uPortal IRC Logs-2008-02-13

Owned by Apereo Infrastructure
Last updated: Feb 13, 2008Version comment

[13:55:08 EST(-0500)] * JASIGLogBot (i=jasigch@jasigch.Princeton.EDU) has joined ##uportal
[15:10:47 EST(-0500)] <athena7> ok, i have a problem
[15:11:13 EST(-0500)] <athena7> i seem to keep fouling up my admin account
[15:11:14 EST(-0500)] <athena7> ack
[15:11:18 EST(-0500)] <EricDalquist> those are never good to have
[15:11:27 EST(-0500)] <EricDalquist> like corrupting the layout?
[15:11:31 EST(-0500)] <athena7> yeah
[15:11:38 EST(-0500)] <EricDalquist> hrm
[15:11:45 EST(-0500)] <athena7> lets see if resetting the layout fixes it
[15:12:42 EST(-0500)] <athena7> what happened to the rest layout option?
[15:13:15 EST(-0500)] <athena7> hm.
[15:13:33 EST(-0500)] <EricDalquist> no link in the theme any more?
[15:13:43 EST(-0500)] <athena7> is there a database command to reset individual layouts?
[15:13:51 EST(-0500)] <athena7> i hate to have to drop and recreate the database
[15:14:01 EST(-0500)] <EricDalquist> I don't know ...
[15:14:13 EST(-0500)] <EricDalquist> oh you could just crn-import the admin user's layout file
[15:15:21 EST(-0500)] <EricDalquist> ant crn-import -Ddir=uportal-impl/src/main/resources/db/enties/......
[15:15:27 EST(-0500)] <EricDalquist> I can't remember the exact file location
[15:15:46 EST(-0500)] <athena7> so somehow the admin and demo layouts and a few others broke
[15:15:49 EST(-0500)] <athena7> but the student one is fine
[15:16:02 EST(-0500)] <athena7> i've had this happen a few times now after running the deployPortletApp task
[15:16:08 EST(-0500)] <athena7> which doesn't make a whole lot of sense
[15:16:18 EST(-0500)] <EricDalquist> hrm
[15:16:22 EST(-0500)] <EricDalquist> out of perm gen error?
[15:16:41 EST(-0500)] <EricDalquist> or does restarting tomcat not help?
[15:16:47 EST(-0500)] <athena7> nope
[15:16:59 EST(-0500)] <athena7> there's a null pointer exception on the structure transform, it looks like
[15:17:00 EST(-0500)] <athena7> yikes.
[15:17:26 EST(-0500)] <athena7> Caused by: java.lang.NullPointerException
[15:17:26 EST(-0500)] <athena7> at org.jasig.portal.layout.dlm.RDBMDistributedLayoutStore.getFragmentNodeInfo(RDBMDistributedLayoutStore.java:922)
[15:17:30 EST(-0500)] <EricDalquist> uhg, that's not good
[15:17:43 EST(-0500)] <athena7> it's strange that it's not happening immediately
[15:17:53 EST(-0500)] <athena7> especially since a couple accounts are ones i'd never logged into
[15:18:15 EST(-0500)] <athena7> it could be some error in the xsl, i suppose, but it doesn't look simple to track down
[15:19:24 EST(-0500)] <athena7> http://uportal.pastebin.com/m3467a9cc
[15:20:05 EST(-0500)] <athena7> oh wait
[15:20:22 EST(-0500)] <athena7> actually this one may be a problem with the calendar portlet
[15:20:33 EST(-0500)] <EricDalquist> really?
[15:20:42 EST(-0500)] <athena7> it looks like this error is slightly different from the other one i got
[15:20:51 EST(-0500)] <athena7> so maybe there's hope (smile)
[15:23:26 EST(-0500)] <EricDalquist> (smile)
[15:23:30 EST(-0500)] <athena7> i'll have to look into the other error later if it comes back again
[15:23:41 EST(-0500)] <athena7> strange
[15:35:28 EST(-0500)] <athena7> ok, still having problems
[15:35:29 EST(-0500)] <athena7> http://uportal.pastebin.com/d67fec535
[15:35:52 EST(-0500)] <athena7> the line mentioned is <xsl:apply-templates/>
[15:37:18 EST(-0500)] <EricDalquist> hrm
[15:38:03 EST(-0500)] <EricDalquist> first step is to make line 922 easier to debug
[15:38:07 EST(-0500)] <EricDalquist> from: Element node = defs[i].view.layout.getElementById(sId);
[15:38:09 EST(-0500)] <EricDalquist> to:
[15:38:39 EST(-0500)] <EricDalquist> final FragmentDefinition fragmentDefinition = defs[i];
[15:38:39 EST(-0500)] <EricDalquist> final UserView view = fragmentDefinition.view;
[15:38:39 EST(-0500)] <EricDalquist> final Document layout = view.layout;
[15:38:39 EST(-0500)] <EricDalquist> Element node = layout.getElementById(sId);
[15:38:55 EST(-0500)] <EricDalquist> then we can at least know which of the four dereferences are null
[15:38:58 EST(-0500)] <athena7> yeah
[15:39:32 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[15:39:47 EST(-0500)] <EricDalquist> my least favorite java code to debug is stuff like that
[15:39:52 EST(-0500)] <athena7> no kidding
[15:39:59 EST(-0500)] <EricDalquist> NPEs always pop up and are evil to track down
[15:40:13 EST(-0500)] <athena7> yeah
[15:40:20 EST(-0500)] <athena7> and it's not code i'm familiar with either
[15:40:46 EST(-0500)] <EricDalquist> well if you change that and find out the new line # I can see what help I can give
[15:40:58 EST(-0500)] <athena7> sure, doing it now
[15:44:55 EST(-0500)] <athena7> ok, the NPE is on final Document layout = view.layout;
[15:47:00 EST(-0500)] <athena7> i'm cursing myself at the moment because this somehow seems familiar
[15:49:27 EST(-0500)] <EricDalquist> so the view is null ...
