[xwiki-devs] [VOTE] Important: Strategy to fix failing tests and stability
Hi devs, Here's the situation: * We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help. Here's what I propose as a drastic and temporary measure till we get better: 1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC. 3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build. 4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents. Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing. Thanks -Vincent
+1 on all counts. In addition to more build engines we should investigate the build speed and how we can improve it. Caleb On 05/31/2011 10:54 AM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC. 3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build. 4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
+1 sounds good On Tue, May 31, 2011 at 16:54, Vincent Massol <[email protected]> wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC. 3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build. 4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
Is 4) really needed if everyone is supposed to help stabilize the release anyway ?
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
-- Thomas Mortagne
On May 31, 2011, at 7:22 PM, Thomas Mortagne wrote:
+1 sounds good
On Tue, May 31, 2011 at 16:54, Vincent Massol <[email protected]> wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC. 3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build. 4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
Is 4) really needed if everyone is supposed to help stabilize the release anyway ?
I thought about it but put it as an extra safety measure but indeed if 1) works we don't need it. Thanks -Vincent
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
On 05/31/2011 05:54 PM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC.
3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build.
Does this apply to Selenium 1 tests?
4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
I think (1) and (2) are hard to achieve because tests take too long to run. CI is at least 2h behind the latest commit. What are we supposed to do while CI runs the tests? I think we should fix the real problem now: add more agents and review the tests to see why they take so long to run. I'm willing to spend time to make the WYSIWYG tests run faster. So +0. Thanks, Marius
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
On Jun 1, 2011, at 6:30 PM, Marius Dumitru Florea wrote:
On 05/31/2011 05:54 PM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC.
3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build.
Does this apply to Selenium 1 tests?
4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
I think (1) and (2) are hard to achieve because tests take too long to run. CI is at least 2h behind the latest commit. What are we supposed to do while CI runs the tests? I think we should fix the real problem now: add more agents and review the tests to see why they take so long to run. I'm willing to spend time to make the WYSIWYG tests run faster.
BTW on this topic XWiki SAS has purchased some new build server today so that we can have several new agents on it (it should be able to run 4 agents). I'll let you know once it's set up. Thanks -Vincent
So +0.
Thanks, Marius
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
On 06/02/2011 12:35 AM, Vincent Massol wrote:
On Jun 1, 2011, at 6:30 PM, Marius Dumitru Florea wrote:
On 05/31/2011 05:54 PM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly. 2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC.
3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build.
Does this apply to Selenium 1 tests?
4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
I think (1) and (2) are hard to achieve because tests take too long to run. CI is at least 2h behind the latest commit. What are we supposed to do while CI runs the tests? I think we should fix the real problem now: add more agents and review the tests to see why they take so long to run. I'm willing to spend time to make the WYSIWYG tests run faster.
BTW on this topic XWiki SAS has purchased some new build server today so that we can have several new agents on it (it should be able to run 4 agents). I'll let you know once it's set up.
That's great news! Should we set up a Selenium Grid ( http://selenium-grid.seleniumhq.org/ )? Thanks, Marius
Thanks -Vincent
So +0.
Thanks, Marius
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
_______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
On 05/31/2011 04:54 PM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly.
-0. Running all tests takes a lot of time, we can't wait 2 hours after each change to check that it works before committing. -1 for forcing everyone to stop their work when somebody else broke some tests. +1 for a different strategy: Whoever commits something should check the next build, and if something fails, he should stop what he's doing and work on fixing the problem. The build should not stay broken for more than 6 working hours.
2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC.
+1, but for this to work we have to reach a stable state by default, meaning that a failed build should be a rare exception.
3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build.
+1
4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
+0, the term should not be strict. If all tests are passing, then there's no need for such a long stabilization period. If releases should happen on Monday, then the branch could be created Friday morning.
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
+1.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
-- Sergiu Dumitriu http://purl.org/net/sergiu/
On Jun 6, 2011, at 11:20 AM, Șergiu Dumitriu wrote:
On 05/31/2011 04:54 PM, Vincent Massol wrote:
Hi devs,
Here's the situation:
* We've been having a hard time releasing on time and the main issue is test stability. We lag by at least a week and we even release with failing tests, causing regressions. * It's not the role of the release manager to fix tests before releasing * It's not normal that some people spend time fixing issues caused by others and that others continue to work on the next thing they are working on. Everyone needs to help.
Here's what I propose as a drastic and temporary measure till we get better:
1) It's forbidden to commit anything till all tests are passing (unit AND functional tests), unless what is committed is about fixing tests. On the exceptional case when a committer absolutely needs to commit even though tests are failing he needs to ask the permission explicitly.
-0. Running all tests takes a lot of time, we can't wait 2 hours after each change to check that it works before committing.
I didn't mean to wait. I meant to check ci.xwiki.org and verify that the last build was succesful.
-1 for forcing everyone to stop their work when somebody else broke some tests.
+1 for a different strategy: Whoever commits something should check the next build, and if something fails, he should stop what he's doing and work on fixing the problem. The build should not stay broken for more than 6 working hours.
It's not just the next build, it's all other builds since tests can flicker. Also it can take time for a CI error to bubble up (may require several builds). It's not fun to fix tests and even less fun to fix other's tests. Especially when other people are having fun working on the next stuff... We also need to stabilize and make tests passing the norm. So right now nobody should commit anything till all tests are passing and consistently. This is not the case. Thanks -Vincent
2) When tests are failing, everyone should stop what they're doing and help stabilize again. We synchronize on IRC.
+1, but for this to work we have to reach a stable state by default, meaning that a failed build should be a rare exception.
3) Flickering tests can be marked as @Ignore and a jira issue created to stabilize the build.
+1
4) Release Manager creates a release branch 1 week before the release to let everyone stabilize the build
+0, the term should not be strict. If all tests are passing, then there's no need for such a long stabilization period. If releases should happen on Monday, then the branch could be created Friday morning.
On a long term we need to work on improving our CI so that functional tests are built faster. One idea is: more agents and functional tests spread on several agents.
+1.
Here's my +1 to apply this now for master (3.2-SNAPSHOT leading to 3.2M1), which means not committing anything more till we have all functional tests passing.
Thanks -Vincent
-- Sergiu Dumitriu http://purl.org/net/sergiu/ _______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs
participants (5)
-
Caleb James DeLisle -
Marius Dumitru Florea -
Thomas Mortagne -
Vincent Massol -
Șergiu Dumitriu