uPortal IRC Logs-2008-09-16

[08:52:11 EDT(-0400)] * athena7 (n=athena7@adsl-99-149-83-32.dsl.wlfrct.sbcglobal.net) has joined ##uportal
[09:12:51 EDT(-0400)] * anastasiac (n=team@142.150.154.160) has joined ##uportal
[09:15:12 EDT(-0400)] * colinclark (n=colin@bas1-toronto09-1279544154.dsl.bell.ca) has joined ##uportal
[09:24:21 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[10:28:57 EDT(-0400)] * michelled (n=team@142.150.154.197) has left ##uportal
[10:51:43 EDT(-0400)] <athena7> help i can't stop playing with the new gallery widget in confluence
[10:58:49 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:19:45 EDT(-0400)] * holdorph (n=holdorph@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[11:22:50 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[11:45:40 EDT(-0400)] * michelled (n=team@142.150.154.197) has joined ##uportal
[11:50:57 EDT(-0400)] <EricDalquist> athena7: SVN is fixed
[11:51:14 EDT(-0400)] <athena7> thanks eric!
[11:51:39 EDT(-0400)] <athena7> i was doing a merge and thought maybe that didn't seem quite right
[11:51:45 EDT(-0400)] <EricDalquist> yeah
[11:51:49 EDT(-0400)] <athena7> congrats on the release, that's awesome
[11:51:54 EDT(-0400)] <EricDalquist> I forgot to switch my local copy back before doing the update
[11:52:17 EDT(-0400)] <EricDalquist> svn merge -r newrev:oldrev .
[11:52:26 EDT(-0400)] <EricDalquist> reverts the changes between new and old
[11:52:38 EDT(-0400)] <EricDalquist> in case you ever need to do that
[11:52:48 EDT(-0400)] <EricDalquist> thanks
[11:56:02 EDT(-0400)] <athena7> ah ok, so i could use that to un-update the vendor/current branch?
[11:56:15 EDT(-0400)] <EricDalquist> well a svn update will work now
[11:56:27 EDT(-0400)] <EricDalquist> I did that then a svn commit to 'revert' the bad commit
[11:56:39 EDT(-0400)] <athena7> ah
[11:57:04 EDT(-0400)] <athena7> i think my choices are either to revert the update i did and run it again, or to just import in a release-v2
[11:57:07 EDT(-0400)] <athena7> that sound right?
[11:57:46 EDT(-0400)] <athena7> by the way, i think we're int he process of getting the search portlet all set up
[11:57:49 EDT(-0400)] <EricDalquist> oh so you already did an import ... got ya
[11:58:01 EDT(-0400)] <athena7> once i get it committed i'll send a note to the lists
[11:58:05 EDT(-0400)] <EricDalquist> at that point I'd probably do a new svn export of the jasig repo
[11:58:06 EDT(-0400)] <athena7> yeah - just the vendor side
[11:58:08 EDT(-0400)] <athena7> yeah
[11:58:10 EDT(-0400)] <athena7> makes sense
[11:58:11 EDT(-0400)] <EricDalquist> then do a new vendor update
[11:58:42 EDT(-0400)] <athena7> http://www.ja-sig.org/wiki/display/PLT/TabbedSearchPortlet
[11:58:46 EDT(-0400)] <athena7> the new gallery widget is neat
[11:59:09 EDT(-0400)] <EricDalquist> oh yeah it is
[12:03:12 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[12:57:09 EDT(-0400)] <athena7> i'm seeing a bunch of org.jasig.portal.PortalException: java.lang.Exception: Unable to find User Profile for user 13 and profile 1
[12:57:16 EDT(-0400)] <athena7> where user 13 is the system user
[12:57:21 EDT(-0400)] <athena7> any ideas on what would cause that?
[12:57:34 EDT(-0400)] <athena7> we're having problems with the guest user not picking up content from DLM correctly
[12:57:44 EDT(-0400)] <EricDalquist> hrm
[12:57:49 EDT(-0400)] <EricDalquist> this is post 3.0.2 upgrade?
[12:58:10 EDT(-0400)] <athena7> no
[12:58:11 EDT(-0400)] <athena7> 3.0.1
[12:58:40 EDT(-0400)] <EricDalquist> so this is 3.0.1 with the default data?