[15:50:13 EST(-0500)] <EricDalquist> do you have any log messages like "Problem occurred activating"
[15:50:33 EST(-0500)] <EricDalquist> FragmentActivator:130 is one place it could get set to null
[15:50:57 EST(-0500)] <athena7> hm, yes, i actually do
[15:51:24 EST(-0500)] <EricDalquist> yeah
[15:51:29 EST(-0500)] <athena7> for the portlets tab, which would explain why it's only certain users
[15:51:30 EST(-0500)] <athena7> Caused by: java.lang.NullPointerException
[15:51:30 EST(-0500)] <athena7> at org.jasig.portal.RDBMChannelRegistryStore.getChannelDefinition(RDBMChannelRegistryStore.java:504)
[15:51:31 EST(-0500)] <EricDalquist> that looks like the only place it could get set to null actually
[15:52:53 EST(-0500)] <EricDalquist> can you pastebin that full stack?
[15:52:57 EST(-0500)] <athena7> sure
[15:53:30 EST(-0500)] <athena7> http://uportal.pastebin.com/def9630
[15:54:19 EST(-0500)] <EricDalquist> interesting
[15:54:21 EST(-0500)] <EricDalquist> that is a nice bug
[15:54:31 EST(-0500)] <EricDalquist> if (channelDef.isPortlet()) {
[15:54:31 EST(-0500)] <athena7> i'm glad you like it (tongue)
[15:54:31 EST(-0500)] <EricDalquist> final IPortletDefinition portletDefinition = this.portletDefinitionRegistry.getPortletDefinition(channelPublishId);
[15:54:31 EST(-0500)] <EricDalquist> final IPortletPreferences portletPreferences = portletDefinition.getPortletPreferences();
[15:54:35 EST(-0500)] <EricDalquist> so
[15:54:41 EST(-0500)] <EricDalquist> I'm guessing you have a portlet in your layout
[15:54:55 EST(-0500)] <EricDalquist> actually .,..
[15:54:57 EST(-0500)] <EricDalquist> hrm
[15:54:58 EST(-0500)] <EricDalquist> weird
[15:55:02 EST(-0500)] <EricDalquist> that shouldn't be possible
[15:55:08 EST(-0500)] <EricDalquist> there is a channel definition
[15:55:10 EST(-0500)] <EricDalquist> for a portlet
[15:55:17 EST(-0500)] <EricDalquist> that doesn't have a corresponding portlet definition
[15:55:26 EST(-0500)] <athena7> that is odd
[15:55:51 EST(-0500)] <EricDalquist> any idea which portlet it is?
[15:56:00 EST(-0500)] <athena7> nope but let me add a debug statement
[15:58:16 EST(-0500)] <EricDalquist> http://uportal.pastebin.com/d37791563
[15:58:25 EST(-0500)] <EricDalquist> a replacement chunk for that section of code
[15:58:42 EST(-0500)] <EricDalquist> have you done any playing in UP_CHANNEL directly?
[15:59:12 EST(-0500)] <athena7> nope
[15:59:22 EST(-0500)] <EricDalquist> very weird
[15:59:22 EST(-0500)] <athena7> ok, i have to run to a meeting
[15:59:27 EST(-0500)] <EricDalquist> np
[15:59:31 EST(-0500)] <athena7> but i'll try that chunk of code when i get home
[15:59:31 EST(-0500)] <athena7> thanks
[15:59:31 EST(-0500)] <EricDalquist> catch ya later
[15:59:35 EST(-0500)] <athena7> i'll let you know how it goes
[15:59:40 EST(-0500)] <EricDalquist> thanks
[16:23:54 EST(-0500)] * michelled (n=team@142.150.154.199) has left ##uportal
[16:40:31 EST(-0500)] * athena7 (n=athena7@c-24-2-189-230.hsd1.ct.comcast.net) has joined ##uportal
[16:41:02 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[16:48:56 EST(-0500)] <athena7> i'm back eric - i had the wrong time for my meeting
[16:48:57 EST(-0500)] <athena7> argh.
[16:49:04 EST(-0500)] <athena7> anyway, it's the library search that's causing the problem
[16:49:18 EST(-0500)] <EricDalquist> weird
[16:49:31 EST(-0500)] <EricDalquist> I'll have to revisit that code
[16:49:40 EST(-0500)] <athena7> now, i'd deleted that one from my layout as an admin
[16:50:18 EST(-0500)] <athena7> which may or may not be relevant - i'm guessing it is
[16:50:40 EST(-0500)] <EricDalquist> hrm ...
[16:50:43 EST(-0500)] <EricDalquist> that could be
[16:50:47 EST(-0500)] <EricDalquist> I'll keep that in mind
[16:50:52 EST(-0500)] <EricDalquist> my clean up code may have done the wrong thing
[16:51:15 EST(-0500)] <athena7> i'll try and see if i can reliably cause it, and see if it makes a difference whether it's deleted via the preferences channel or the ajax interface
[16:58:12 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[17:16:13 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[17:35:15 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[17:55:18 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[18:24:33 EST(-0500)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[19:00:11 EST(-0500)] * athena7 (n=athena7@c-24-2-189-230.hsd1.mn.comcast.net) has joined ##uportal
[20:41:29 EST(-0500)] * colinclark (n=colin@bas1-toronto09-1279336376.dsl.bell.ca) has joined ##uportal