uPortal IRC Logs-2008-07-25

Owned by Apereo Infrastructure
Last updated: Jul 25, 2008Version comment

[08:17:31 EDT(-0400)] * athena7 (n=athena7@adsl-75-58-125-195.dsl.wlfrct.sbcglobal.net) has joined ##uportal
[08:48:51 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[09:53:44 EDT(-0400)] <dstn> helluva address for the hotel at JA-SIG unconference
[09:53:49 EDT(-0400)] <dstn> 666 Wisconsin Avenue
[09:57:29 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:00:28 EDT(-0400)] <athena7> lol
[10:00:51 EDT(-0400)] <athena7> dstn what're you guys's plans for exchange email in the portal?
[10:00:59 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has left ##uportal
[10:02:05 EDT(-0400)] <dstn> athena7: current plans are a small fix to determine preferred mailbox. Long term, move to mail portlet or other solution
[10:02:21 EDT(-0400)] <athena7> as far as CAS integration goes though?
[10:06:04 EDT(-0400)] <dstn> athena7: sorry, I misread
[10:06:18 EDT(-0400)] <athena7> no it's cool, that was a valid interpretation of my vague statement (smile)
[10:06:47 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[10:09:18 EDT(-0400)] <dstn> athena7: My understanding is exchange doesn't support CAS so I'm honestly not sure. Ideally we want to be able to have a mail portlet in the portal that pulls from exchange but I am unsure as to the details of that design at this point. There was discussion of using the web services of exchange but again, there is still the CAS issue.
[10:09:38 EDT(-0400)] <athena7> ah ok
[10:38:16 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[11:03:31 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:14:06 EDT(-0400)] <higpetter> damn... I did something, and now my form always makes "invalid submit"...
[11:15:14 EDT(-0400)] <higpetter> DEBUG [RegistrationsCertificateController] <Invalid submit detected: Form object not found in session (in session-form mode)>
[11:15:21 EDT(-0400)] <higpetter> where did my form object go then?
[12:31:48 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:17:34 EDT(-0400)] * KWhat_Work (n=abarker@abarker.callutheran.edu) has joined ##uportal
[13:20:28 EDT(-0400)] <KWhat_Work> anyone here?
[13:20:38 EDT(-0400)] <EricDalquist> mostly
[13:20:47 EDT(-0400)] <EricDalquist> it is Friday ... and technically the afternoon here (smile)
[13:21:19 EDT(-0400)] <KWhat_Work> haha yes i understand
[13:22:27 EDT(-0400)] <KWhat_Work> this is kind of a long shot and not even uportal related but have you ever seen this error: Caused by: java.lang.SecurityException: class "org.jaxen.XPath"'s signer information does not match signer information of other classes in the same package
[13:22:59 EDT(-0400)] <EricDalquist> no,
[13:23:03 EDT(-0400)] <EricDalquist> never played with signed jars
[13:23:06 EDT(-0400)] <EricDalquist> at least that I know of
[13:23:17 EDT(-0400)] <KWhat_Work> i would rather not but oracle is bending me over again
[13:23:23 EDT(-0400)] <EricDalquist> :/
[13:24:28 EDT(-0400)] <KWhat_Work> cant wait to get on uportal
[13:24:39 EDT(-0400)] <KWhat_Work> my stress and frustration levels going to drop like 90%
[13:24:43 EDT(-0400)] <EricDalquist> are you switching from oracle?
[13:24:49 EDT(-0400)] <KWhat_Work> yah
[13:24:55 EDT(-0400)] <EricDalquist> how's it going?
[13:25:03 EDT(-0400)] <KWhat_Work> we are still in discovery stage
[13:25:05 EDT(-0400)] <EricDalquist> ah
[13:25:24 EDT(-0400)] <KWhat_Work> its not going to be for at least 6 months but the higher ups agree that its the next step
[13:25:34 EDT(-0400)] <EricDalquist> that's good
[13:26:13 EDT(-0400)] <KWhat_Work> indeed
[13:30:57 EDT(-0400)] <dstn> grrrr
[13:31:46 EDT(-0400)] <dstn> quartz leaves running threads and then tomcat never shuts down
[13:32:02 EDT(-0400)] <EricDalquist> using it via spring?
[13:32:23 EDT(-0400)] <dstn> yea
[13:32:36 EDT(-0400)] <dstn> I set destroy-method="destroy" and it still doesn't quit
[13:32:49 EDT(-0400)] <dstn> very frustrating
[13:33:14 EDT(-0400)] <EricDalquist> what are the threads doing?
[13:33:20 EDT(-0400)] <EricDalquist> are they actually running
[13:33:23 EDT(-0400)] <EricDalquist> or just waiting?
[13:34:24 EDT(-0400)] * dstn does not know much at all about threads
[13:34:54 EDT(-0400)] <dstn> I know its quartz though cause when I remove it, tomcat shuts down fine
[13:34:54 EDT(-0400)] <EricDalquist> what OS are you on?
[13:35:01 EDT(-0400)] <dstn> ubuntu
[13:35:08 EDT(-0400)] <EricDalquist> with quartz enabled
[13:35:13 EDT(-0400)] <EricDalquist> shutdown tomcat
[13:35:22 EDT(-0400)] <EricDalquist> when it seems to be 'done' but is still running
[13:35:29 EDT(-0400)] <EricDalquist> issue a 'kill -3 PID'
[13:35:42 EDT(-0400)] <dstn> k, let me try, one sec
[13:35:45 EDT(-0400)] <EricDalquist> that will dump the current stack traces of all active threads to stdout
[13:35:47 EDT(-0400)] <EricDalquist> so catalina.out
[13:35:54 EDT(-0400)] <dstn> ahhhh
[13:36:07 EDT(-0400)] <EricDalquist> the other option is to connect to the JVM via jconsole
[13:36:08 EDT(-0400)] <dstn> I saw that kill -3 command in the mist of my searches...
[13:36:19 EDT(-0400)] <EricDalquist> and the threads view lets you see the stack of any thread
[13:38:32 EDT(-0400)] <dstn> whoa, lots of output
[13:38:45 EDT(-0400)] <EricDalquist> yup
[13:40:49 EDT(-0400)] <dstn> all quartz threads are TIMED_WAITING
[13:41:09 EDT(-0400)] <EricDalquist> strange
[13:41:17 EDT(-0400)] <athena7> argh, i refactored all this stuff in the duke mail portlet and never checked it in (sad)
[13:42:08 EDT(-0400)] <dstn> bare with me here, what is TIMED_WAITING?
[13:42:37 EDT(-0400)] <EricDalquist> can you pastebin the entire stack trace for that thread?
[13:42:45 EDT(-0400)] <dstn> ya, sure one sec
[13:46:13 EDT(-0400)] <dstn> http://uportal.pastebin.com/d1b0adde5
[13:47:13 EDT(-0400)] <EricDalquist> so TIMED_WAIT is a result of a Thread asking to sleep for some specified ammount of time
[13:47:35 EDT(-0400)] <dstn> ok, that makes sense
[13:48:52 EDT(-0400)] <EricDalquist> nothing obvious in there as to why quartz is stopping shutdown
[13:49:46 EDT(-0400)] <dstn> well shoot
[13:50:00 EDT(-0400)] * dstn learn something though
[13:50:03 EDT(-0400)] <dstn> learned*
[14:55:26 EDT(-0400)] <dstn> YES!
[14:55:36 EDT(-0400)] <dstn> found a solution (smile)
[14:55:41 EDT(-0400)] <EricDalquist> what was the problem?
[14:56:16 EDT(-0400)] <dstn> if I make them daemon threads they shutdown with tomcat
[14:56:21 EDT(-0400)] <dstn> org.quartz.scheduler.makeSchedulerThreadDaemon = true
[14:56:28 EDT(-0400)] <dstn> org.quartz.threadPool.makeThreadsDaemons = true
[14:56:53 EDT(-0400)] <EricDalquist> ah
[14:57:00 EDT(-0400)] <EricDalquist> some properties that need to be set?
[14:57:01 EDT(-0400)] <dstn> mind me that I don't know much about threads but the daemon threads shutdown with the jvm
[14:57:17 EDT(-0400)] <dstn> ya
[14:58:17 EDT(-0400)] <EricDalquist> so the JVM assumes daemon threads can just be stopped when the JVM is being shut down
[14:58:27 EDT(-0400)] <EricDalquist> it will wait for non-daemon threads
[14:58:33 EDT(-0400)] <EricDalquist> just a marker flag I believe
[15:00:04 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[15:07:33 EDT(-0400)] <dstn> ya, that's what I was reading
[15:44:56 EDT(-0400)] <dstn> eeek
[15:44:59 EDT(-0400)] <dstn> java.lang.UnsupportedClassVersionError: Bad version number in .class file
[15:45:29 EDT(-0400)] <EricDalquist> something compiled for a newer jdk than you're running on
[15:46:41 EDT(-0400)] <dstn> must be a friday
[15:46:44 EDT(-0400)] <dstn> I'm error prone
[15:46:46 EDT(-0400)] <EricDalquist> (smile)
[15:48:00 EDT(-0400)] <EricDalquist> still on ubuntu?
[15:48:09 EDT(-0400)] <dstn> do you know how to look at a class and see what version it was compiled with?
[15:48:13 EDT(-0400)] <dstn> yes...
[15:48:49 EDT(-0400)] <EricDalquist> the exception should have given you a #
[15:48:52 EDT(-0400)] <EricDalquist> liker 49.0
[16:21:37 EDT(-0400)] <EricDalquist> so her is a fun one
[16:21:40 EDT(-0400)] <dstn> misconfiguration on my part...eclipse was set to 6.0 and was auto building so on ant compile everything was already built
[16:21:52 EDT(-0400)] <EricDalquist> we're going to need a way to flag portlets as being part of uPortal
[16:22:04 EDT(-0400)] <EricDalquist> or make people re-configure the applicationId for all the included portlets
[16:22:14 EDT(-0400)] <EricDalquist> since people deploy to many names other than uPortal
[16:22:36 EDT(-0400)] * lennar1 (n=sparhk@183.sub-75-211-132.myvzw.com) has joined ##uportal
[16:25:03 EDT(-0400)] <dstn> not sure i understand ya
[16:25:21 EDT(-0400)] * dstn doesn't know up3 yet
[16:25:39 EDT(-0400)] <EricDalquist> so in up3.1 we're starting to include admin tools written as portlets
[16:25:47 EDT(-0400)] <EricDalquist> so they exist in the uPortal webapp
[16:26:04 EDT(-0400)] <EricDalquist> when you deploy portlets to uportal you specifiy the webapp name and the portlet name
[16:26:18 EDT(-0400)] <EricDalquist> so we don't want to just hardcode uPortal as the webapp name for these
[16:26:31 EDT(-0400)] <EricDalquist> since people deploy uPortal under a variety of webapp names
[16:26:32 EDT(-0400)] <dstn> ahhh
[16:26:39 EDT(-0400)] <EricDalquist> so we need to detect the current webapp name and use that
[16:26:46 EDT(-0400)] <EricDalquist> so I think uP3.1 is going to require servlet 2.5
[16:26:55 EDT(-0400)] <EricDalquist> which allows us to find out the current servlet context nam,e
[16:27:04 EDT(-0400)] <EricDalquist> but still need to add a flag in channel manager or soemthing
[16:27:37 EDT(-0400)] <dstn> so only 2.5 allows that...dang
[16:27:50 EDT(-0400)] <EricDalquist> yup
[16:28:01 EDT(-0400)] <EricDalquist> so Tomcat 6 for uP 3.1
[16:28:03 EDT(-0400)] <EricDalquist> not a big deal
[16:28:09 EDT(-0400)] <EricDalquist> or at least it shouldn't be
[16:35:08 EDT(-0400)] <dstn> I've been wanting to look at 6 for a while
[16:36:26 EDT(-0400)] <EricDalquist> not much different than TC5
[16:36:37 EDT(-0400)] <EricDalquist> they disable the shared classloader in the default download which is annoying
[16:36:52 EDT(-0400)] <EricDalquist> but just a one liner in catalina.properties to re-enable it
[16:37:14 EDT(-0400)] <dstn> hmm, wonder why they did that
[16:37:21 EDT(-0400)] <dstn> are they moving away from that or what?
[16:37:37 EDT(-0400)] <EricDalquist> they don't have anything in there by default
[16:37:49 EDT(-0400)] <EricDalquist> and it speeds things up a bit to have one less classloader to look through
[16:37:58 EDT(-0400)] <EricDalquist> and most webapps don't need it
[16:38:04 EDT(-0400)] <dstn> ah
[16:38:41 EDT(-0400)] <dstn> shared would be something like portlet-api right?
[16:39:28 EDT(-0400)] <EricDalquist> classes in shared are accessible by all webapps but not the container
[16:39:43 EDT(-0400)] <EricDalquist> but in good practice you only put things there that you only want one copy of
[16:39:55 EDT(-0400)] <EricDalquist> or that need to share static data between all webapps
[16:40:07 EDT(-0400)] <EricDalquist> so the pluto container and descriptor parsing code goes there
[16:40:10 EDT(-0400)] <EricDalquist> since all portlets need it
[16:40:16 EDT(-0400)] <EricDalquist> and they all need to access the same instance of it
[16:40:19 EDT(-0400)] <dstn> right ok, that makes sense
[16:40:23 EDT(-0400)] <EricDalquist> portlet-api too
[16:56:59 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has left ##uportal
[18:20:26 EDT(-0400)] * EricDalquist loves this new machine ... 'ant clean initportal' is only a 2 minute task now (big grin)
[18:23:00 EDT(-0400)] <lennar1> lol
[18:23:18 EDT(-0400)] <EricDalquist> important when you're doing work with pluto bug-fixes
[18:23:31 EDT(-0400)] <lennar1> careful... the uPortal world may suffering longer waits due to eric's uber laptop(smile)
[18:23:47 EDT(-0400)] <EricDalquist> lol
[18:23:59 EDT(-0400)] <EricDalquist> (and its just a desktop)
[18:24:04 EDT(-0400)] <lennar1> just knowing how fast it could run makes my wait appear longer(smile)
[18:24:30 EDT(-0400)] <EricDalquist> one of our todos in our upgrade is actually going to be crn import/export performance
[18:24:39 EDT(-0400)] <EricDalquist> which talking to drew there is a bit of work to be done
[18:24:49 EDT(-0400)] <EricDalquist> and could save 30+ seconds on the default init
[18:25:00 EDT(-0400)] <EricDalquist> and much more when we have to migrate 80k users (tongue)
[18:25:49 EDT(-0400)] <lennar1> (smile)
[18:26:31 EDT(-0400)] <lennar1> hmm... one wonders how long it will take to migrate Pearson's nearly 2 million users.
[18:26:48 EDT(-0400)] <EricDalquist> 'a while'
[18:26:51 EDT(-0400)] <lennar1> (smile)
[18:26:52 EDT(-0400)] <EricDalquist> you can quote me on that (tongue)
[18:26:56 EDT(-0400)] <lennar1> sweet
[18:27:25 EDT(-0400)] <EricDalquist> anyone here familiar with the mysteries of CPDs?
[18:27:57 EDT(-0400)] * lennar1 made a very small change about a year ago
[18:28:01 EDT(-0400)] <lennar1> doubt that counts(sad)
[18:28:28 EDT(-0400)] <EricDalquist> I'm trying to figure out how to add a new checkbox to the portlet publishing workflow
[18:28:31 EDT(-0400)] <EricDalquist> I kind of have it working
[18:28:42 EDT(-0400)] <EricDalquist> I'm having to work backwards from the XSL that renders the thing
[18:29:39 EDT(-0400)] * lennar1 added a new checkbox to his portlet publishing workflow for stripping the chrome off of a portlet
[18:29:50 EDT(-0400)] <EricDalquist> oohh
[18:30:03 EDT(-0400)] <lennar1> I just looked around in the CPDs for an example of a checkbox... then added it in... and... was done
[18:30:04 EDT(-0400)] <EricDalquist> do you happen to have the cpd still?
[18:30:12 EDT(-0400)] <lennar1> sure
[18:30:19 EDT(-0400)] * lennar1 goes to look
[18:30:36 EDT(-0400)] <lennar1> was in 2.6.1
[18:30:44 EDT(-0400)] <EricDalquist> I didn't see an example ...
[18:30:48 EDT(-0400)] <EricDalquist> maybe I looked in the wrong place
[18:30:53 EDT(-0400)] * EricDalquist runs search again
[18:32:48 EDT(-0400)] <lennar1> sorry... am at the NFJS conference and am trying to appear like I am paying attention
[18:32:52 EDT(-0400)] <lennar1> this guy seems touchy
[18:32:57 EDT(-0400)] <EricDalquist> lol
[18:41:12 EDT(-0400)] <lennar1> sorry still looking
[18:41:17 EDT(-0400)] <lennar1> I know I did it
[18:41:27 EDT(-0400)] <EricDalquist> (smile)
[18:41:30 EDT(-0400)] <EricDalquist> I have it showing up
[18:41:41 EDT(-0400)] <EricDalquist> but there is no value in the name attribute of the <input>
[18:41:46 EDT(-0400)] <EricDalquist> which is rather useless
[18:46:42 EDT(-0400)] <lennar1> found it
[18:47:41 EDT(-0400)] <EricDalquist> yay!
[18:47:45 EDT(-0400)] <lennar1> <parameter modify="publish-only">
[18:47:46 EDT(-0400)] <lennar1> <name>hideChrome</name>
[18:47:46 EDT(-0400)] <lennar1> <label>Hide portlet chrome</label>
[18:47:46 EDT(-0400)] <lennar1> <type base="string" input="single-choice" display="drop-down">
[18:47:46 EDT(-0400)] <lennar1> <restriction type="enumeration">
[18:47:46 EDT(-0400)] <lennar1> <value>true</value>
[18:47:48 EDT(-0400)] <lennar1> <value>false</value>
[18:47:50 EDT(-0400)] <lennar1> <defaultValue>false</defaultValue>
[18:47:52 EDT(-0400)] <lennar1> </restriction>
[18:47:54 EDT(-0400)] <lennar1> </type>
[18:47:56 EDT(-0400)] <lennar1> <description>
[18:47:58 EDT(-0400)] <lennar1> True hides the portlets chrome in the layout.
[18:48:00 EDT(-0400)] <lennar1> False allows the chrome to be displayed as normal
[18:48:02 EDT(-0400)] <lennar1> </description>
[18:48:04 EDT(-0400)] <lennar1> </parameter>
[18:48:05 EDT(-0400)] <EricDalquist> ah you used a drop-down
[18:48:18 EDT(-0400)] <lennar1> sorry... thought it was a checkbox(sad)
[18:48:21 EDT(-0400)] <EricDalquist> I'll have to give that a try and see if I have more luck
[18:48:22 EDT(-0400)] <EricDalquist> np
[18:48:30 EDT(-0400)] <EricDalquist> I think there may be a bug in the XSL (tongue)
[18:48:39 EDT(-0400)] <lennar1> perish the thought!
[18:50:24 EDT(-0400)] <EricDalquist> well I'm out
[18:50:26 EDT(-0400)] <EricDalquist> later all
[21:41:53 EDT(-0400)] * lennar1 (n=sparhk@183.sub-75-211-132.myvzw.com) has left ##uportal