Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 26 Next »

[09:19:50 EDT(-0400)] * awills (n=awills@mtw160-5.ippl.jhu.edu) has joined ##uportal
[09:36:32 EDT(-0400)] * grimesp (n=grimesp@134.250.4.177) has joined ##uportal
[09:57:50 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[09:58:10 EDT(-0400)] * bulloche (n=bulloche@134.250.4.77) has joined ##uportal
[10:09:06 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[10:12:42 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[10:31:02 EDT(-0400)] <EricDalquist> anyone here like coming up with method names?
[10:31:50 EDT(-0400)] <EricDalquist> I need to figure out appropriate names to differentiate these two methods:
[10:31:51 EDT(-0400)] <EricDalquist> public Set<IPerson> getPeople(Map<String, Object> query);
[10:31:51 EDT(-0400)] <EricDalquist> public Set<IPerson> getPeople(Map<String, List<Object>> query);
[10:35:33 EDT(-0400)] <EricDalquist> my best idea so far is:
[10:35:34 EDT(-0400)] <EricDalquist> public Set<IPerson> getPeople(Map<String, Object> query);
[10:35:34 EDT(-0400)] <EricDalquist> public Set<IPerson> getPeople(Map<String, List<Object>> query);
[10:37:50 EDT(-0400)] <higpetter> EricDalquist: did I meet you in denver a year ago?
[10:38:05 EDT(-0400)] <EricDalquist> you may have, I was there
[10:38:20 EDT(-0400)] <higpetter> I spoke a lot with Eric Olsson at least
[10:38:47 EDT(-0400)] <higpetter> but now I have problems with portlets that start some times, but not other times...
[10:39:05 EDT(-0400)] <higpetter> I guess they need some classes which some times already has been loaded
[10:39:16 EDT(-0400)] <EricDalquist> that sounds weird
[10:39:22 EDT(-0400)] <EricDalquist> classloading should be pretty consistent
[10:39:34 EDT(-0400)] <EricDalquist> since that is all handled by tomcat
[10:39:54 EDT(-0400)] <higpetter> like I haven't supplied the needed jars, but they do exist in other portlets
[10:40:01 EDT(-0400)] <higpetter> so it depends on what order they are started
[10:40:28 EDT(-0400)] <EricDalquist> well each portlet webapp is in its own classloader
[10:40:39 EDT(-0400)] <higpetter> ....
[10:40:44 EDT(-0400)] <higpetter> then I have no clue
[10:40:49 EDT(-0400)] <EricDalquist> you cannot access classes in webapp A from webapp B
[10:40:56 EDT(-0400)] <EricDalquist> that's part of the servlet spec
[10:42:48 EDT(-0400)] <EricDalquist> do you have stack traces from the problem you can share?
[10:43:45 EDT(-0400)] <higpetter> nope... when they don't start, they don't log anything either... I suspect the logging is the problem
[10:44:26 EDT(-0400)] <EricDalquist> are you sure each portlet WAR includes ALL of the jars it depends upon?
[10:44:38 EDT(-0400)] <higpetter> <higpetter> like I haven't supplied the needed jars, but they do exist in other portlets
[10:45:11 EDT(-0400)] <dstn> getPeopleWithStringObjectMap, getPeopleWithStringListObjectMap ... LOL
[10:45:25 EDT(-0400)] <EricDalquist> thanks dstn (tongue)
[10:45:26 EDT(-0400)] <higpetter> so no, I'm not sure about anything (wink)
[10:46:33 EDT(-0400)] <higpetter> EricDalquist: eclipse finds all classes, if that's a good start
[10:47:08 EDT(-0400)] <EricDalquist> that is
[10:47:17 EDT(-0400)] <EricDalquist> perhaps try one portlet app at a time
[10:47:28 EDT(-0400)] <EricDalquist> and see if each loads with no others (other than the portal) in tomcat
[10:47:36 EDT(-0400)] <EricDalquist> watch the tomcat logs for clues
[10:47:52 EDT(-0400)] <EricDalquist> also have you put anything in shared/lib or common/lib beyond what uPortal puts there?
[10:48:45 EDT(-0400)] <higpetter> only mysql-connector..
[10:50:58 EDT(-0400)] <higpetter> SEVERE: Error listenerStart
[10:50:58 EDT(-0400)] <higpetter> Jul 10, 2008 4:50:28 PM org.apache.catalina.core.StandardContext start
[10:50:58 EDT(-0400)] <higpetter> SEVERE: Context [/HIG-uPortalNewsPortletApp] startup failed due to previous errors
[10:51:07 EDT(-0400)] <higpetter> and nothing in the portlets own lof
[10:51:08 EDT(-0400)] <higpetter> log
[10:51:21 EDT(-0400)] <EricDalquist> huh
[10:51:31 EDT(-0400)] <EricDalquist> at that point I would be checking to ensure the portlet's logging config is correct
[10:51:34 EDT(-0400)] <EricDalquist> and turn logging way up
[10:51:48 EDT(-0400)] <higpetter> it's all the way up...
[10:52:03 EDT(-0400)] <higpetter> I think it the logging listener that fails
[10:53:19 EDT(-0400)] <higpetter> hmm... commons-logging.jar mishmash
[11:01:00 EDT(-0400)] * athena7 (n=athena7@128.187.217.157) has joined ##uportal
[11:03:31 EDT(-0400)] <awills> higpetter it sounds like something that can happen when the code is compiled against one version of commons-logging but a different version is present at runtime
[11:03:58 EDT(-0400)] <EricDalquist> on a logging related note anyone else here ever tried slf4j?
[11:04:11 EDT(-0400)] <EricDalquist> http://www.slf4j.org/
[11:04:14 EDT(-0400)] <awills> and you're right – an "Error listenerStart" error means that a listener in web.xml couldn't start
[11:04:25 EDT(-0400)] <EricDalquist> I've used it a few times and seems much more sane than commons-logging
[11:05:56 EDT(-0400)] <awills> that's cool... is it "more sane" than log4j config as well? (wink)
[11:06:30 EDT(-0400)] <EricDalquist> I haven't actually used the underlying logger included with it
[11:06:37 EDT(-0400)] <EricDalquist> just slf4j -> log4j
[11:06:59 EDT(-0400)] <EricDalquist> oh the other nice thing is no longer needing to do if (log.isDebugEnabled()) type stuff
[11:07:17 EDT(-0400)] <EricDalquist> logger.debug("Temperature set to {}. Old temperature was {}.", t, oldT);
[11:07:35 EDT(-0400)] <EricDalquist> then slf4j does the check and only does the toStrings and string replacement if debug is enabled
[11:07:50 EDT(-0400)] <awills> that's handy
[11:08:08 EDT(-0400)] <awills> you'd still need isDebugEnabled() for looping
[11:08:39 EDT(-0400)] <EricDalquist> yeah if you're logging something really complex
[11:08:54 EDT(-0400)] <awills> only occasionally (tongue)
[11:11:00 EDT(-0400)] <dstn> so I ran release:perform and I don't see anywhere where it runs site:deploy ... do I have to run site:site prior?
[11:11:24 EDT(-0400)] <EricDalquist> oh ... maybe you have to do site manually
[11:11:27 EDT(-0400)] <EricDalquist> let me check my notes
[11:12:33 EDT(-0400)] <dstn> I'm reading that the plugin's default goals include site:deploy
[11:12:49 EDT(-0400)] <EricDalquist> do you have a site deployment location setup in your pom?
[11:13:04 EDT(-0400)] <dstn> but I ran site:deploy on its own and it says site:site has to be run first
[11:13:18 EDT(-0400)] <EricDalquist> really? I've never had to do that
[11:13:21 EDT(-0400)] <dstn> I have the distributionMangement section set up, is there another part?
[11:14:26 EDT(-0400)] <EricDalquist> https://www.ja-sig.org/svn/jasig-parent/trunk/pom.xml

  • No labels