[12:58:42 EDT(-0400)] <athena7> i think they tried to update one of the fragments the guest user includes, then re-import the fragments back into the database
[12:58:49 EDT(-0400)] <athena7> customized layouts
[12:58:55 EDT(-0400)] <athena7> worked well for a while though
[12:59:07 EDT(-0400)] <athena7> i'm wondering if something got corrupted in one of the imports, maybe
[12:59:10 EDT(-0400)] <EricDalquist> without a full description of what they've done I'm not sure
[12:59:21 EDT(-0400)] <athena7> yeah me neither
[12:59:39 EDT(-0400)] <athena7> i'm not really clear enough of the system user's role to know if that error is likely part of the problem though
[12:59:59 EDT(-0400)] <EricDalquist> it is used as the absolute backup
[13:00:08 EDT(-0400)] <EricDalquist> so if the user and default user don't have enough info
[13:00:13 EDT(-0400)] <EricDalquist> the 'system' user is looked at
[13:00:27 EDT(-0400)] <athena7> ok
[13:00:39 EDT(-0400)] <athena7> that's helpful, thanks
[13:02:07 EDT(-0400)] <EricDalquist> I always end up doing a local 'default' import
[13:02:21 EDT(-0400)] <EricDalquist> and looking at how the system user -> profile -> layout stuff look
[13:02:55 EDT(-0400)] <athena7> default import?
[13:03:32 EDT(-0400)] <EricDalquist> so load up a 3.0.1 quickstart
[13:03:35 EDT(-0400)] <EricDalquist> and look at the database
[13:03:51 EDT(-0400)] <EricDalquist> see what the system user looks like in up_user, up_user_profile, etc...
[13:04:01 EDT(-0400)] <EricDalquist> crappy way to have to do it
[13:04:09 EDT(-0400)] <EricDalquist> but if you need a quick fix that's usually the easiest
[13:04:49 EDT(-0400)] <athena7> ah
[13:21:06 EDT(-0400)] * EricDalquist (n=dalquist@bohemia.doit.wisc.edu) has joined ##uportal
[13:23:45 EDT(-0400)] <athena7> happen to remember how and why user profiles get updated to profile_id != 1?
[13:24:04 EDT(-0400)] <EricDalquist> nope
[13:32:10 EDT(-0400)] * MarkRogers (n=MarkRoge@addhcp129.cc.umanitoba.ca) has joined ##uportal
[13:54:22 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[13:55:37 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[13:55:57 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[14:13:40 EDT(-0400)] <awills> are you there EricDalquist?
[14:13:45 EDT(-0400)] <EricDalquist> yup
[14:13:50 EDT(-0400)] <awills> great
[14:14:14 EDT(-0400)] <awills> i'd like to work on UP-2134 if possible: http://www.ja-sig.org/issues/browse/UP-2134
[14:14:36 EDT(-0400)] <awills> is there anything standing in the way of a WPP release at this point?
[14:14:39 EDT(-0400)] <EricDalquist> ok
[14:14:44 EDT(-0400)] <EricDalquist> um
[14:14:54 EDT(-0400)] <EricDalquist> would have time to sort through a diff?
[14:15:06 EDT(-0400)] <EricDalquist> I can create one of our current WPP code to the last code-dump we did
[14:15:10 EDT(-0400)] <EricDalquist> so you can get any other bug-fixes
[14:15:21 EDT(-0400)] <EricDalquist> I know there have been at least a few
[14:15:36 EDT(-0400)] <awills> if you can send it, i'd be happy to sort through it
[14:15:45 EDT(-0400)] <EricDalquist> ok
[14:15:53 EDT(-0400)] <EricDalquist> I'll email it to the portlets list
[14:16:00 EDT(-0400)] <EricDalquist> other than that it should be ready for a 1.0
[14:16:07 EDT(-0400)] <awills> k, cool
[14:20:36 EDT(-0400)] <athena7> i'm seeing something kind of strange in uportal 3.0.1 with doing layout updates
[14:21:04 EDT(-0400)] <EricDalquist> awills: ok sent
[14:21:15 EDT(-0400)] <EricDalquist> if you have questions about specific changes I can probably answer them
[14:21:21 EDT(-0400)] <EricDalquist> athena7: what is the issue?
