Best Of
Re: Physical standby database instance licensing
Hello F1R7Z,
DataGuard is part of "EE" so no need for extra licensing for the "option", but the server hosting the STANDBY (or occasionally the PRIMARY) has to be licensed…
What's more: beware of not using "Active DataGuard" (as it is very easy to do it without realizing) as this is an extra option that requires extra licensing (and is not cheap…)
Something else: welcome to MOSC community (well, you arrived ~6 months ago but it isn't too late to welcome you); note that you can very easily change your "display name" into a more "human-friendly" value, just have a look at and you can also have a look at the follwowing document containing several pointers to other documents:
Best regards,
Bruno Vroman.
(Edited: about RAC, note also the possibility of using "RAC One Node", cheaper than full RAC)
Re: Physical standby database instance licensing
As licensing is a complicated thing, we can just repeat standard licensing rules.
A standby environment must follow the same rule (licensing NUP or CPU) like the primary and in most cases, you do need the same Enterprise Edition Packs on top (e.g. Partitioning, Tuning,Diagnostics,…). You don't need to license RAC option if the standby is on a Single System. Nevertheless - ask your Oracle Sales to approve this.
What is the best way to configure BU level security for an expenses Insights Dashboard?
Hello,
I am in the process of developing a custom Insights Dashboard for the expenses module for my agency. I am trying to work through security testing now. I am trying to determine the best way to assign BU level security for the results in the dashboards. Currently, we have the search definition we are using for the dashboard configured to source-level security, with the query security record SP_BUEX_VW assigned to the ex_sheet_hdr record on the BU. (ex_sheet_hdr is a primary record for the query associated with the search definition we are using) My team thinks that this will work in theory, however, me being newer to security, am unsure. The only reason that I am unsure is because, to the best of my understanding, the query is not executed for each different user every time they access the dashboard. This makes me think that a query security record may not be the best option. My understanding of search framework security is that source-level security within the search framework grants users with the appropriate role assignments complete access to all results derived from a specific search definition; it's an all-or-nothing approach that doesn't filter results further. Am I correct on this or am I misunderstanding? If my understanding is on the right path, then I don't believe source-level security will work for us, but maybe I'm wrong?
I explored how the PS delivered expense search definitions were configured from a security standpoint to compare and give myself a better understanding, and they all appeared to have the same structure. The PS delivered security query EX_SRCH_SECURITY_SES_KIB appears to focus on sheet_id and the oprid of the approver of the sheet_id. Essentially, this security query just finds the sheet id and a corresponding approver that can approve the sheet_id, and if you order the query by the oprid of the approver, you can see a list of all sheet_ids that any one oprid can approve. With this understanding.. I don't know if this solution is what my team needs either. We want executives to be able to view high-level TE data for BUs that they have access too.
We need BU level security, not approver oprid security.. if that makes sense. I created a query in a lower environment, let's call it, te_bu_sec_query that selects the oprid, bu, and descr from the SP_BUEX_OPRVW record and joined on the psoprdefn record to get the oprid's emplid and name. Then I went to the connected query and mapped on this te_bu_sec_query on the BU field. Would this approach better solve my needs? I was thinking that this was the approach my business needs, however, am not 100% certain. Like I stated above, the goal of this dashboard is to allow executives access to high-level TE data/insights for BUs that they have access to, and nothing more. We tried the approach in paragraph 1, but am not 100% that will work. Could anyone provide any other guidance or expertise to help me determine best next steps? Like I said, I'm a little new to peoplesoft security so I want to be cognizant of best practices.
Our Releases:
PT: 8.60.13
OpenSearch v 2.3.
Re: How to download a pdf file stream from a API call response using Orchestrator Connector Service?
Hi Cezar,
Thank you. I will try it and let you know.
