AL32UTF8 on Oracle 12c

[Dan_Wentworth]

Has anyone used the Oracle AL32UTF8 character set for an OnBase 16 implementation?

I would be be interested in any risks\issues encountered as well as what modules were implemented.

Thanks

Dan

[Madhu_Vatti]

For our onbase, we use below for oracle 12c RAC multitenant.

Region Character Set
                 US WE8MSWIN1252

[Ryan_Coyne · Accepted solution]

Hello Dan,

Thank you for using Community. The AL32UTF8 Character Set is supported with OnBase on Oracle, but only for Unicode systems. ANSI systems would still use one of the Character Sets listed in the Database Reference Guide such as WE8MSWIN1252 mentioned above. Unicode is relatively new functionality for OnBase (beginning with OnBase 14 for SQL Server and OnBase 15 SP1 for Oracle) which called for significant changes in the software behind the scenes, resulting in some limitations for existing modules. A Module Reference Guide is now published for each OnBase release detailing the Unicode Considerations (OnBase 16) of a given OnBase release, including known limitations for existing modules as well as a background on Unicode with OnBase at a general level.

Because of these limitations, the ability to install a new system with Unicode or convert an existing system to Unicode does require the involvement of Hyland - even for Partners. This also helps to ensure that the business need for Unicode is valid, and that there are no available workarounds that could be used in the more common ANSI setup. If you are interested in discussing Unicode options for a new or existing customer, I'd recommend reaching out to your First Line of Support.

Thanks,
Ryan

[Nelson_Raimond]

Hi Ryan, what do I have to understand for a Unicode or ANSI system? Are you referring to the OS that hosts the database? Oracle stands the following notes:
• Oracle recommends AL32UTF8 even for pure ASCII data, because it is much less costly to start with the right multilingual character set instead of to migrate the database later when it has accumulated giga- or terabytes of data and additional languages have to be supported. In the Internet area, it is not wise to assume that the database will only ever have to support English.
• US7ASCII is a particularly bad character set, because hardly any client platform is really US7ASCII. If you insist on a single-byte character set, use WE8MSWIN1252 for the database character set.

[Ryan_Coyne]

Hi Nelson,

The distinction of Unicode vs. ANSI in the context of OnBase refers to the application/schema configuration, which is outlined in the Unicode Considerations document linked above. This is independent of the operating system used for the Oracle Database. By default, an OnBase solution is deployed as ANSI, but beginning with OnBase 15 (for Oracle) can be installed as Unicode compliant with assistance from Hyland and/or your First Line of Support. Details for implementation/migration paths between ANSI and Unicode are also outlined in the Unicode Considerations document.

Despite Oracle's stance that all databases regardless of data should use the AL32UTF8 Character Set, Hyland Software does not test ANSI implementations of OnBase against AL32UTF8, and use of that Character Set with an ANSI OnBase solution would not be supported. As you stated, WE8MSWIN1252 would be the recommended Character Set for a US region ANSI OnBase database.

I'd recommend reaching out to your First Line of Support to discuss your particular OnBase implementation and character storage needs to identify the best next steps.