Debunking common myths in CA Log Analyzer
Myth 1
I was a CA Log Analyzer user not so long ago, and the first thing I heard when I started using it for DML reports is that I forcefully needed to have my tables with DATA CAPTUTE CHANGES in order to see the full results from my REDO/UNDO reports.
Wrong!
I was later shown that my view was very narrow and there's a way to overcome this issue by using Image Copy level of detail, which is another option in the DML Activity Report Options:
By using this option, CA Log Analyzer will leverage the information contained in my previous image copies and reconstruct all information it needed to give a full report even if my tables were defined with DATA CAPTURE NONE.
Myth 2
Then, after learning about Image Copy of detail, I believed that using Image Copy level of detail in my reports would have complete dependency on the existing Image Copies, and so, I wouldn't be able to report on recent events that occurred after an Image Copy was taken (i.e. the end range from my report would be current time or very close to it).
Wrong! (again)
CA Log Analyzer has the answer to this situation as well, but how?
Introducing Partial Update Reconstruction (PUR)
CA Log Analyzer can use the same tablespace VSAM to compensate for the information it needs when reporting on objects defined with DATA CAPTURE NONE, this is called Partial Update Reconstruction (PUR).
What do I need for this and how does this happen?
When the reporting range is the current time or close to it, or an image copy was taken shortly after the reporting range, CA Log Analyzer can take advantage and use the current VSAM or this aforementioned copy made after the end range.
This new processing also brings added benefits such as a shorter execution time for your jobs.
The following illustration shows the concept:
What do I need to turn this feature on?
CA Log Analyzer in Release 20, and the following PTF numbers applied:
SO08802
SO08803
SO08804
How can I turn it on?
There are several ways to do so:
- By a new PARMLIB option called PURSRC, with the following possible values:
PURSRC(A | B | U)
- A: PUR is activated and CA Log Analyzer will take advantage of this feature.
- B (default value): PUR is disabled. CA Log Analyzer continues to operate as usual, with no changes.
- U: CA Log Analyzer will select the most efficient way to obtain the information source for your report.
- Can also be activated through syntax if you want to use it in individual jobs:
PURSRC(AFTER | BEFORE | AUTO)
- AFTER: PUR is activated and CA Log Analyzer will take advantage of this feature.
- BEFORE (default value): PUR is disabled. CA Log Analyzer continues to operate as usual, with no changes.
- AUTO: CA Log Analyzer will select the most efficient way to obtain the information source for your report.
- If you're generating the job through panels, there is an option in the panels Profile section too:
From the main menu, go to Profile > Log Reporting Profile Variables (option 2), and on the bottom right corner you will see the following option:
Where the same options apply:
B (BEFORE, default value): PUR is deactivated and CA Log Analyzer performs as usual.
A (AFTER): PUR is activated and CA Log Analyzer will take advantage of reading the tablespace VSAM for your reports.
U (AUTO): CA Log Analyzer will determine the most effective option for you.
We hope you will find this information useful. We will be posting more content to inform you about useful features within our Backup & Recovery products for Db2.
If you would like to know more or talk about how we can meet your current and future needs, or you have something in mind that you would like to see in your products, don't hesitate and reach out, we're here for you.
Javier Estrada Benavides
Product Owner for Backup and Recovery Suite for Db2 for z/OS
Mainframe Division | Broadcom