[ome-devel] image display problem

[Bernd Jagla]

I found the problem:

During the installation I was asked to give the URL to OMEIS
(http://127.0.0.1/cgi-bin/omeis), which I left as it is since I thought that
this should be valid if all servers are on one machine. I now changed it be
my machine's IP address and it works...

Thanks for all the kind help.

Bernd

|-----Original Message-----
|From: Ilya Goldberg [mailto:igg at nih.gov]
|Sent: Tuesday, September 12, 2006 12:34 PM
|To: Bernd Jagla
|Cc: 'Curtis Rueden'; ome-devel at lists.openmicroscopy.org.uk
|Subject: Re: [ome-devel] image display problem
|
|Hi Guys
|All of the errors in the log that you sent are in fact normal. More
|comments inline.
|
|You also said:
|"The problem I have is that the images are not displayed on the screen"
|Do you get the thumbnails?  What do they look like - an icon with a
|question mark inside?  A black square?  What happens if you click the
|lower-right corner of the square/icon/thumbnail - does it launch the
|SVG image viewer?
|Click on the upper-left corner of the square/icon/thumbnail, and you
|should get the image detail view.
|In the lower-right corner of this page (you may need to scroll down)
|it says "View with template:" followed by a drop-down menu, which
|says "detail".  Select "large" and click the Go button.  Still
|nothing?  Black square, or icon with question mark?
|
|If instead of images, you get an icon with a question mark, then
|something may be wrong with the image server (OMEIS).
|If you get a black square, it means that the default rendering
|settings were no good for your images.  You can fix this yourself
|temporarily by launching the SVG viewer, adjusting the color mapping
|and clicking "Save settings".
|You can adjust these settings specifically in the InCell importer,
|which is probably a better idea.
|
|Launching VisBio from the image view should work, but I'm not the
|best source if this is a problem.  It uses Java Web-Start to do its
|magic.
|Shoola does not detect that its session expired, and gives a non-
|informative error.  You really have to not do anything with your
|logged in session from any client for over a half hour for the
|session to expire.  Every request refreshes the session, so it
|shouldn't expire in normal use.
|
|Give a little more info about images not displaying, and we should be
|able to resolve that.
|The InCell importer should have set the display options.  You can
|check this from the image detail view (select "detail" from the "View
|with template" menu when looking at an image).  On the upper-right
|click on "Image import" in the "Module Executions" table.  The
|Outputs table should have DisplayChannel and DisplayOptions listed in
|the "Untyped Outputs" column.  Click on DisplayOptions.  These two
|outputs specify how planes in your image are mapped to RGB channels
|for display.  Each plane in your image that's used for display is
|listed in "DisplayChannel" with its white level and black level.  The
|DisplayOptions table sets how each DisplayChannel is mapped to RGB.
|These two STs control how the image is displayed by the web UI and
|any other client that reads these settings.
|
|Try importing one or more of the test images here:
|http://openmicrosocpy.org/TestImages/
|(maybe try the whole 011B1_Morph* set, which is a partial screening
|set from MetaMorph).  Are these displayed properly?
|
|-Ilya
|
|
|
|On Sep 8, 2006, at 10:43 AM, Bernd Jagla wrote:
|
|> Curtis,
|>
|> I can see the image even though I have to adjust the intensities...
|>
|> I see this LDAP message in the error_log of httpd:
|> Can anyone make sense of this?
|>
|> Here is the error log from the following steps:
|> Remove /OME (rm -rf)
|> Dropdb ome
|> Sudo perl install -y
|> ome import ~/allOMEs/inca/Screening.ome
|> (=> no problems)
|> ome import ~/test/Ai_compound_import.ome
|> => problems with
|> OME::Tasks::LSIDManager::getRemoteObject() is not implemented. at
|> /usr/lib/perl5/site_perl/5.8.6/OME/ImportExport/ModuleImport.pm
|> line 183
|
|This is a warning (not an error), and is normal and expected.
|
|
|>
|> Could it be that this error comes up when doing the second import
|> within a
|> certain timeframe? There is no error log entry when this error occurs.
|>
|> Thanks,
|> Bernd
|>
|> Ok here is finally the snippet from the error_log:
|>
|>
|> [Fri Sep 08 09:31:21 2006] [notice] Digest: generating secret for
|> digest
|> authentication ...
|> [Fri Sep 08 09:31:21 2006] [notice] Digest: done
|> [Fri Sep 08 09:31:21 2006] [notice] LDAP: Built with OpenLDAP LDAP SDK
|> [Fri Sep 08 09:31:21 2006] [notice] LDAP: SSL support unavailable
|
|This is probably from mod_auth_ldap (not a part of OME).  Anyway,
|this doesn't look like an error.
|
|> [Fri Sep 08 09:31:22 2006] [notice] mod_python: Creating 4 session
|> mutexes
|> based on 150 max processes and 0 max threads.
|> [Fri Sep 08 09:31:22 2006] [notice] Apache/2.0.54 (Fedora)
|> configured --
|> resuming normal operations
|> Use of uninitialized value in print at
|> /home/bernd/allOMEs/06_08_18/OME/src/perl2/mod_perl_test.pl line 6.
|> Use of uninitialized value in print at
|> /home/bernd/allOMEs/06_08_18/OME/src/perl2/mod_perl_test.pl line 6.
|
|These appear sporadically and are generally harmless.  We try to get
|rid of them as we see them, but these apparently we missed.
|
|> [Fri Sep 08 09:31:22 2006] [error] [client 127.0.0.1] In PID 10585,
|> Error
|> calling OMEIS: Method parameter missing
|> [Fri Sep 08 09:31:26 2006] [error] [client 127.0.0.1] In PID 10601,
|> Error
|> calling OMEIS: Method parameter missing
|
|These are the expected responses from OMEIS if the test is
|successful.  So when installing, this is a good thing.
|
|> [Fri Sep 08 09:31:44 2006] [notice] SIGHUP received.  Attempting to
|> restart
|> [Fri Sep 08 09:31:44 2006] [warn] The ScriptAlias directive in
|> /OME/conf/httpd2.omeis-dev.conf at line 2 will probably never match
|> because
|> it overlaps an
|>  earlier ScriptAlias.
|
|This one always comes up, and we haven't figured out what to do about
|it.  Anyway, its harmless because the installer copied the omeis
|binary to whatever ScriptAlias was set to (otherwise the above
|expected "Missing parameter" error would not have been generated by
|OMEIS).
|
|> Use of uninitialized value in concatenation (.) or string at
|> /home/bernd/allOMEs/06_08_18/OME/src/perl2/OME/Install/
|> ApacheConfigTask.pm
|> line 75.
|
|Again, one of the ones we missed.  Harmless.
|
|> Processing OME-startup.pl
|> DBD::Pg::st execute failed: ERROR:  permission denied for relation
|> experimenters
|
|During the server startup, it tries to get an OME session as the user
|who launched apache.  Usually this is root, and usually this fails,
|so it tries to get a session as the Apache user:
|
|> Tring to get a session as the Apache user...
|
|And it succeeds as the Apache user, so all is well because it goes on
|to pre-load the STs:
|
|> Pre-loading SemanticTypes...
|> WARNING: Explicit deletion of OME::Session at /OME/conf/OME-
|> startup.pl line
|> 76
|> OME::Session __destroySession: $self->__destroySession();
|>  at /OME/conf/OME-startup.pl line 76
|
|This is also normal because we explicitly shut down the temporary DB
|handle we allocated to pre-load the STs.
|
|
|> Pre-loading classes for OME::Web...
|
|This is the last thing the pre-load script does, so all is well.
|
|> [Fri Sep 08 09:31:47 2006] [notice] Digest: generating secret for
|> digest
|> authentication ...
|> [Fri Sep 08 09:31:47 2006] [notice] Digest: done
|> [Fri Sep 08 09:31:47 2006] [notice] LDAP: Built with OpenLDAP LDAP SDK
|> [Fri Sep 08 09:31:47 2006] [notice] LDAP: SSL support unavailable
|
|These are probably from mod_auth_ldap - has nothing to do with OME.
|
|> [Fri Sep 08 09:31:48 2006] [notice] mod_python: Creating 4 session
|> mutexes
|> based on 150 max processes and 0 max threads.
|
|These are probably from mod_python - also has nothing to do with OME.
|
|
|> [Fri Sep 08 09:31:48 2006] [notice] Apache/2.0.54 (Fedora)
|> configured --
|> resuming normal operations
|
|And this means we're good to go.
|
|
|>
|> |-----Original Message-----
|> |From: ctrueden.wisc at gmail.com [mailto:ctrueden.wisc at gmail.com] On
|> Behalf Of
|> |Curtis Rueden
|> |Sent: Friday, September 08, 2006 9:40 AM
|> |To: Bernd Jagla
|> |Cc: ome-devel at lists.openmicroscopy.org.uk
|> |Subject: Re: [ome-devel] image display problem
|> |
|> |Hi Bernd,
|> |You could try reading the binary .ome files with VisBio from disk,
|> to see
|> |how things look.
|> |-Curtis
|> |
|> |
|> |On 9/7/06, Bernd Jagla < baj2107 at columbia.edu
|> <mailto:baj2107 at columbia.edu>
|> |> wrote:
|> |
|> |	Hi there,
|> |
|> |
|> |
|> |	I managed to get the images from the INCell imported again with the
|> |new version of OME. The images are in OMEIS (verified). They 16
|> bit Tiffs.
|> |
|> |	The problem I have is that the images are not displayed on the
|> screen.
|> |They cannot be exported to VisBio. I get an XML from the download
|> with some
|> |binary data but can't confirm that this is what I want since I
|> don't know
|> |how to read them. I believe that the attached error log has
|> something to do
|> |with this. I couldn't see anything interesting in the http error
|> logs.
|> |
|> |
|> |
|> |	Please advise where look for possible solutions.
|> |
|> |
|> |
|> |	Thanks,
|> |
|> |
|> |
|> |	Bernd
|> |
|> |
|> |	_______________________________________________
|> |	ome-devel mailing list
|> |	ome-devel at lists.openmicroscopy.org.uk
|> |	http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
|> |
|> |
|> |
|> |
|> |
|>
|>
|> _______________________________________________
|> ome-devel mailing list
|> ome-devel at lists.openmicroscopy.org.uk
|> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-devel
|>