[14:21:27 EDT(-0400)] <athena7> now that everything's working again
[14:21:45 EDT(-0400)] <athena7> we tried removing the google search portlet from the fragment-layout file we had, which is imported by the guest user
[14:21:59 EDT(-0400)] <athena7> after importing it via the import/export tools and restarting
[14:22:22 EDT(-0400)] <athena7> the google portlet doesn't show up if you log in as the fragment owner
[14:22:29 EDT(-0400)] <athena7> but it does show up in the unauthenticated guest view
[14:22:40 EDT(-0400)] <EricDalquist> hrm
[14:23:16 EDT(-0400)] <awills> thanks eric
[14:23:42 EDT(-0400)] <EricDalquist> awills: no problem
[14:23:51 EDT(-0400)] <athena7> i guess i need to find a vanilla instance and test this to see if it happens in other environments
[14:24:43 EDT(-0400)] <awills> can you do an initportal and see if the portlet still shows up? i wonder if it had been moved in the unauthenmticated guest layout?
[14:28:15 EDT(-0400)] <athena7> i don't think it could have been . . .
[14:28:33 EDT(-0400)] <athena7> it's using the guest evaluator, so it only shows up when you're actually a guest, not when the username equals "guest"
[14:28:40 EDT(-0400)] <awills> yeah, i thought they were locked-down
[14:28:44 EDT(-0400)] <athena7> so when i log in as the user "guest", the fragment doesn't appear
[14:29:04 EDT(-0400)] <athena7> that was my first thought too, but i can't think how it would have ever been modified
[14:29:09 EDT(-0400)] <awills> so what user are you when you see it?
[14:29:34 EDT(-0400)] <athena7> unauthenticated guest
[14:29:40 EDT(-0400)] <athena7> (as opposed to logged-in-as-guest)
[14:30:20 EDT(-0400)] <awills> it's still the same logical user, though, right? the same rows on UP_LAYOUT_STRUCT?
[14:30:24 EDT(-0400)] <athena7> right
[14:30:41 EDT(-0400)] <awills> was the portal running the whole time all this was done?
[14:30:41 EDT(-0400)] <athena7> but potentially different DLM fragment inclusions
[14:30:46 EDT(-0400)] <athena7> don't know
[14:31:50 EDT(-0400)] <awills> yeah potentially... and is that true in this case? does the guest user account return true on the GuestUserEvaluator?
[14:32:30 EDT(-0400)] <athena7> from the behavior i'm seeing, it looks like it doesn't
[14:32:35 EDT(-0400)] <athena7> which is sort of confusing
[14:33:34 EDT(-0400)] * awills agrees
[14:33:48 EDT(-0400)] <awills> but it's not a total surprise
[14:34:02 EDT(-0400)] <athena7> yeah
[14:34:23 EDT(-0400)] <athena7> trying this out now on the uportal trunk
[14:34:32 EDT(-0400)] <athena7> should really be trying on up 3.0.1, but i'd need to find one
[14:34:44 EDT(-0400)] <awills> is there a chance the unauthenticated layout is cached from before the change? can you try restarting to see if it updates?
[14:35:08 EDT(-0400)] <athena7> well, that worked
[14:35:16 EDT(-0400)] <athena7> they restarted a whole bunch of times
[14:35:35 EDT(-0400)] <awills> what worked?
[14:35:45 EDT(-0400)] <athena7> removing the channel from a fragment for a guest
[14:36:09 EDT(-0400)] <awills> ah, from the guest-lo layout you mean?
[14:36:19 EDT(-0400)] <athena7> yeah
[14:38:56 EDT(-0400)] <athena7> ok, interesting
[14:39:09 EDT(-0400)] <athena7> the guests fragment shows up while logged in as guest in up3 trunk
[14:39:21 EDT(-0400)] <athena7> so there may be something still horked in this database
[14:39:22 EDT(-0400)] <athena7> ugh.
[14:48:32 EDT(-0400)] <awills> athena7 that's surprising
[14:48:57 EDT(-0400)] <athena7> i think we've decided to give up and do an ant initdb and hope it doesn't happen again
[14:49:24 EDT(-0400)] <awills> http://uportal.pastebin.com/m7742aa5
[14:50:02 EDT(-0400)] <awills> the logged-in guest user shouldn't get the frag unless its SecurityContext returns false for isAuthenticated()
[14:50:36 EDT(-0400)] <athena7> interesting
[14:51:11 EDT(-0400)] <athena7> which class is this in again?
[14:51:29 EDT(-0400)] <awills> org.jasig.portal.security.provider.PersonImpl
[14:51:49 EDT(-0400)] * awills hopes that's the right impl for the logged-in guest user
[14:52:12 EDT(-0400)] <athena7> good question, actually
[14:52:38 EDT(-0400)] <awills> it really should be... i can't think how it wouldn't
[14:53:16 EDT(-0400)] <athena7> yeah me neither
[14:53:25 EDT(-0400)] <athena7> wow, i have no idea how i'm getting this fragment then
[14:59:32 EDT(-0400)] <athena7> uhoh, i see what happened.
[14:59:52 EDT(-0400)] <athena7> the guest layout is cached and shared between the logged in and not logged in versions
[15:00:25 EDT(-0400)] <athena7> so if you log in as guest before hitting the unauthenticated page, both the user types get the no-fragments version
[15:00:38 EDT(-0400)] <athena7> view the guest page, then log in as guest, and both get the fragment-y one
[15:01:05 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[15:02:01 EDT(-0400)] <athena7> that's probably not what we ideally want to happen
[16:19:22 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[16:27:47 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[16:30:57 EDT(-0400)] <awills> so eric... is it the case that WPP from JA-SIG is mavenized, but WPP at UW isn't?
[16:31:23 EDT(-0400)] <EricDalquist> yup
[16:31:38 EDT(-0400)] <EricDalquist> you can probably just ignore all the build.* related changes
[16:31:56 EDT(-0400)] <awills> sure thing
[16:49:07 EDT(-0400)] * colinclark (n=colin@142.150.154.101) has joined ##uportal
[17:31:58 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has joined ##uportal
[17:54:45 EDT(-0400)] <EricDalquist> hey awills
[17:54:57 EDT(-0400)] <EricDalquist> is there a way in crn to read a plain-text file?
[17:56:42 EDT(-0400)]

[17:56:48 EDT(-0400)] <EricDalquist> ah
[17:57:08 EDT(-0400)] <EricDalquist> wonderful thanks!
[17:57:20 EDT(-0400)] <awills> and then you can <line-iterator> through it if you like
[17:57:27 EDT(-0400)] <EricDalquist> perfect
[17:57:41 EDT(-0400)] <EricDalquist> we are going to have a list of usernames that are to be exported
[17:57:58 EDT(-0400)] <EricDalquist> so I need to load it into a Set in crn and check if a username exists in that set before actually dumping data related to it
[17:58:13 EDT(-0400)] <awills> yeah, that sounds good
[17:58:25 EDT(-0400)] <awills> bbl, meeting
[18:36:58 EDT(-0400)] <EricDalquist> that worked, it is a little confusing that it uses the directory the script was run from when other things like <properties> use the CONTEXT attribute
[18:37:10 EDT(-0400)] <EricDalquist> is that intended?
[18:50:58 EDT(-0400)] <awills> no, it probably wasn't... i guess it's just a byproduct of these things being created over a long period of time. I have in mind to create a ResourceHelper class that defines CONTEXT and LOCATION reagents in a standard way... thean work toward a complete collection of tasks that leverage them
[18:51:27 EDT(-0400)] <EricDalquist> ah cool
[18:51:49 EDT(-0400)] <EricDalquist> so I have export scripts that can dump our 2.5 install and then load into 3.0
[18:51:53 EDT(-0400)] <awills> do you think there's anything in the changes to portlet.xml (from your patch) that needs to come forward? it doesn't look like it
[18:52:01 EDT(-0400)] <EricDalquist> when I get a chance to breath in another week or two I'll be posting it all
[18:52:03 EDT(-0400)] <awills> that's great news
[18:52:08 EDT(-0400)] <EricDalquist> probably not
[18:52:33 EDT(-0400)] <awills> if you get a chance maybe give it a quick perusal... just that 1 file
[18:52:42 EDT(-0400)] <awills> but it doesn't look like any of it applies
[18:52:48 EDT(-0400)] <EricDalquist> yeah, looks like its all just local uw stuff
[18:52:53 EDT(-0400)] <awills> yeah
[18:54:15 EDT(-0400)] <awills> fyi the ResourceHelper class will also be a godsend in supporting 'classpath://' resources throughout
[18:54:48 EDT(-0400)] <EricDalquist> yup
[18:59:34 EDT(-0400)] * jessm (n=Jess@c-76-19-199-61.hsd1.ma.comcast.net) has joined ##uportal
[19:13:08 EDT(-0400)] <EricDalquist> and bash/linux gurus here?
[19:14:03 EDT(-0400)] * lennar1 waggles his fingers at sememmon and holdorph
[19:14:28 EDT(-0400)] <EricDalquist>
[19:14:31 EDT(-0400)] <EricDalquist> [dalquist@tumble svn-admin]$ sudo cp src/svnkit_1.2.0.4820/svnkit.jar /svn/crucible_test/crucible_home/lib/svn/svnkit-1.2.0.4820.jar
[19:14:31 EDT(-0400)] <EricDalquist> cp: cannot stat `src/svnkit_1.2.0.4820/svnkit.jar': No such file or directory
[19:14:31 EDT(-0400)] <EricDalquist> [dalquist@tumble svn-admin]$ sudo cp src/svnkit-1.2.0.4820/svnkit.jar /svn/crucible_test/crucible_home/lib/svn/svnkit-1.2.0.4820.jar
[19:14:41 EDT(-0400)] <EricDalquist> the first line of that is copying the command from an echo from a script
[19:14:49 EDT(-0400)] <EricDalquist> running the command in the script fails with the same error
[19:14:58 EDT(-0400)] <EricDalquist> the second line is typing the whole command by hand
[19:15:13 EDT(-0400)] <athena7> what does the script think the current directory is?
[19:15:22 EDT(-0400)] <EricDalquist> why oh why does the first one (copied and pasted) fail and the second one doesn't?
[19:15:23 EDT(-0400)] <EricDalquist> .
[19:15:35 EDT(-0400)] <EricDalquist> copying/moving other things from that same src/ directory works
[19:16:01 EDT(-0400)] <athena7> ok that is kind of weird
[19:18:10 EDT(-0400)] <holdorph> I'm in agreement with athena7, I'd at least rule out the pwd issues as possibilities first, even if you're fairly certain it's not it
[19:18:29 EDT(-0400)] <holdorph> I'd move to absolute paths, to rule it out, and debug the next step from there
[19:19:03 EDT(-0400)] <holdorph> for the next step, I'd start to try to verify what the 'effective' environment is of the two commands, including path,pwd,shell,effective userid, etc.
[19:20:25 EDT(-0400)] <EricDalquist> [dalquist@tumble svn-admin]$ sudo cp /home/dalquist/svn-admin/src/svnkit_1.2.0.4820/svnkit.jar /svn/crucible_test/crucible_home/lib/svn/svnkit-1.2.0.4820.jar
[19:20:25 EDT(-0400)] <EricDalquist> cp: cannot stat `/home/dalquist/svn-admin/src/svnkit_1.2.0.4820/svnkit.jar': No such file or directory
[19:20:31 EDT(-0400)] <EricDalquist> [dalquist@tumble svn-admin]$ sudo cp /home/dalquist/svn-admin/src/svnkit-1.2.0.4820/svnkit.jar /svn/crucible_test/crucible_home/lib/svn/svnkit-1.2.0.4820.jar
[19:20:44 EDT(-0400)] <EricDalquist> same thing, scripted version fails
[19:20:56 EDT(-0400)] <EricDalquist> if I copy and paste the string from the scripted version it fails
[19:21:20 EDT(-0400)] <EricDalquist> if I re-type the string (just the path to the file being copied, not the destination) it works
[19:21:38 EDT(-0400)] <EricDalquist> my only other thought is the svnkit zip extracts with weird issues
[19:24:05 EDT(-0400)] <EricDalquist> oh
[19:24:06 EDT(-0400)] <EricDalquist> well
[19:24:08 EDT(-0400)] <EricDalquist> time to go home
[19:38:40 EDT(-0400)] <Sememmon> strange..
[20:03:30 EDT(-0400)] * lennar1 (n=sparhk@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[20:03:46 EDT(-0400)] * awills (n=awills@wsip-98-174-242-39.ph.ph.cox.net) has left ##uportal
[22:51:28 EDT(-0400)] * MarkRogers (n=MarkRoge@wnpgmb0810w-ad08-125-123.dynamic.mts.net) has joined ##uportal