Global Variable translated to different characters

[Former Member]

We recently upgraded BODS from 4.2 SP1 to 4.2 SP4. Now we are facing issues in those DataFlows where are used Hierarchy Flattering functions, that Global Variables are wrongly translated from text to weird Asian characters. E.g. varchar variable: 'abc' is transformed to '扡c'. As data source/target is used HANA. The issue is caused whenever there is any table from HANA datastore 'directly' connected to HANA (attached picture). When used ODBC connection to HANA than it works fine, but then we are loosing Bulk load options.

Attached ATL (with .txt extenstion) has password: abc

Versions:

DS: 4.2 SP4

HANA: 1.0 SP09

HANA Client ODBC: 1.00.09 (tested also on 1.)

After_upgrade_testing.txt.zip

[Former Member]

Ok, sorry for a basic question but I can't find anywhere that explains this.

Does anyone know how to translate the version number to a patch level?

We're on 14.2.4.668

I know that's 4.2 SP4 but what does the 668 equate to in patch levels?

[karen_neeb]

Hi Vanda,

668 is the build number. If you search the SAP Notes/KBAs for 14.2.4.668, you'll find the KBA release notes for that version. 14.2.4.668 corresponds to SP4 (patch 0).

SAP Data Services 4.2 Support Pack 4 (Version 14.2.4.668) Announcement

Regards,

Karen

[Former Member]

It looks like there are some notes that address this, 2152186 for 4.2 SP 3 and 2151959 for 4.2 SP 4.

I just ran into the same issue today, so I'm hoping one of those fixes it.

[Former Member]

Applying 4.2 SP4 Patch1 solved the problem.

For 4.2 SP3 support recommend to apply Patch4.

[former_member106536]

There is a bug in the most recent version that corrupts global variables.

I dont seem to recall jobs working when ran in debug, but we saw the same strange characters in gvars all over the place, in different jobs.  Some worked some didnt.  There seemed to be no rhyme or reason to which would be corrupt. 

SAP provided us a special patch to fix it.  Actually 2 patches since there was also a bad access violation error introduced.

[meenakshi_raina]

Hi Dusan,

How are you trying to do the dataflows ? Is there any option coming to regenerate it. Regeneration of object basically works after the upgrade.

Regards

Meenakshi

[Former Member]

The problem appears even when I regenerate the DF's or DataStores. I suspect some discrepancy between HANA and DS, maybe driver (code page).

When I debug the WF the variable is behaving normally, but once it's send to HANA it's somehow translated to weird Chinese/Japanese characters.

EDIT: I forgot mention that our DS Repository is stored on MS SQL Server 2008 R2 - connected via ODBC v 6.01.7601.

[former_member198401]

Hi Dusan,

Never seen this kind of behavior.

Can you please set the Code Page to us-ascii under Advanced options in HANA Datastore. Not sure if it will work or not

Regards

Arun Sasi

[Former Member]

Hi Arun,

unfortunately I can't find any option to set code page when using HANA Data store (screen 1 & 2 in above attachment Datastores.jpg)

D.