Page Comparison

[09:18:29 CDT(-0500)] <EricDalquist> yay the jarjar'd resource server works
[09:18:49 CDT(-0500)] <EricDalquist> I tweaked the jarjar config so both yui and rhino end up under the org.jasig.resourceserver package
[09:19:05 CDT(-0500)] <EricDalquist> so any other inclusion of either library won't break the resource server APIs
[11:54:28 CDT(-0500)] <EricDalquist> so athena1 the new filters stuff
[11:54:36 CDT(-0500)] <EricDalquist> how do I trigger that via ant?
[11:55:11 CDT(-0500)] <athena1> -Denv=name
[11:55:16 CDT(-0500)] <EricDalquist> got it
[11:55:17 CDT(-0500)] <EricDalquist> thanks
[11:55:21 CDT(-0500)] <athena1> so if you make a filters/eric.properties
[11:55:22 CDT(-0500)] <athena1> yep
[11:55:27 CDT(-0500)] <athena1> same flags for both ant and maven
[11:57:06 CDT(-0500)] <EricDalquist> working on converting the live nightly build to use it
[11:57:59 CDT(-0500)] <athena1> oh yay
[11:58:22 CDT(-0500)] <EricDalquist> also got some docs put up for other projects: https://wiki.jasig.org/display/JCH/Request+a+EC2+Live+Build+Site
[11:59:21 CDT(-0500)] <athena1> oh yay
[11:59:34 CDT(-0500)] <athena1> so i need to figure out what to do to create a umobile quickstart
[11:59:38 CDT(-0500)] <athena1> is all that stuff scripted now?
[11:59:50 CDT(-0500)] <EricDalquist> yes
[11:59:54 CDT(-0500)] <EricDalquist> cd into assembly
[12:00:09 CDT(-0500)] <EricDalquist> and run "ant clean quickstart"
[12:00:25 CDT(-0500)] <EricDalquist> that will create a quickstart based on whatever you have checked out and configured
[12:00:55 CDT(-0500)] <athena1> awesome
[12:01:07 CDT(-0500)] <athena1> is there any way to get custom SQL tables and data into the database?
[12:01:23 CDT(-0500)] <EricDalquist> not out of the box
[12:01:31 CDT(-0500)] <athena1> ok
[12:01:33 CDT(-0500)] <EricDalquist> you could stick them in data.xml
[12:01:34 CDT(-0500)] <athena1> figured that was the answer
[12:01:37 CDT(-0500)] <athena1> hmm
[12:01:39 CDT(-0500)] <EricDalquist> if you're comfortable with doing that
[12:01:41 CDT(-0500)] <EricDalquist> er tables.xml
[12:01:45 CDT(-0500)] <athena1> kind of doubt that'd work well
[12:01:48 CDT(-0500)] <athena1> all hibernate-managed
[12:01:52 CDT(-0500)] <EricDalquist> ah
[12:10:46 CDT(-0500)] <EricDalquist> once I get the live nightly working again I'm going to look at search
[12:11:23 CDT(-0500)] <athena1> sweet!
[12:24:48 CDT(-0500)] <athena1> you're using the maven release plugin now right?
[12:35:39 CDT(-0500)] <EricDalquist> yes
[12:35:45 CDT(-0500)] <athena1> ok, cool
[12:35:51 CDT(-0500)] <athena1> guess i need to change all the artifact names
[12:36:21 CDT(-0500)] <EricDalquist> heh ... yeah if you did a release from your fork that would be interesting
[12:36:28 CDT(-0500)] <athena1> yeah lets not do that
[12:44:21 CDT(-0500)] <holdorph> Pearson is starting to see this problem with the UP_USER guest record disapparing.... http://jasig.275507.n4.nabble.com/guest-account-is-missing-from-the-UP-USER-table-td1678212.html
[12:44:35 CDT(-0500)] <holdorph> does anyone know of something that might cause this?
[12:44:43 CDT(-0500)] <EricDalquist> yes
[12:44:52 CDT(-0500)] <EricDalquist> I think we even got the fix back into svn
[12:44:52 CDT(-0500)] <holdorph> it's happened twice in less then 2 weeks
[12:44:55 CDT(-0500)] <EricDalquist> yeah
[12:45:03 CDT(-0500)] <EricDalquist> let me see if I can find the commit
[12:45:08 CDT(-0500)] <holdorph> thanks.
[12:47:52 CDT(-0500)] <EricDalquist> !@#!@ broken fisheye
[12:48:01 CDT(-0500)] <EricDalquist> so here is the jira issue: https://issues.jasig.org/browse/UP-3060
[12:48:08 CDT(-0500)] <EricDalquist> the cause is really bizzar
[12:48:17 CDT(-0500)] <EricDalquist> bad context classloader causes an xpath to fail
[12:48:25 CDT(-0500)] <EricDalquist> which makes the portal think the layout is corrupt
[12:48:31 CDT(-0500)] <EricDalquist> which causes it to try and reset the layout
[12:48:45 CDT(-0500)] <EricDalquist> and reseting the guest layout == BAD since the reset is done by just deleting the user
[12:49:42 CDT(-0500)] <EricDalquist> having to do a svnlog to find the commit
[12:50:03 CDT(-0500)] <holdorph> Much appreciated. this will save a ton of time.
[12:50:23 CDT(-0500)] <holdorph> the jira says it affects 3.2.4. So that means everyone on a 'released' version of uportal is affected?
[12:50:38 CDT(-0500)] <holdorph> because pearson is experiencing this on 3.1. (forgot the final number)
[12:50:52 CDT(-0500)] <EricDalquist> yes
[12:51:03 CDT(-0500)] <EricDalquist> this likely affects every released version up to 3.2.4
[12:51:17 CDT(-0500)] <EricDalquist> you can edit the issue and add 3.1 to the affects list
[12:51:17 CDT(-0500)] <holdorph> wow. surprised it doesn't come up even more often.
[12:51:32 CDT(-0500)] <EricDalquist> its a really weird ordering issue
[12:51:47 CDT(-0500)] <holdorph> ok, i'll determine the exact version Pearson is on and add that to the issue.
[12:52:12 CDT(-0500)] <EricDalquist> essentially a callback from a portlet needs to hit this one API before a non-portlet callback hits it for a user's session
[12:52:36 CDT(-0500)] <EricDalquist> made me also realize an issue that we've likely had in all versions of uPortal
[12:52:51 CDT(-0500)] <EricDalquist> and that is we never switch the context classloader back to the portals during portlet SPI callbacks
[12:53:00 CDT(-0500)] <EricDalquist> and we probably should be doing that
[12:54:25 CDT(-0500)] <EricDalquist> looks like revs 24018-24019 include the fix
[12:54:59 CDT(-0500)] <holdorph> Thank you very much. I appreciate the effort to look that up. I owe you a beer.
[12:55:05 CDT(-0500)] <EricDalquist> no problem
[12:55:31 CDT(-0500)] <EricDalquist> it was a "fun" issue to debug when we started seeing it
[12:57:53 CDT(-0500)] <holdorph> hmm... my irc client is acting up. restarting.
[13:42:04 CDT(-0500)] <EricDalquist> athena1: can you remind of the state of the search stuff right now?
[13:42:22 CDT(-0500)] <athena1> i only see results for the GSA search service
[13:42:23 CDT(-0500)] <EricDalquist> out of the box the cms portlet and person dir both have code in place to respond to search requests
[13:42:27 CDT(-0500)] <athena1> don't see any portlet search results
[13:42:37 CDT(-0500)] <athena1> are you getting appropriate results back?
[13:42:40 CDT(-0500)] <EricDalquist> but you're not seeing results
[13:42:43 CDT(-0500)] <athena1> yeah
[13:42:48 CDT(-0500)] <EricDalquist> I'm going to start looking now
[13:42:51 CDT(-0500)] <athena1> i just see the results from the service-based ones
[13:42:52 CDT(-0500)] <athena1> ok
[13:42:53 CDT(-0500)] <EricDalquist> I just wanted to make sure I was on the right page
[13:42:55 CDT(-0500)] <athena1> yep
[13:43:01 CDT(-0500)] <athena1> actually the GSA service is commented out by default
[13:43:10 CDT(-0500)] <athena1> so i guess interesting to see whether you get portlet results w/ that off
[13:43:12 CDT(-0500)] <athena1> just on the off chance
[13:43:19 CDT(-0500)] <EricDalquist> ok
[13:51:35 CDT(-0500)] <EricDalquist> yay we're working again: https://up40-nightly.jasig.org/f/u16l1s4/normal/render.uP
[14:01:28 CDT(-0500)] <athena1> yay!
[14:13:14 CDT(-0500)] <EricDalquist> I think eclipse knows its Friday
[14:13:27 CDT(-0500)] <EricDalquist> right after lunch it decided to stop playing nice