Core Content Only

There has been some good discussion going on over on the Oracle ECM Forums regarding Accounts. It started out as Accounts and Access Control Lists but it has morphed a little into some discussion about how to use Accounts in general and some possible structures people are using to layout their account hierarchies.

Check out these discussions about accounts in the Oracle ECM Forums:

http://forums.oracle.com/forums/thread.jspa?threadID=1001039&tstart=0

One of the easiest ways to see what might be going wrong, or right, in Content Server is the use of trace sections on the System Audit page. Trace sections allow you a great variety of control over what kind of information shows up in the server output.

See this former blog entry for more detailed information about tracing:

Content Server Tracing and Creating Your Own Custom Trace Sections

Refreshing the Server Output to show you what is going on can get annoying though. If you add this entry to General Configuration or config.cfg and restart you can get server output to be written to disk:

UseRedirectedOutput=true

To find the log on the disk look under the directory that is appropriate for your Operating System.

Windows: <install dir>/bin/IdcServerNT.log

Linux: <install dir>/etc/log

Now that you have found the log, it would be easier to keep an eye on what’s going on by using something like the tail command in Linux/UNIX. The tail command, with the -f switch will continuously show the data being added to the log file. In Linux you simply use this command:

tail -f /etc/log

There are several applications available that add similar functionality for windows:

http://tailforwin32.sourceforge.net/
http://www.baremetalsoft.com/baretail/index.php
http://www.withdata.com/tail4win.html

Grab the Web URL Map Extras Component at the Downloads page.

In my Oracle OpenWorld 2009 demo I showed how to use the Web URL Map feature of Content Server to make images and renditions easier to use in Cascading Style Sheets. This post is about that topic and one of the components I demonstrated during that session.

One way to make referencing images (and especially renditions) in Cascading Style Sheets and other places in your web sites is to use the Web URL Mapping tool that comes with Content Server. Check out Kyle’s blog, he has a whole post about this feature.

I have several of these mappings that I always like to add to a server. So, I made a component that will add these mappings to the server when you install the component. Also, I wanted an example component showing how you can read from a defined map and add it to another map. Hence, the WebMapUrlExtras component was born and is available here.

WebMapUrlExtras adds the following four mappings.

w/
pass in a dDocName and get back the web rendition of an image

n/
pass in a dDocName and get back the native file

qs/
pass in a query string and get back the quick search results

r/
pass in a dDocName and get back the web rendition, or you can optionally specify the rendition by tagging on a ?Rendtion=<rendition name>

Prior to the component your Web Map URL configuration might look like this:

After installing the component and restarting it should look similar to this:

The other day I was using an Oracle database tablespace supplied by a dba for a development environment and when I went to rebuild the full text index the Repository Manager greeted me with this nifty little error:

ORA-01658: unable to create initial extent for segment in tablespace

Fun huh?

Ok, this is a development box, lots of space. I thought this was all setup to expand on need, but let's check it out. First, we need to go to the database via something like Toad, SQL Developer or JDeveloper and execute a Query similar to this one:

select file_name, bytes, autoextensible, maxbytes from dba_data_files where tablespace_name='DEVWCM_SYSTEM';

You're tablespace name will vary, BUT REMEMBER: the value you supply for tablespace name is case sensitive

Running this query I find out that autoextend is NOT turned on, and further exploration yields a maxed out datafile. Fine, let's get autoextend turned ON. Here's two samples:

alter database datafile
'C:\Oracle\app\oradata\orcl\DEVWCM_SYSTEM.DBF'
autoextend on;
alter database datafile 'C:\Oracle\app\oradata\orcl\DEVWCM_SYSTEM.DBF'
autoextend on next 100m maxsize 2000m;

In the first case we're going to simply turn on autoextend and let it ride. In case number two you can see some extra instructions including how much to extend and a limit.

And did this take care of the problem? Yep. Sweet, a fresh, clean, working index.

Out on the Oracle ECM Forums a question came up about how to find out which index is currently active (from Java). It's quite simple really. Just use the getActiveProperty method of the ActiveIndexState class.

Like so:

String activeIndexRoot = ActiveIndexState.getActiveProperty("ActiveIndex"); 

To see this in action grab this Sample Component to Find Your UCM Search Active Index.

While my blog output the last couple months has been lackluster at best the community at large has benefited from a new author in the arena. Take a look at http://blogs.oracle.com/kyle. Leave a comment. Welcome Kyle Hatlestad to our blog-o-sphere.

I am adding Kyle to the blog roll here. He has already produced a series of very nice articles and I strongly encourage you to take a gander and add his site to your RSS Reader. To give you a sample, here are the titles of his first eight posts:

2009.08.31 – How to set up a Public Document Library in WebCenter Spaces
2009.08.21 – Positioning Sections in Site Studio Designer
2009.08.20 – Site Studio Naming Best Practices
2009.08.18 – UCM Black Hole Check In
2009.08.14 – Allow Contributors To Add Fragments To Your UCM Site Studio Pages
2009.08.06 – UCM Custom Skins and Layouts
2009.07.23 – Global Rules for Standard Check In and Search Only
2009.07.21 – Displaying Native Content With Site Studio 10gR4 in External Apps

Excellent work so far. Kyle has been involved in various aspects of the community for a while. I see his name in forums and blogs occasionally. This is a great next step for him, and a benefit for all of us learning and using Content Server on a weekly basis.

In an effort to learn more about Kyle I have been sending him questions and tabulating his answers to form a kind of "Web Interview". I wanted to introduce him as a community participant, point out his great work and demonstrate to people just how approachable he is. Here's some of the questions I asked and his responses.

Where have you worked other than Oracle (did you come over with the Stellent acquisition)?

Yes, I came over as part of the acquisition of Stellent. I joined Stellent (which was formally named IntraNet Solutions) in July of 1998. Prior to Stellent, I worked at Camax which was a Computer Aided Manufacturing (CAM) software company for 3 years.

Where is home?

I work in Eden Prairie, MN which was the Stellent headquarters. I live about 20 miles north and was born and raised in Minnesota.

Can you describe what you do now for Oracle and the previous roles you have played in the history of Stellent?

I work in the sales organization in a department known as the Enterprise Solution Group. Our charter is to put together reusable sales assets that can be used during sales cycles. Prior to sales, I was in the Quality Assurance department doing testing for several years. I was also a consultant for several years and did go to customers to do initial installs and configurations of systems.

Is there anyone you can single out has influential in your career development?

Frank Radichel who is a Vice President in Software Development at Oracle and was head of R&D under Stellent has been a big influence on me. I've worked directly for him when I was in the Quality Assurance group and have always worked with him in the other various roles that I've had (Tech Support, Consulting, Sales). He's always been a great mentor and someone I could always turn to with questions.

What products do you work with on a semi-regular basis at Oracle besides UCM (your recent blog post indicates WebCenter Spaces)?

The major ones within what we call the 'Enterprise 2.0' group. They include the portal products of WebCenter Spaces, WebCenter Interaction, and WebLogic Portal. I also work a lot with Information Rights Management and Oracle Capture. Outside of the E 2.0 products, I've done work with BPEL Process Manager, Siebel, AutoVue, & BI Publisher. It's basically the laundry list of products that integrate with UCM. Which seem to be growing every day.

Within UCM what are your favorite two products and why?

I generally like to work with Site Studio. Web Content Management offers a lot of flexibility and creativity. I also like working with the core Content Server which has proven to have a great architecture over the years.

Outside of UCM what are your favorite two Oracle products?

WebCenter Spaces is a great product to work with. It's a very advanced product with a lot of flexibility and control. Provides a great portal/workspace environment that we use internally on a day-to-day basis with our own projects.

And I guess the other product would have to be the database. Under Stellent I was often intimidated when working with the Oracle Database, but have really enjoyed digging into all the things it can do and benefits it provides. Things like Secure Files with compression, deduplication, and encryption of unstructured data, the Application Express (APEX) interface, Oracle Text searching, etc.

What's one piece of functionality that you would like to see added to Content Server? (I get asked this on support follow-up calls all the time!)

Yeah…this is a tough one because it's always changing depending on what I'm working on at the time and what's missing.

While it's not technically functionality, I think a redesign of the UI would be nice. I think functionally it can do most of the things customers need (or can be customized to), but the general UI can be difficult to use for the non-technical user.

One enhancement I've always wanted to be included since day one has been the ability to store and present comments in Workflow. For both rejection and approval. I've been asked too many times then I can count if you can add comments when you approve in workflow.

What do you feel is the hardest part about customizing Content Server?

Tracing resource includes across different components and patches. Resource includes are great in that they are modular and reusable, but given the number of core and extra components that generally get loaded, it can be cumbersome at times to find the code you need to override for your component.

Do you have any advice to share with those that have just purchased Content Server and are intimidated by "the possibilities" it provides them?

Well, I've always tried to encourage customers to start small and try to do quick win projects first. It helps with the learning curve and encourages adoption. I've seen too many initial big projects that get bogged down in politics, lack of understanding, and poor planning.

Also, I tell every customer to buy the book "The Definitive Guide to Stellent Content Server Development" by Brian "Bex" Huff. Even though it was written prior to the acquisition, the majority of it still holds true for UCM 10gR3 and is a tremendous resource.

Please tell us your vision for the future of Content Management (2-3 years out). What are some forthcoming trends, technologies, etc.?

I think it's something that Billy Cripe has been saying for a few years now and that's "invisible ECM". It's the concept that ECM is sitting behind the systems that businesses use to do their day-to-day work. So it will be less important that it be a primary system that users go directly to and use but rather transparent to them as they are executing their business process. So I see a lot more integrations down the line that tie Content Management systems closer to these business systems.

Here's to hoping that Kyle can find the time to continue sharing his insights with all of us in the community for a long time to come. Thanks for the info Kyle, keep the posts coming!

When attempting to install or uninstall Content Server or the related services on a Windows box you may occasionally run into some problems. Whatever the scenario is, you may get to a place where you need to manually uninstall the services.

Note: As you will recall, uninstalling Content Server is generally a very simple task of finding the bin directory on the command line and issuing this command:

Installer UnregisterAll

This should cleanup everything. When it doesn't, you can attempt these maneuvers:

-cs install-\bin\IdcServerNT -uninstall -cs install-\admin\bin\IdcAdminNT -uninstall -ibr install-\bin\IdcRefineryNT -uninstall

If this doesn't work, you can also try this beauty:

sc delete "service name"

And finally, you can go hack the registry at your own risk:

1. Open command prompt
2. Type "regedit"
3. Press enter
4. Navigate to 'HKEY_LOCAL_MACHINE/SYSTEM/CurrentControlSet/Services'
5. Find service name you wish to remove
6. Select it
7. Delete it

A question was asked on the Oracle ECM forums about what the various values for Release State meant. I knew which states some of the single character values represented, but not all of them. I was interested in finally finding out this answer for myself once and for all so I dug in. What follows is what I found as a result:

N
apStateNew
This item was newly added to the system

Y
apStateCurrent
This refers to the revision status of the content item (so it's the most recent revision of the content item)

O
apStateOld
This refers to the revision status of the content item (so it's not the most recent revision of the content item)

E
apStateWorkflow
This means the content items is currently in a workflow

R
apStateProcessing
This means the content items is currently being processed by the indexer

U
apStateUpdate
This means the content item is being updated

I
apStateIndexing
This means the content items is currently being indexed

A recent client has users who were getting frustrated about the case sensitivity of the UCM logins. Their old system did not have this restriction (feature?) and they wanted this removed. The more of these types of things you remove the less secure the login becomes, but depending on each situation the risk/reward analysis plays to either side of the fence.

This turns out to be a fairly easy task to accomplish. First, the users in question need to be changed to "Global" in the User Admin applet.

NOTE: This means you cannot perform this action with SYSADMIN.

Having switched the appropriate users to Global, you need to add this setting via Admin Server (under General Configuration) or directly to the confing.cfg file. Once added, do not forget to restart the services.

AllowCaseInsensitiveLogin=true   

The default query form setting out-of-the-box is the expanded "standard" form. New with 10gR3, you also get the Query Builder form that allows non-technical users to build more advanced queries. Query Builder also includes some new advanced options for power users.

One thing you may want to do at some point is change the default query form view. Users can set their preferred query form view in their profile. Once a user sets this in the profile it will override anything you or your admin set as the system default.

To see where all this query form view decision making is happening we will take a quick peek at the "determine_search_form_type" dynamic html include:

<@dynamichtml determine_search_form_type@> <$if not searchFormType$>  <$searchFormType = utGetValue("pne_portal", "searchFormType")$> <$endif$> <$if not searchFormType and DefaultSearchFormType$>  <$searchFormType = DefaultSearchFormType$> <$endif$> <$if not searchFormType$>  <$searchFormType = "standard"$> <$endif$> <$if DisabledSearchFormTypes$>  <$DisabledSearchFormTypes = DisabledSearchFormTypes$> <$else$>  <$DisabledSearchFormTypes = ""$> <$endif$> <$if DisabledSearchFormTypes like ("*" & searchFormType & "*")$>  <$if DefaultSearchFormType$>   <$searchFormType = DefaultSearchFormType$>  <$else$>   <$searchFormType = "standard"$>  <$endif$> <$endif$><@end@> 

Upon examination of this include you can see that you can use DefaultSearchFormType as a configuration variable in Admin Server or [install]/config/config.cfg to control the default query form view. Also note, they provide a variable to disable a particular query form view, DisabledSearchFormTypes.

Out-of-the-box, possible values we can set are "standard" which is the default, and "queryBuilder".