ICEfaces
  1. ICEfaces
  2. ICE-6330

When Liferay session expires, listen.icepush requests bombard the server and max out the CPU

    Details

    • Type: Bug Bug
    • Status: Closed
    • Priority: Critical Critical
    • Resolution: Cannot Reproduce
    • Affects Version/s: 2.0-Beta2
    • Fix Version/s: 2.0.0
    • Component/s: Framework
    • Labels:
      None
    • Environment:
      Liferay 6.0.5 + ICEfaces 2.0.0-Beta2 + PortletFaces Bridge 2.0.0-BETA2

      Description

      After viewing a portlet with Ajax Push enabled, and then waiting for the Liferay session to expire (see attached screenshot), the browser bombards the server (maxing out the CPU) with repeated requests like the following:

      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "POST /user/joe.bloggs/dashboard?_1_WAR_intramailportlet_javax.faces.resource=listen.icepush&p_p_cacheability=cacheLevelPage&p_p_col_count=1&p_p_col_id=column-2&p_p_id=1_WAR_intramailportlet&p_p_lifecycle=2&p_p_mode=view&p_p_state=normal HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /c/portal/login?redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal&p_l_id=0 HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /web/guest/home?p_p_state=maximized&p_p_mode=view&saveLastPath=0&_58_struts_action=%2flogin%2flogin&p_p_id=58&p_p_lifecycle=0&_58_redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal HTTP/1.1" 304 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "POST /user/joe.bloggs/dashboard?_1_WAR_intramailportlet_javax.faces.resource=listen.icepush&p_p_cacheability=cacheLevelPage&p_p_col_count=1&p_p_col_id=column-2&p_p_id=1_WAR_intramailportlet&p_p_lifecycle=2&p_p_mode=view&p_p_state=normal HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /c/portal/login?redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal&p_l_id=0 HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /web/guest/home?p_p_state=maximized&p_p_mode=view&saveLastPath=0&_58_struts_action=%2flogin%2flogin&p_p_id=58&p_p_lifecycle=0&_58_redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal HTTP/1.1" 304 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "POST /user/joe.bloggs/dashboard?_1_WAR_intramailportlet_javax.faces.resource=listen.icepush&p_p_cacheability=cacheLevelPage&p_p_col_count=1&p_p_col_id=column-2&p_p_id=1_WAR_intramailportlet&p_p_lifecycle=2&p_p_mode=view&p_p_state=normal HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /c/portal/login?redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal&p_l_id=0 HTTP/1.1" 302 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "GET /web/guest/home?p_p_state=maximized&p_p_mode=view&saveLastPath=0&_58_struts_action=%2flogin%2flogin&p_p_id=58&p_p_lifecycle=0&_58_redirect=%2fuser%2fjoe.bloggs%2fdashboard%3f_1_WAR_intramailportlet_javax.faces.resource%3dlisten.icepush%26p_p_cacheability%3dcacheLevelPage%26p_p_col_count%3d1%26p_p_col_id%3dcolumn-2%26p_p_id%3d1_WAR_intramailportlet%26p_p_lifecycle%3d2%26p_p_mode%3dview%26p_p_state%3dnormal HTTP/1.1" 304 -
      127.0.0.1 - - [03/Dec/2010:20:37:38 +0000] "POST /user/joe.bloggs/dashboard?_1_WAR_intramailportlet_javax.faces.resource=listen.icepush&p_p_cacheability=cacheLevelPage&p_p_col_count=1&p_p_col_id=column-2&p_p_id=1_WAR_intramailportlet&p_p_lifecycle=2&p_p_mode=view&p_p_state=normal HTTP/1.1" 302 -



        Activity

        Hide
        Neil Griffin added a comment -

        Screen shot of Liferay session expiration

        Show
        Neil Griffin added a comment - Screen shot of Liferay session expiration
        Hide
        Deryk Sinotte added a comment -

        I'm unable to reproduce this. I've tried with both Beta 2 and the trunk using chat with different browsers and I never see any kind of "push storm" when the session expires. Is there any other information you can provide to replicate this?

        Show
        Deryk Sinotte added a comment - I'm unable to reproduce this. I've tried with both Beta 2 and the trunk using chat with different browsers and I never see any kind of "push storm" when the session expires. Is there any other information you can provide to replicate this?
        Hide
        Deryk Sinotte added a comment -

        I took a look at trying to recreate this issue using our own sample portlets. I put a portlet from Component Showcase (no push) and an instance of Chat (with push) on the same portal page, used them both, and then waited for the session to expire. There was no request storm and the CPU usage remained low so I was unable to reproduce. Probably require a specific test case and/or more specific instructions on how to replicate.

        Show
        Deryk Sinotte added a comment - I took a look at trying to recreate this issue using our own sample portlets. I put a portlet from Component Showcase (no push) and an instance of Chat (with push) on the same portal page, used them both, and then waited for the session to expire. There was no request storm and the CPU usage remained low so I was unable to reproduce. Probably require a specific test case and/or more specific instructions on how to replicate.

          People

          • Assignee:
            Deryk Sinotte
            Reporter:
            Neil Griffin
          • Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved: