0
./data-model-cmd.sh uploadrpd -I mytest.rpd -W weblogic -SI ssi -U weblogic -P weblogic

I get the following error when I try to install the appropriate file.

Error is : Operation Failed. An exception occured during execution, please check server logs.

Here

  • RPD source file in windows server.
  • The position I want to install linux server.
  • I could not find in the log directory.

In the same version 2 server installed. Is there any idea?

m2h2
  • 33
  • 1
  • 5

4 Answers4

1

I am no expert in obiee, but this is how I upload my rpd in obiee12c (and I think that is what you are asking about):

./datamodel.sh uploadrpd -I /fullPathToRPD/RPD_NAME.rpd -U userNameOBIEE -P passwordOBIEE -W passwordOfRPD -SI ssi

I hope it helps

Sindre
  • 75
  • 1
  • 6
  • Many times I've uploaded RPD files. I'm trying to transfer files from Windows server to Linux Server here. Maybe that is the problem. I couldn't find this a very relevant topic. Thanks for your attention. – m2h2 Oct 01 '16 at 22:57
0

To upload the RPD in OBIEE 12c, use the documented process: uploadrpd through datamodel.sh/datamodel.cmd, which is available through the client tools on Windows.

This is the only supported way to do it. You can run it remotely from your client machine. You do not need to manually copy the RPD to the server.

Robin Moffatt
  • 30,382
  • 3
  • 65
  • 92
0

I had this problem on a new install of OBIEE 12.2.1.4 and, after a fun morning, managed to track it down to OBIEE case sensitivity to server name.

A RedHat Linux 7 server was provided to me. At build time it had an /etc/hosts file with an entry for the local machine. This was all in caps:

192.168.123.456 MACHINENAME.DOMAIN.FOO.BAR MACHINENAME

It also had an /etc/hostname in lower case:

machinename

OBIEE Installed fine - all working. Came to upload RPD and saw OP's error message.

Also saw this in the REST log:

cat $DOMAIN_HOME/bi/servers/bi_server1/logs/bi-lcm-rest.log.0

...    oracle.bi.lcm.pojo.si.rpd.RpdHelper INFO - getCustomizationSessionFromLocalOBIS: localNodeHostName info ------->machinename:Null
...

Comparing with another server where I was able to successfully upload the RPD I saw that this should be:

...info ------->machinename:<port>

i.e. where there should be a valid port number for the OBIS service I had Null.

Other threads suggested that similar errors were caused by an incorrect server name somewhere.

To resolve I did the follwing:

Commented out the entry for the local server from /etc/hosts:

# 192.168.123.456 MACHINENAME.DOMAIN.FOO.BAR MACHINENAME

Changed the /etc/hostname entry to be capitalised:

MACHINENAME

Rebooted and it worked.

Ideally I'd go back and rebuild with both /etc/hosts and /etc/hostname changed to lower case. But I haven't time to test this now.

Hope this saves someone else a morning of debugging.

RandomPav
  • 1
  • 1
-1

Uploading the rpd is a two step process. Ideally you want to copy your rpd to the customization folder which is: 1)$ORACLE_HOME/user_projects/domain/bi/bidata/service_instances/ssi/metadata/datamodel/customization. 2) go to bitools/bin:./data-model-cmd.sh uploadrpd -I customization/rpd -SI ssi -U weblogic -p password

You will be prompted to enter the rpd password.

As per the error received verify the bi_server log and bi-lcm-rest log for the root cause. Chances are that the BI Server restarted. Else look at the BI process in FMW-EM to ensure that all process are up.

  • _"copy your rpd to the customization folder which is: 1)$ORACLE_HOME/user_projects/domain/bi/bidata/service_instances/ssi/metadata/datamodel/customization."_ <-- NO NO NO. This is **completely and utterly incorrect**, and dangerous, information. That folder is internal to OBIEE. RPD files end up there as a result of the `data-model-cmd`, but go through various steps on the way. The method you're describing here (manually copying the RPD to a folder) was correct 10g. Doing this in 12c may well break your OBIEE instance by stopping BI Server from booting. – Robin Moffatt Jan 29 '17 at 11:17
  • Hello Robin, so what you are saying is that manually copying the binary file to the customization folder is not the best approach in 12c. What else would be the best approach? I believe that could be a bigger issue as it could corrupt the service instance (ssi) which would then require to reset the service instance using wls. Can you confirm what would be your approach? – Kodima Zeebra Jan 30 '17 at 16:15
  • The correct approach is the one that is documented - data-model-cmd: https://docs.oracle.com/middleware/12212/biee/BIEMG/GUID-C0BBA452-CC54-43A3-9A61-F1ADCCBEBDDB.htm#upload_repos_cmd – Robin Moffatt Jan 30 '17 at 19:36
  • Thanks for the prompt response Robin. We have been experiencing errors deploying the rpd as we were manually copying the rpd to the customization folder then execute the data-model.sh uploadrpd command which would failed. I ran the wlst to reset the service instance then ran data-model.sh uploadrpd with -I /tmp/rpd....and it work fine. I will have to monitor this method. By the way I'm a big fan of your blogs. Thanks again. – Kodima Zeebra Jan 30 '17 at 21:34