUCCESS
This pull request looks good

rsafonseca · 2015-06-19T19:10:06Z

Mentioned issue fixed in #493
:D

rsafonseca · 2015-06-19T19:27:37Z

Ever since i first saw this failure I wondered... why is this considered a success?!?

=== TestName: test_02_sys_template_ready | Status : EXCEPTION ===
+----------------------------+---------+
| test_02_sys_template_ready | Success |
+----------------------------+---------+

Fixed in the latest commit.

rsafonseca · 2015-06-19T21:04:05Z

It seems latest fix uncovered tests that were failing but actually reporting success!

asfbot · 2015-06-19T23:44:29Z

cloudstack-pull-requests #512 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-20T02:28:22Z

cloudstack-pull-requests #516 FAILURE
Looks like there's a problem with this pull request

rsafonseca · 2015-06-21T02:05:34Z

I'd say this now downloads around 99% of project dependencies before build starts
Mostly just maven-remote-resources-plugin stuff is still downloaded, plus very few others that somehow are not picked-up by the script runs... hopefully 99% will be enough to have travis never fail with download error/timeouts.. will launch some load on travis to see :)

asfbot · 2015-06-21T02:13:22Z

cloudstack-pull-requests #517 ABORTED

asfbot · 2015-06-21T02:15:52Z

cloudstack-pull-requests #518 ABORTED

asfbot · 2015-06-21T02:24:01Z

cloudstack-pull-requests #519 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-21T02:25:29Z

cloudstack-pull-requests #522 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-21T02:33:42Z

cloudstack-pull-requests #530 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-21T02:35:05Z

cloudstack-pull-requests #531 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-21T02:35:28Z

cloudstack-pull-requests #532 FAILURE
Looks like there's a problem with this pull request

asfbot · 2015-06-21T02:36:43Z

cloudstack-pull-requests #525 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:36:43Z

cloudstack-pull-requests #523 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:36:43Z

cloudstack-pull-requests #520 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:36:43Z

cloudstack-pull-requests #521 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:37:36Z

cloudstack-pull-requests #529 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:37:43Z

cloudstack-pull-requests #524 SUCCESS
This pull request looks good

asfbot · 2015-06-21T02:46:26Z

cloudstack-pull-requests #533 FAILURE
Looks like there's a problem with this pull request

Add extra echo command to make sure there's data in the pom

Add some verbosity for status checking

…uts on this phase Travis is using maven 3.2.5 and this script was setting an alternate maven path before running script Setting CATALINA_BASE and CATALINA_HOME is of no consequence while running jetty, just adds more problems

…is unavailable Check if DNS resolution is working properly when failure occurs

…rough the while loop. This will make sure nc gets restared to retry in case something nasty is happening in the running process host command not present in build env, switch to getent second phase of dep download turning some error, print failure log

This version didn't like missing <version> tag on artifacts

Sometimes after a full successful run, travis will fail due to not being able to parse tests output

Add urandom as random source in before_script.sh Remove commented lines in before_script.sh Remove commented lines in install.sh

…wn during test! Revert the fix to throw failed test on exception, will fix that in a separate branch so all the other fixes can be committed already and not raise failure on an old intermittent issue that was hidden

…ersion too low for marvin, nosetests wouldn't run and result would be success

…e dependency:resolve-plugins mojo This adds a couple more dark magic sprinkles to make sure no dependencies will be downloaded at build time

Add download for midonet and juniper stuff, this increases runtime for a few minutes, may remove in future if travis show too much overhead

DaanHoogland · 2015-06-21T10:36:53Z

LGTM waiting for travis to agree

asfbot · 2015-06-21T10:44:41Z

cloudstack-pull-requests #548 SUCCESS
This pull request looks good

asfbot · 2015-06-21T10:44:41Z

cloudstack-pull-requests #549 SUCCESS
This pull request looks good

Signed-off-by: Daan Hoogland <daan@onecht.net> This closesapache#482

Fixes#466 Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>

rsafonseca added 4 commits June 21, 2015 12:21

Second phase wan't running properly due to path issue
5b2e54e
Add extra echo command to make sure there's data in the pom

It seems the network failures last for a bit.. allow some sleeping time
270aa26
Add some verbosity for status checking

Don't sleep for a fixed amount of time, no use in continuing if repo …
f84c5a6
…is unavailable Check if DNS resolution is working properly when failure occurs

rsafonseca force-pushed the fixtravis branch from 3fcfe11 to 4383646Compare June 21, 2015 10:30

rsafonseca added 2 commits June 21, 2015 12:30

Fix phase2 deps download for maven 3.2.5
140d75a
This version didn't like missing <version> tag on artifacts

rsafonseca force-pushed the fixtravis branch from 4383646 to cfbb1edCompare June 21, 2015 10:32

rsafonseca added 16 commits June 21, 2015 12:33

Add some retry and debug to python packages install
defd820
Sometimes after a full successful run, travis will fail due to not being able to parse tests output

Can't run travis_wait from script, this just throws an error
964cbf8

Tomcat download is not doing anything except waste time and disk
6973aaf

Remove some more unwanted stuff in install.sh
2628cf6

Add urandom as random source in install.sh
47678ee
Add urandom as random source in before_script.sh Remove commented lines in before_script.sh Remove commented lines in install.sh

Use upstart to start mysql and get rid of the warnings
99a8ff2

Print log data on after_failure.sh
679c50b

Fix erroneous assert expression to get debug from that line
804fd0a

Travis reported another false negative, if paramiko was present but v…
24ba8cb
…ersion too low for marvin, nosetests wouldn't run and result would be success

Allow proper retry in pip install command
664f9d9

Manually add another dependency to be downloaded before build
21aa92a

Some of the plugin dependencies were'nt being properly resolved on th…
91a7efe
…e dependency:resolve-plugins mojo This adds a couple more dark magic sprinkles to make sure no dependencies will be downloaded at build time

Improve plugin dependency download
ca21071
Add download for midonet and juniper stuff, this increases runtime for a few minutes, may remove in future if travis show too much overhead

Check return code on plugin runs
4d44baa

Add apache license header to script
cfbb1ed

asfgit closed this in 6fd49e4Jun 21, 2015

kishankavala pushed a commit to kishankavala/cloudstack that referenced this pull request Jun 25, 2015
Add apache license header to script
92839d4
Signed-off-by: Daan Hoogland <daan@onecht.net> This closesapache#482

yadvr pushed a commit that referenced this pull request Jan 20, 2021
component: Adding contextual info for resource status (#482)
56ca733
Fixes#466 Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>

Uh oh!

Conversation

rsafonseca commented Jun 18, 2015

Uh oh!

rsafonseca commented Jun 18, 2015

Uh oh!

asfbot commented Jun 18, 2015

Uh oh!

asfbot commented Jun 18, 2015

Uh oh!

rsafonseca commented Jun 18, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

DaanHoogland commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

asfbot commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

rsafonseca commented Jun 19, 2015

Uh oh!

asfbot commented Jun 19, 2015

Uh oh!

asfbot commented Jun 20, 2015

Uh oh!

rsafonseca commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

DaanHoogland commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

asfbot commented Jun 21, 2015

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants