Uploaded image for project: 'Jenkins'
  1. Jenkins
  2. JENKINS-18641

People View does Not Populate (makeStaplerProxy)

    • Icon: Bug Bug
    • Resolution: Fixed
    • Icon: Major Major
    • core
    • Windows Server 2008, Debian 7, Ubuntu 12.04, CentOS 6.4

      On entering the "People" view on Jenkins, some page elements load while the list of people is not. Instead a message displays along the lines of:

      makeStaplerProxy('/jenkins/$stapler/bound/7544a1d9-a6f5-44ea-8a6f-ccfd839f03e9','65215ee9-8e24-4f45-96b9-4e34a01b79cf',['news']), display);

          [JENKINS-18641] People View does Not Populate (makeStaplerProxy)

          Observing the same issue on Windows 7. The issue appeared in after upgrading to 1.520.

          Wolf Wolfswinkel added a comment - Observing the same issue on Windows 7. The issue appeared in after upgrading to 1.520.

          Same thing on CentOS 6.4

          Florian Rosenberg added a comment - Same thing on CentOS 6.4

          Shannon Kerr added a comment -

          Same thing on Ubuntu 12.04. Reverted to 1.518 and the issue went away. Can anyone confirm this still exists in the latest 1.522?

          Shannon Kerr added a comment - Same thing on Ubuntu 12.04. Reverted to 1.518 and the issue went away. Can anyone confirm this still exists in the latest 1.522?

          sogabe added a comment -

          I can confirm this in the 1.522 and 1.524-SNAPSHOT, too.

          sogabe added a comment - I can confirm this in the 1.522 and 1.524-SNAPSHOT, too.

          Love Nyberg added a comment -

          Same thing on Debian 7, Jenkins ver. 1.522

          Love Nyberg added a comment - Same thing on Debian 7, Jenkins ver. 1.522

          sogabe added a comment -

          I created pull a request https://github.com/jenkinsci/jenkins/pull/844 .
          It'll fix this issue but may be workaround. I'm not expert of Javascript.

          sogabe added a comment - I created pull a request https://github.com/jenkinsci/jenkins/pull/844 . It'll fix this issue but may be workaround. I'm not expert of Javascript.

          Jesse Glick added a comment -

          Jesse Glick added a comment - 1.521 introduced Stapler 1.213 with: https://github.com/stapler/stapler/commit/48cdf0f3242c6a79bb398e9115c0db7526c42a50

          Shannon Kerr added a comment -

          Jesse, I thought we first saw this behavior in the 1.520 release though? I can go back and try it again.

          Shannon Kerr added a comment - Jesse, I thought we first saw this behavior in the 1.520 release though? I can go back and try it again.

          Jesse Glick added a comment - - edited

          Not sure. Cannot reproduce even in trunk on Chrome on Linux. Wrote an HTMLUnit test for it, but that passes in trunk too. Perhaps the <script> tag is only inserted under certain circumstances?

          Display seems broken in IE 8, but that appears to be a different problem (in sortable.js), present even in 1.509.2.

          Jesse Glick added a comment - - edited Not sure. Cannot reproduce even in trunk on Chrome on Linux. Wrote an HTMLUnit test for it, but that passes in trunk too. Perhaps the <script> tag is only inserted under certain circumstances? Display seems broken in IE 8, but that appears to be a different problem (in sortable.js ), present even in 1.509.2.

          Shannon Kerr added a comment -

          OK, I'm still on 1.518 (there was another bug in 519 that I'm avoiding). Can anyone confirm that 1.523 does not have this problem any longer?

          Shannon Kerr added a comment - OK, I'm still on 1.518 (there was another bug in 519 that I'm avoiding). Can anyone confirm that 1.523 does not have this problem any longer?

          Jesse Glick added a comment -

          Cannot work on this if I cannot reproduce the problem to begin with.

          Jesse Glick added a comment - Cannot work on this if I cannot reproduce the problem to begin with.

          Shannon Kerr added a comment -

          btw, My client is latest Chrome on Win 7. My host is Ubuntu 12.04 Server VM.

          Shannon Kerr added a comment - btw, My client is latest Chrome on Win 7. My host is Ubuntu 12.04 Server VM.

          dogfood added a comment -

          Integrated in jenkins_main_trunk #2716
          JENKINS-18641 Test passing in 1.520. (Revision 50c2bb39008d2da60a957b2c8b5ef7d8e09802cc)

          Result = SUCCESS
          Jesse Glick : 50c2bb39008d2da60a957b2c8b5ef7d8e09802cc
          Files :

          • test/src/test/java/hudson/model/AsynchPeopleTest.java

          dogfood added a comment - Integrated in jenkins_main_trunk #2716 JENKINS-18641 Test passing in 1.520. (Revision 50c2bb39008d2da60a957b2c8b5ef7d8e09802cc) Result = SUCCESS Jesse Glick : 50c2bb39008d2da60a957b2c8b5ef7d8e09802cc Files : test/src/test/java/hudson/model/AsynchPeopleTest.java

          sogabe added a comment - - edited

          @Shannon, Did you install JQuery plugin?

          Jenkins #A with JQuery Plugin -> reproduced
          Jenkins #B without Jquery plugin -> not reproduced

          sogabe added a comment - - edited @Shannon, Did you install JQuery plugin? Jenkins #A with JQuery Plugin -> reproduced Jenkins #B without Jquery plugin -> not reproduced

          Jesse Glick added a comment -

          I was thinking it had to be something like that. Not sure how to capture this in the existing functional test, or if we need to; can just leave it for other possible regressions in ProgressiveRendering usage.

          @sogabe if you can reproduce with the JQuery plugin installed, but not after applying your patch, then go ahead and merge it!

          Jesse Glick added a comment - I was thinking it had to be something like that. Not sure how to capture this in the existing functional test, or if we need to; can just leave it for other possible regressions in ProgressiveRendering usage. @sogabe if you can reproduce with the JQuery plugin installed, but not after applying your patch, then go ahead and merge it!

          Shannon Kerr added a comment -

          @sogabe, yes, I have it installed. Good catch.

          Shannon Kerr added a comment - @sogabe, yes, I have it installed. Good catch.

          Code changed in jenkins
          User: ssogabe
          Path:
          core/src/main/resources/lib/layout/progressiveRendering.jelly
          http://jenkins-ci.org/commit/jenkins/c2db894ad5ee41a8fd6148bc840e2b82780944f1
          Log:
          JENKINS-18641 <st:bind> tag adds <script> tag. This extra tag caused syntax error.

          SCM/JIRA link daemon added a comment - Code changed in jenkins User: ssogabe Path: core/src/main/resources/lib/layout/progressiveRendering.jelly http://jenkins-ci.org/commit/jenkins/c2db894ad5ee41a8fd6148bc840e2b82780944f1 Log: JENKINS-18641 <st:bind> tag adds <script> tag. This extra tag caused syntax error.

          Code changed in jenkins
          User: Seiji Sogabe
          Path:
          core/src/main/resources/lib/layout/progressiveRendering.jelly
          http://jenkins-ci.org/commit/jenkins/8c8219114dcfd5bb2b63715f6aa33465a584a91b
          Log:
          Merge pull request #844 from ssogabe/JENKINS-18641

          [FIXED JENKINS-18641] <st:bind> tag adds <script> tag. This extra tag caused syntax error if JQuery plugin enabled.

          Compare: https://github.com/jenkinsci/jenkins/compare/1ecf4b14d7be...8c8219114dcf

          SCM/JIRA link daemon added a comment - Code changed in jenkins User: Seiji Sogabe Path: core/src/main/resources/lib/layout/progressiveRendering.jelly http://jenkins-ci.org/commit/jenkins/8c8219114dcfd5bb2b63715f6aa33465a584a91b Log: Merge pull request #844 from ssogabe/ JENKINS-18641 [FIXED JENKINS-18641] <st:bind> tag adds <script> tag. This extra tag caused syntax error if JQuery plugin enabled. Compare: https://github.com/jenkinsci/jenkins/compare/1ecf4b14d7be...8c8219114dcf

          Code changed in jenkins
          User: ssogabe
          Path:
          changelog.html
          http://jenkins-ci.org/commit/jenkins/0f515fe81064ad414f166beb82a6d47d59f61e21
          Log:
          updated changelog for JENKINS-18641

          SCM/JIRA link daemon added a comment - Code changed in jenkins User: ssogabe Path: changelog.html http://jenkins-ci.org/commit/jenkins/0f515fe81064ad414f166beb82a6d47d59f61e21 Log: updated changelog for JENKINS-18641

          dogfood added a comment -

          Integrated in jenkins_main_trunk #2721
          JENKINS-18641 <st:bind> tag adds <script> tag. This extra tag caused syntax error. (Revision c2db894ad5ee41a8fd6148bc840e2b82780944f1)
          updated changelog for JENKINS-18641 (Revision 0f515fe81064ad414f166beb82a6d47d59f61e21)

          Result = UNSTABLE
          s.sogabe : c2db894ad5ee41a8fd6148bc840e2b82780944f1
          Files :

          • core/src/main/resources/lib/layout/progressiveRendering.jelly

          s.sogabe : 0f515fe81064ad414f166beb82a6d47d59f61e21
          Files :

          • changelog.html

          dogfood added a comment - Integrated in jenkins_main_trunk #2721 JENKINS-18641 <st:bind> tag adds <script> tag. This extra tag caused syntax error. (Revision c2db894ad5ee41a8fd6148bc840e2b82780944f1) updated changelog for JENKINS-18641 (Revision 0f515fe81064ad414f166beb82a6d47d59f61e21) Result = UNSTABLE s.sogabe : c2db894ad5ee41a8fd6148bc840e2b82780944f1 Files : core/src/main/resources/lib/layout/progressiveRendering.jelly s.sogabe : 0f515fe81064ad414f166beb82a6d47d59f61e21 Files : changelog.html

          Jesse Glick added a comment -

          Is this really the right fix? https://wiki.jenkins-ci.org/display/JENKINS/AJAX+with+JavaScript+proxy recommends using <st:bind> inside <script>.

          Jesse Glick added a comment - Is this really the right fix? https://wiki.jenkins-ci.org/display/JENKINS/AJAX+with+JavaScript+proxy recommends using <st:bind> inside <script> .

          Shannon Kerr added a comment -

          Is the fix going to change?

          Shannon Kerr added a comment - Is the fix going to change?

          Shannon Kerr added a comment - - edited

          I still see this in 1.524:

          Includes all known “users”, including login identities which the current security realm can enumerate, as well as people mentioned in commit messages in recorded changelogs.

          makeStaplerProxy('/$stapler/bound/12572177-4402-44ca-8049-11c6c791c655','b3eb158c-2164-4cee-bd76-ca4fe6ccb71c',['news']), display);

          NM, I see the fix is not in the 1.524 changelog.

          Shannon Kerr added a comment - - edited I still see this in 1.524: Includes all known “users”, including login identities which the current security realm can enumerate, as well as people mentioned in commit messages in recorded changelogs. makeStaplerProxy('/$stapler/bound/12572177-4402-44ca-8049-11c6c791c655','b3eb158c-2164-4cee-bd76-ca4fe6ccb71c', ['news'] ), display); NM, I see the fix is not in the 1.524 changelog.

          Yves Schumann added a comment - - edited

          Right, still existing on 1.524 running on Ubuntu 12.04 and accessed with Chrome, FF, Safari and IE

          Yves Schumann added a comment - - edited Right, still existing on 1.524 running on Ubuntu 12.04 and accessed with Chrome, FF, Safari and IE

          sogabe added a comment -

          this will be fixed in 1.525.

          sogabe added a comment - this will be fixed in 1.525.

          Code changed in jenkins
          User: Jesse Glick
          Path:
          test/src/test/java/hudson/model/AsynchPeopleTest.java
          http://jenkins-ci.org/commit/jenkins/50c2bb39008d2da60a957b2c8b5ef7d8e09802cc
          Log:
          JENKINS-18641 Test passing in 1.520.

          SCM/JIRA link daemon added a comment - Code changed in jenkins User: Jesse Glick Path: test/src/test/java/hudson/model/AsynchPeopleTest.java http://jenkins-ci.org/commit/jenkins/50c2bb39008d2da60a957b2c8b5ef7d8e09802cc Log: JENKINS-18641 Test passing in 1.520.

          Just upgraded to 1.525. Unfortunately the issue is still present for me.

          Wolf Wolfswinkel added a comment - Just upgraded to 1.525. Unfortunately the issue is still present for me.

          William Whittle added a comment - - edited

          I'm also seeing this on 1.525 running on CentOS 6.3 (EC2 AMI: ami-eb0b8082). I don't see the issue when running the same Jenkins build on Windows 7.

          William Whittle added a comment - - edited I'm also seeing this on 1.525 running on CentOS 6.3 (EC2 AMI: ami-eb0b8082). I don't see the issue when running the same Jenkins build on Windows 7.

          Shannon Kerr added a comment -

          Confirmed this bug is not Fixed. I am running 1.525 on Ubuntu 12.04 Server + Chrome browser:

          Includes all known “users”, including login identities which the current security realm can enumerate, as well as people mentioned in commit messages in recorded changelogs.

          makeStaplerProxy('/$stapler/bound/a7d06408-0c93-483a-8910-c36b2295fa1f','154d0cb7-97a2-4f82-abda-bd46caf6ff29',['news']), display);

          Shannon Kerr added a comment - Confirmed this bug is not Fixed. I am running 1.525 on Ubuntu 12.04 Server + Chrome browser: Includes all known “users”, including login identities which the current security realm can enumerate, as well as people mentioned in commit messages in recorded changelogs. makeStaplerProxy('/$stapler/bound/a7d06408-0c93-483a-8910-c36b2295fa1f','154d0cb7-97a2-4f82-abda-bd46caf6ff29', ['news'] ), display);

          Shannon Kerr added a comment -

          Does not appear to be fixed in 1.525.

          Shannon Kerr added a comment - Does not appear to be fixed in 1.525.

          Jesse Glick added a comment -

          I think there is nothing wrong with the fix, it is just in 1.526, not 1.525 which is in fact identical to 1.524.

          Jesse Glick added a comment - I think there is nothing wrong with the fix, it is just in 1.526, not 1.525 which is in fact identical to 1.524.

          Shannon Kerr added a comment -

          OK, thanks Jesse. When I looked at the package size for Ubuntu, I did see the 1.525 was slightly larger than 1.524, so I didn't make that connection.

          Shannon Kerr added a comment - OK, thanks Jesse. When I looked at the package size for Ubuntu, I did see the 1.525 was slightly larger than 1.524, so I didn't make that connection.

          Jesse Glick added a comment -

          The same bug applies to other users of <st:bind>. @rsandell suggests adding to layout.jelly:

          <st:adjunct assumes="org.kohsuke.stapler.framework.prototype.prototype"
                          includes="org.kohsuke.stapler.json2"/>
          

          right after the existing include of org.kohsuke.stapler.bind. Needs to be tested for effectiveness (after reverting the original fix of this issue).

          Jesse Glick added a comment - The same bug applies to other users of <st:bind> . @rsandell suggests adding to layout.jelly : <st:adjunct assumes= "org.kohsuke.stapler.framework.prototype.prototype" includes= "org.kohsuke.stapler.json2" /> right after the existing include of org.kohsuke.stapler.bind . Needs to be tested for effectiveness (after reverting the original fix of this issue).

          Actually, I think BindTag's inclusion logic is wrong.

          The idea is that we need either prototype or JQuery for <st:bind> to work correctly.

          If the current page already includes Prototype, there's no need to have "org.kohsuke.stapler.json2" adjunct, as we don't use JSON.stringify()

          In Jenkins, every page automatically gets prototype, so there's never a need for this inclusion.

          I'm going to fix this by tweaking BindTag.java.

          A test can be written by simulating an insertion of jQuery adjunct into the page, so it should be testable.

          Kohsuke Kawaguchi added a comment - Actually, I think BindTag's inclusion logic is wrong. The idea is that we need either prototype or JQuery for <st:bind> to work correctly. If the current page already includes Prototype, there's no need to have "org.kohsuke.stapler.json2" adjunct, as we don't use JSON.stringify() In Jenkins, every page automatically gets prototype, so there's never a need for this inclusion. I'm going to fix this by tweaking BindTag.java. A test can be written by simulating an insertion of jQuery adjunct into the page, so it should be testable.

          Code changed in jenkins
          User: Kohsuke Kawaguchi
          Path:
          core/pom.xml
          http://jenkins-ci.org/commit/jenkins/7e8bbea3336dc65a10c2fd69a20beac20bc6fca3
          Log:
          [FIXED JENKINS-18641]

          Eliminated the wrong script tag generation from <st:bind/>

          Compare: https://github.com/jenkinsci/jenkins/compare/738a1d116409...7e8bbea3336d

          SCM/JIRA link daemon added a comment - Code changed in jenkins User: Kohsuke Kawaguchi Path: core/pom.xml http://jenkins-ci.org/commit/jenkins/7e8bbea3336dc65a10c2fd69a20beac20bc6fca3 Log: [FIXED JENKINS-18641] Eliminated the wrong script tag generation from <st:bind/> Compare: https://github.com/jenkinsci/jenkins/compare/738a1d116409...7e8bbea3336d

          dogfood added a comment -

          Integrated in jenkins_main_trunk #2795

          Result = UNSTABLE

          dogfood added a comment - Integrated in jenkins_main_trunk #2795 Result = UNSTABLE

          Stapler 1.217 cause some regressions JENKINS-19281 and JENKINS-19168

          Philippe Jandot added a comment - Stapler 1.217 cause some regressions JENKINS-19281 and JENKINS-19168

            sogabe sogabe
            zulu12 Wilfredo Velazquez
            Votes:
            8 Vote for this issue
            Watchers:
            18 Start watching this issue

              Created:
              Updated:
              Resolved: