uPortal IRC Logs-2008-09-23

Owned by Apereo Infrastructure
Last updated: Sept 23, 2008Version comment
10 min read

[03:08:05 EDT(-0400)] * mad (n=chatzill@pcit-8752.HIG.SE) has joined ##uportal
[08:52:04 EDT(-0400)] * athena7 (n=athena7@adsl-99-149-83-32.dsl.wlfrct.sbcglobal.net) has joined ##uportal
[08:55:27 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[09:04:01 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[09:37:38 EDT(-0400)] * chrisdoyle (n=chrisdoy@mtw160-1.ippl.jhu.edu) has joined ##uportal
[09:38:05 EDT(-0400)] <chrisdoyle> eric, are you using 3.0.2 as the base for your up3 upgrade at wisconsin?
[09:38:45 EDT(-0400)] <EricDalquist> 3.1
[09:38:55 EDT(-0400)] <EricDalquist> our current vendor branch is off 3.1.0-M1
[09:39:26 EDT(-0400)] <chrisdoyle> wow, bleeding edge – i am in the process of applying all the patches to get us to 3.0.2 – would you recommend also making the leap to 3.1.0-M1 as well?
[09:39:54 EDT(-0400)] <EricDalquist> well ... it should be stable but I have a lot of local mods too :/
[09:40:42 EDT(-0400)] <EricDalquist> there are some nice features such as the swapper portlets
[09:41:04 EDT(-0400)] <EricDalquist> most of the 3.1 dev to date has been driven by the work I'm doing here to get us moved to 3
[09:41:04 EDT(-0400)] <chrisdoyle> ok...i'll wrap up the 3.0.2 patches and probably touch base once i start looking at 3.1.0 then...thought it might be good to try and get on the same page as you guys
[09:41:12 EDT(-0400)] <chrisdoyle> gotcha
[09:51:01 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[10:18:06 EDT(-0400)] * bulloche (n=bulloche@134.250.4.77) has joined ##uportal
[10:41:02 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[11:12:00 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:17:12 EDT(-0400)] * holdorph (n=holdorph@uni1.unicon.net) has joined ##uportal
[11:40:21 EDT(-0400)] <athena7> hm, i'm getting an error with the current trunk build
[11:40:29 EDT(-0400)] <athena7> Caused by: java.lang.ClassNotFoundException: org.jasig.portal.tools.dbloader.HibernateDbLoader
[11:40:45 EDT(-0400)] <athena7> looks like it's coming from persistenceContext.xml?
[11:40:51 EDT(-0400)] <EricDalquist> oh
[11:40:58 EDT(-0400)] <EricDalquist> I probably committed a change there that shouldn't be
[11:41:22 EDT(-0400)] <athena7> oh (smile)
[11:41:42 EDT(-0400)] <EricDalquist> try editing persistenceContext.xml and commenting out the hibernatedbloader bean
[11:42:04 EDT(-0400)] <athena7> ok
[11:48:32 EDT(-0400)] <athena7> ack
[11:48:41 EDT(-0400)] <athena7> now i'm getting crn import errors
[11:48:55 EDT(-0400)] <EricDalquist> uhg
[11:49:27 EDT(-0400)] <EricDalquist> this is what I get for having like four local mods in my local trunk at once (sad)
[11:49:32 EDT(-0400)] <athena7> lol
[11:49:46 EDT(-0400)] <EricDalquist> honestly I'm not sure what could be causing those
[11:49:53 EDT(-0400)] <EricDalquist> I can look into it later this afternoon
[11:50:05 EDT(-0400)] <athena7> ok, thanks
[11:50:23 EDT(-0400)] <athena7> i'm trying to get the fluid stuff updated so they can test
[11:59:28 EDT(-0400)] * holdorph (n=holdorph@uni1.unicon.net) has joined ##uportal
[13:55:20 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:56:30 EDT(-0400)] <awills> hey folks (smile) how does everyone feel about UP-1876: http://www.ja-sig.org/issues/browse/UP-1876
[13:56:45 EDT(-0400)] <EricDalquist> looks really neat
[13:56:55 EDT(-0400)] <athena7> woo!
[13:57:02 EDT(-0400)] <EricDalquist> I wanted to take a look and see if there was a way to do it as a filter instead of adding more code to LoginServlet
[13:57:10 EDT(-0400)] <awills> i know we get a lot of interest in this feature from clients
[13:57:12 EDT(-0400)] <EricDalquist> but I have no idea how feasible that would be
[13:57:36 EDT(-0400)] <chrisdoyle> sounds pretty cool, Drew!
[13:57:52 EDT(-0400)] <awills> it would be totally feasible... i struggled on that point for a bit... it's a tradeoff between login servlet vs. web.xml complexity
[13:58:00 EDT(-0400)] <awills> oh hey chris (smile)
[13:58:06 EDT(-0400)] <EricDalquist> +1 for web.xml any day
[13:58:43 EDT(-0400)] <awills> what i think we could use is some spring-configurable login pipeline for this and other things like it
[13:58:49 EDT(-0400)] <EricDalquist> it also lets you access Spring resources more easily as you could write your fitler as a Spring bean and use DelegatingFilterProxy
[13:58:55 EDT(-0400)] <EricDalquist> well
[13:59:04 EDT(-0400)] <EricDalquist> what we really need is to rm -Rf all the security code
[13:59:07 EDT(-0400)] <EricDalquist> and replace it with acegie
[13:59:10 EDT(-0400)] <chrisdoyle> i see you are taking advantages of the permissions table again (like you did for impersonate) – cool
[13:59:13 EDT(-0400)] <awills> +1 for that
[13:59:16 EDT(-0400)] <EricDalquist> which would give us that fitler based auth system
[13:59:50 EDT(-0400)] <awills> yeah permissions is good – last thing I'd want to do for this is invent a bunch of new primitives (wink)
[14:00:02 EDT(-0400)] <awills> that sounds great
[14:00:15 EDT(-0400)] <awills> for now, i can do it as a filter... shouldn't take long
[14:00:19 EDT(-0400)] <EricDalquist> but that's a pretty significant ammount of work
[14:00:23 EDT(-0400)] <awills> yeah
[14:00:28 EDT(-0400)] <EricDalquist> but yeah, I've been meaning to reply about doing it as a fitler
[14:00:43 EDT(-0400)] <EricDalquist> then it isn't as sensitive to nuking LoginServlet someday in the future too
[14:00:54 EDT(-0400)] <awills> yeah that's no trouble... 3-0-patches and trunk/ ok?
[14:00:54 EDT(-0400)] <EricDalquist> it is very nice code though
[14:00:57 EDT(-0400)] <EricDalquist> yup
[14:01:02 EDT(-0400)] <EricDalquist> both places
[14:01:07 EDT(-0400)] <awills> good deal
[14:01:09 EDT(-0400)] <EricDalquist> we'll likely use it here
[14:02:14 EDT(-0400)] <awills> here's a suggestion i thought of – create a PAGS group for people logging in at public terminals and do a MAX_INACTIVE DENY at 5 min
[14:02:36 EDT(-0400)] <chrisdoyle> wow – that's a great idea –
[14:02:53 EDT(-0400)] <chrisdoyle> we are getting ready to use IP-based pags groups here
[14:02:58 EDT(-0400)] <awills> CO had something like that, but it was hard-coded straight through the framework & theme
[14:03:07 EDT(-0400)] <chrisdoyle> i'll be sure to pass along that idea when we get there
[14:03:22 EDT(-0400)] <EricDalquist> nice
[14:03:32 EDT(-0400)] <awills> DENYs trump GRANTs, so they act as a ceiling
[14:04:56 EDT(-0400)] <awills> so i'm doing coopretive dev for (probably) the rest of this month, so I actually have time to run around knocking these sorts of things out, based on subscriber priorities
[14:05:08 EDT(-0400)] <awills> it's pretty nice
[14:05:59 EDT(-0400)] <EricDalquist> coll
[14:06:00 EDT(-0400)] <EricDalquist> cool
[14:06:21 EDT(-0400)] <EricDalquist> I think I'm going to be able to bundle up my import/export changes tomorrow night
[14:06:43 EDT(-0400)] <EricDalquist> so I can post that for folks to start working on re-integrating back with 2.x/3.0/trunk
[14:07:06 EDT(-0400)] <athena7> ooh yay
[14:07:14 EDT(-0400)] <awills> very nice
[14:07:26 EDT(-0400)] <athena7> now if that sad postgres bug could fix itself someday i'd be all set
[14:07:34 EDT(-0400)] <awills> you think org.jasig.portal.security is a good place for the Filter impl?
[14:07:38 EDT(-0400)] <EricDalquist> yup
[14:07:45 EDT(-0400)] <EricDalquist> yeah athena7 I really wish I knew what was going on there
[14:07:48 EDT(-0400)] <athena7> me too
[14:07:51 EDT(-0400)] <athena7> couldn't figure it out
[14:08:08 EDT(-0400)] <awills> i looked at that a bit jen... looks like postgres wants you to code for transactions differently
[14:08:23 EDT(-0400)] <athena7> by the way eric, is it possible to have portlets as part of the uportal webapp in 3.0 or do we need to wait for 3.1?
[14:08:25 EDT(-0400)] <athena7> weird
[14:08:36 EDT(-0400)] <awills> they want you to use savepoints
[14:08:37 EDT(-0400)] <athena7> is it fixable?
[14:08:50 EDT(-0400)] <athena7> wow, really?? i use postgres all the time and i've never run across that
[14:08:52 EDT(-0400)] <EricDalquist> athena7: you can do it in 3.0
[14:09:01 EDT(-0400)] <athena7> ok, good - is it just that we need to use tomcat 6?
[14:09:04 EDT(-0400)] <EricDalquist> you just have to edit the portlet app path by hand
[14:09:06 EDT(-0400)] <EricDalquist> nope
[14:09:20 EDT(-0400)] <EricDalquist> so the 3.1 feature was adding the concept of a 'framework portlet'
[14:09:40 EDT(-0400)] <EricDalquist> if a portlet is marked as such the portal automatically figures out the portlet app path (ex: /uPortal)
[14:09:50 EDT(-0400)] <athena7> oh! ok
[14:10:00 EDT(-0400)] <EricDalquist> so that if you're deploying uPortal to say /portal you don't have to go muck with all of the framework portlets by hand
[14:10:15 EDT(-0400)] <athena7> that makes sense
[14:10:15 EDT(-0400)] <EricDalquist> portlets work just find in the uPortal webapp in 2.x or 3.x
[14:10:26 EDT(-0400)] <EricDalquist> just pre 3.1 you have to manually change that webapp path
[14:10:35 EDT(-0400)] <athena7> i filtered out that context location already i think so i won't have to manually change it (tongue)
[14:10:59 EDT(-0400)] <athena7> though i would in a vanilla portal
[14:19:10 EDT(-0400)] <athena7> so it'll work ok in tomcat 5.5 too, right?
[14:19:18 EDT(-0400)] <EricDalquist> yup
[14:19:49 EDT(-0400)] <EricDalquist> the need for TC6 was for Servlet 2.5 which added an API that allows a servlet to determine the context path without having to have a current request object
[14:19:56 EDT(-0400)] <athena7> ok, cool
[14:33:00 EDT(-0400)] <awills> got to run an errand... i'll be back on in the afternoon
[14:33:08 EDT(-0400)] <chrisdoyle> cya
[14:36:33 EDT(-0400)] <chrisdoyle> so while i was manually modifying some of the files for our 3.0.2 upgrade today, i noticed a couple of lingering whirlycache configurations in log4j.properties:
[14:36:34 EDT(-0400)] <chrisdoyle> log4j.logger.com.whirlycott.cache=WARN, R
[14:36:50 EDT(-0400)] <chrisdoyle> shall i jira and remove?
[14:36:58 EDT(-0400)] <EricDalquist> yup
[14:37:01 EDT(-0400)] <chrisdoyle> k
[14:38:26 EDT(-0400)] <holdorph> wait
[14:38:36 EDT(-0400)] <holdorph> what do you mean, you want to remove those?
[14:38:48 EDT(-0400)] <EricDalquist> yeah
[14:38:49 EDT(-0400)] <holdorph> they were added for a reason. do you know if that reason has gone away?
[14:38:49 EDT(-0400)] <chrisdoyle> for version 3, yes
[14:38:52 EDT(-0400)] <EricDalquist> it has
[14:39:01 EDT(-0400)] <EricDalquist> whirly cache does not exist in 3.x
[14:39:19 EDT(-0400)] <holdorph> ok. but so help me, if I start seeing whilycache lines in my log again, i'm going to be very annoyed (smile)
[14:39:33 EDT(-0400)] <EricDalquist> lol
[14:39:57 EDT(-0400)] <chrisdoyle> shouldn't see them in up3 (smile)
[14:40:44 EDT(-0400)] <EricDalquist> unless you choose to switch back to whirlycache (tongue)
[14:40:52 EDT(-0400)] <chrisdoyle> lol
[14:41:18 EDT(-0400)] <holdorph> you're just taunting me now.
[14:41:19 EDT(-0400)] <chrisdoyle> i can't live without my ehcache / jconsole now...love it
[14:49:26 EDT(-0400)] <athena7> whirlycache went down the whirlydrain?
[14:49:34 EDT(-0400)] <EricDalquist> oh yeah
[14:49:46 EDT(-0400)] <athena7> (smile)
[14:49:48 EDT(-0400)] <athena7> i like ehcache
[14:49:54 EDT(-0400)] <athena7> and spring-modules-cache
[14:50:02 EDT(-0400)] <EricDalquist> (smile)
[14:50:21 EDT(-0400)] <EricDalquist> I'm just happy I got spring-modules-cache and jpa/hibernate using the same cachemanager
[14:50:33 EDT(-0400)] <EricDalquist> lots more flexibility that way and easier to config
[14:50:59 EDT(-0400)] <chrisdoyle> yes, yes
[15:04:39 EDT(-0400)] * travisennis (n=travisen@199.8.168.88) has joined ##uportal
[15:29:49 EDT(-0400)] <chrisdoyle> UP-2177 Remove unused Whirlycache log4j configurations - recorded in JIRA and committed to trunk and rel-3-0-patches
[15:37:33 EDT(-0400)] <athena7> EricDalquist, you get a chance to look at the trunk at all?
[16:15:39 EDT(-0400)] <EricDalquist> athena7: not yet :/
[16:15:47 EDT(-0400)] <EricDalquist> I'll see in another hour or so
[16:15:50 EDT(-0400)] <athena7> thanks
[16:16:21 EDT(-0400)] <athena7> i just want to see if we can get things set up with the new fluid trunk so they can test
[16:16:30 EDT(-0400)] <athena7> i appreciate the help (smile)
[16:18:23 EDT(-0400)] <EricDalquist> I may just end up committing the new dbloader and import scripts to fix it (tongue)
[16:19:20 EDT(-0400)] <athena7> woo!
[16:22:51 EDT(-0400)] * travisennis (n=travisen@199.8.168.88) has left ##uportal
[16:40:27 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[16:46:55 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[16:47:21 EDT(-0400)] <awills> so this is pretty funny: http://uportal.pastebin.com/m685033bf
[16:49:14 EDT(-0400)] <chrisdoyle> where did all that come from?
[16:49:25 EDT(-0400)] <awills> svn update
[17:13:37 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[17:15:11 EDT(-0400)] <awills> is HibernateDbLoader.java something that needs to be svn added? http://uportal.pastebin.com/m35780658
[17:15:27 EDT(-0400)] <EricDalquist> it was accidentally added
[17:15:37 EDT(-0400)] <chrisdoyle> gotta bolt, see yall lata
[17:15:38 EDT(-0400)] <EricDalquist> if you comment it out in persistenceContext.xml it will start
[17:15:45 EDT(-0400)] <EricDalquist> there is a problem with an import script though
[17:16:15 EDT(-0400)] <EricDalquist> what would people think of my just committing the hibernate dbloader and update import scripts before I head out tonight
[17:16:29 EDT(-0400)] <EricDalquist> the uP3 trunk export scripts still need work but the import ones all work correctly
[17:16:38 EDT(-0400)] <awills> cya chris
[17:16:52 EDT(-0400)] * chrisdoyle (n=chrisdoy@mtw160-1.ippl.jhu.edu) has left ##uportal
[17:17:05 EDT(-0400)] <EricDalquist> I'm afraid that I'm not going to find the time to fix what I screwed up in SVN until some time tomorrow otherwise :/
[17:17:32 EDT(-0400)] <holdorph> sounds more appropriate to roll the changes back in svn
[17:17:39 EDT(-0400)] <EricDalquist> that it does
[17:17:43 EDT(-0400)] <holdorph> rather then have something halfway done and/or broken
[17:17:49 EDT(-0400)] <EricDalquist> but I don't know which change broke it (sad)
[17:17:52 EDT(-0400)] <awills> we're talking trunk right? that should be fine... we should have time to sort out loose ends on 3.1 code
[17:17:57 EDT(-0400)] <EricDalquist> I do know what I have locally works
[17:18:18 EDT(-0400)] <EricDalquist> the problem is from me having local mods for 3 different things in one project
[17:18:22 EDT(-0400)] <EricDalquist> and trying to commit just one of them
[17:18:28 EDT(-0400)] <awills> yes, that is confusing
[17:18:44 EDT(-0400)] * holdorph bemoans yet another thing Perforce did better then svn
[17:18:56 EDT(-0400)] <EricDalquist> what's that? Have less stressed users?
[17:19:22 EDT(-0400)] <holdorph> no, you had to explicitly 'edit' before you could change a file. And when you did, you could mark which 'changelist' the edit was for
[17:19:36 EDT(-0400)] <holdorph> so you could have 3 'changelists' in one project, that were maintained separately
[17:19:41 EDT(-0400)] <EricDalquist> ah
[17:20:09 EDT(-0400)] <EricDalquist> what happened if one file was in two changesets?
[17:20:28 EDT(-0400)] <holdorph> then you'd be "no worse off" then you are in svn
[17:20:55 EDT(-0400)] <holdorph> at least when files didn't overlap, you could keep the changes next to each other, but separate
[17:21:06 EDT(-0400)] <EricDalquist> yeah, the problem here was an overlapping file
[17:21:19 EDT(-0400)] <EricDalquist> and I missed the other change in my compare before I made the commit
[17:21:29 EDT(-0400)] <EricDalquist> but there is some other problem with an import script
[17:21:34 EDT(-0400)] <EricDalquist> and I have no idea when it was introduced
[17:21:52 EDT(-0400)] <EricDalquist> nightly builds would be very handy (tongue)
[17:22:12 EDT(-0400)] <holdorph> well if you dont have an easy way to roll back, then 'less broken' is better then not
[17:23:21 EDT(-0400)] <athena7> less broken would be good
[17:23:33 EDT(-0400)] <EricDalquist> ok
[17:23:43 EDT(-0400)] <EricDalquist> I'll see if I can unbreak things before I head out tonight
[17:23:52 EDT(-0400)] <EricDalquist> and I'll send an update email to uportal-dev when doing so
[17:23:59 EDT(-0400)] <athena7> thanks
[17:24:04 EDT(-0400)] <EricDalquist> yup
[17:24:05 EDT(-0400)] <athena7> if you don't get to it, life will go on too though
[17:52:10 EDT(-0400)] <EricDalquist> ok ... just got the channel manager work done ... starting to commit things to hopefully fix trunk!
[17:56:03 EDT(-0400)] <athena7> oh nice! is this the multi-preferences stuff?
[17:56:21 EDT(-0400)] <EricDalquist> yeah
[17:56:25 EDT(-0400)] <EricDalquist> multi-pref support in the import
[17:56:28 EDT(-0400)] <EricDalquist> and in channel manager UI
[17:56:39 EDT(-0400)] <EricDalquist> I'll be including it in my set of commits over the next 20 minutes
[17:57:06 EDT(-0400)] * athena7 cheers
[17:57:12 EDT(-0400)] <athena7> that's so excellent (smile)
[17:57:49 EDT(-0400)] <EricDalquist> also creates a new 3.1 .channel file format
[17:57:55 EDT(-0400)] <EricDalquist> that supports something like:
[17:58:25 EDT(-0400)] <EricDalquist> <portletPreferences><preference><name/><readonly/><values><value/><value/></values>.....
[17:58:37 EDT(-0400)] <EricDalquist> no more use of the PORTLET. prefix (tongue)
[17:58:56 EDT(-0400)] <athena7> nice!
[18:00:04 EDT(-0400)] <athena7> that'll be less confusing
[18:00:09 EDT(-0400)] <EricDalquist> yeah
[18:00:15 EDT(-0400)] <EricDalquist> the channelmanager ui is ok
[18:00:18 EDT(-0400)] <EricDalquist> not really wonderful
[18:00:20 EDT(-0400)] <EricDalquist> but at least functional
[18:04:14 EDT(-0400)] <athena7> hey that sounds like a good place to start
[18:22:43 EDT(-0400)] <EricDalquist> ok athena7
[18:22:53 EDT(-0400)] <EricDalquist> all those changes are committed ....
[18:23:01 EDT(-0400)] <athena7> YAY
[18:23:03 EDT(-0400)] * athena7 cheers
[18:23:11 EDT(-0400)] <EricDalquist> just watch when you do the update
[18:23:14 EDT(-0400)] <EricDalquist> it is a lot of files
[18:23:25 EDT(-0400)] <athena7> ok, i'll do that tonight
[18:23:30 EDT(-0400)] <athena7> hopefully get the fluid js file updated
[18:23:32 EDT(-0400)] <EricDalquist> I just sent an email to the dev list explaining all of it
[18:23:45 EDT(-0400)] <EricDalquist> awills: UWs updated import scripts are included
[18:23:56 EDT(-0400)] <EricDalquist> I'll post the 2.x export package as soon as I can
[18:26:03 EDT(-0400)] <holdorph> i thought you said above you thought export might not be working with what you had so far, but you didn't say that in the email
[18:26:07 EDT(-0400)] <holdorph> is export working?
[18:26:37 EDT(-0400)] <EricDalquist> good catch, the export should work except for portlet preferences
[18:27:33 EDT(-0400)] <EricDalquist> actually no
[18:27:37 EDT(-0400)] <EricDalquist> they will work
[18:27:49 EDT(-0400)] <EricDalquist> because export in 3.0 uses the portlet APIs directly for those
[18:27:56 EDT(-0400)] <EricDalquist> and none of that changed
[18:28:06 EDT(-0400)] <EricDalquist> so export should still work
[18:28:41 EDT(-0400)] <EricDalquist> import just supports more data in the entity files than the current export writes out (namely dlm nodes and parameters in the layout)
[20:10:46 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[22:21:20 EDT(-0400)] * apetro-_ (n=apetro@828host77.starwoodbroadband.com) has joined ##uportal
[22:39:12 EDT(-0400)] * agherna (n=agherna@12-214-178-128.client.mchsi.com) has joined ##uportal
[22:41:23 EDT(-0400)] * agherna_ (n=agherna@vpn3-144145.near.uiuc.edu) has joined ##uportal