Thank you
Sabu M
Re: Block details are not shown through dba_extents view
While the DBMS_ROWID package correctly identifies the specific data block containing your row (block 112433), the block_id column in the dba_extents view only marks the starting block of a larger, contiguous allocation of space known as an extent. Your specific block resides within this extent but is not necessarily the first one, which is why a direct equality check fails. To resolve this and locate the correct extent, you must query for a range that contains your block number by checking if it falls between the extent's start block (block_id) and its calculated end block (block_id + blocks - 1). The correct query would be:
SELECT * FROM dba_extents WHERE file_id = 1 AND 112433 BETWEEN block_id AND (block_id + blocks - 1);
Cheers
Arsalan
Re: Changing domain & logins
You have to use My Oracle Support for this. You need to open a ticket and provide the original address/domain and the new one. I believe there is a knowledge paper on this
Re: Chained ROWS in UNIFIED AUDIT tablespace
If you have chained rows , recreating the table is often recommended to optimize storage. You can use CTAS to create a copy of the table, then swap it back. This approach will also defragment the table.
CREATE TABLE AUDSYS.AUD$UNIFIED_TEMP TABLESPACE AUDIT_TS AS SELECT * FROM AUDSYS.AUD$UNIFIED;
DROP TABLE AUDSYS.AUD$UNIFIED;
ALTER TABLE AUDSYS.AUD$UNIFIED_TEMP RENAME TO AUD$UNIFIED;
rebuild indexes and shrink tablespace then.
Re: Chained ROWS in UNIFIED AUDIT tablespace
A couple of questions:
What are you checking when you say you have 50GB of audit data? The table has three CLOB columns, and many people forget to add in the size of the LOB segments when they report the table size.
John's comment about chained rows is highly significant in this case - the varchar2() columns in the aud$unified table have a total length declaration of a little over 70KB; it's not likely that any rows actually use that much space, but the rows could still be very long if you've enabled any interesting features of Oracle. On top of that, the CLOBs are all declared "enable storage in row" - one is about the RLS (FGAC) information, one is the SQL_TEXT, and one is the SQL_BINDs, so in many cases you might have code that squeezes below the 4000 byte limit that allows the CLOB to stay in the row. Bottom line - chained rows will not go away if you rebuild the table.
(It's also possible that the 49pct reclaimable space is NOT reclaimable - just because it's Oracle's own space advisor that's reporting it there's no good reason to assume the code is accurate.)
What is your current retention, and what did you reduce it from - it's possible to imagine a pattern where (e.g.) you had 9 months of audit stored and reduced it to one month - but that might mean you have 400GB (8 months) of empty space at the start of the file(s) and 50GB at the ends of the file, blocking the shrink. Maybe you just need to wait a while until Oracle has started reusing space the start of file for the next partition(s) and the time comes when you (or the admin package) can drop the partitions that are currently blocking the shrink.
There is an odd glitch in the "clean_audit_trail" procedure, by the way. If you give it a timestamp it deletes all data prior to that timestamp - but it's clever enough to drop any complete partitions that have a high_value less than that timestamp; however, if you give it a timestamp that is exactly the high_value for a partition the code deletes all the data in the partition when it could drop the partition - so you need to add one second to the high_value if you want to drop the partition rather than working hard to delete everything.
Regards
Jonathan Lewis
Re: Unable to locate PUM 53 and PUM 54 delta
Hi Safi,
Welcome to the Financial Management - PSFT (MOSC) Community and thank you for posting your question!
You can review the KM doc id - PeopleSoft Update Manager (PUM) Home Page (Doc ID 1641843.2) - then go to the FSCM Update Image Home Page link and under the Virtual Box - you will see the FSCM 9.2.053 Update Image Contents List which includes all the bug fixes for all PUM images.
I hope the information provided has helped. If your question has been successfully addressed, please mark this reply with a YES on the question listed at the bottom of the post ('Did this answer the question?'). Only the user who created the initial question can perform this action. This will enable other Community Users to quickly identify the solution suggested, and see that this post has been successfully answered.
Regards
Vaishali N
Re: The attribute SUPPLIER_INVOICE_AMOUNT has invalid usage. Approver list may not be genarated
Hi,
Even if AME rule does not use the attribute in condition, all attributes would be evaluated first and then conditions satisfying would be evaluated.
So, as mentioned, correct the query used on these attributes to avoid the exception.
Regards,
Sirisha


