Chat transcript of askCA: CA Test Data Manager - January 2018 which took place on January 30, 2018.
Melanie Giuliani to everyone: HI everyone - welcome to askCA! if you have any questions, we're ready to take them now! You can start by typing your questions into the chat window, and selecting "send to: everyone"
William Lindsay to everyone: Is the Profiling on the Mainframe still useful? Does it work and is it supported? Can you profile JSON, XML or RR pairs? This is for shops that are more weighted towards using APIS as opposed to DataModels.
Mahesh to everyone: Hi we are planning to upgrade out TDM Infrastructure, We are planning to get high-end servers and replace current infrastructure.
Mahesh to everyone: Need guidance form CA to plan this. what is the best way forward
Manjunath Narayan to everyone: @Mahesh, Please email me, I can help you to upgrade TDM..
Mahesh to everyone: @Manjunath Thank you, I will send email
Manjunath Narayan to everyone: @Mahesh, You can check the below link , we have lot of information on upgrade as well.. , But we can get connected in email..
William Lindsay to everyone: Are noSQL DB like DynamoDB or Mongo in the pipeline for Profiling?
William Lindsay to everyone: Can you ask the Deployment Packaging team to backup the OrientDB before installing. The test data models get deleted.
Debaditya Chatterjee to everyone: Hi @William. Yes we have plans to support MongoDB and other noSQL data sources. We are still working on the timelines
William Lindsay to everyone: Also there is a capbility in OrientDB to schedule backups, but is not enabled by default and it is usesful
Debaditya Chatterjee to everyone: @William Lindsay Thanks for your suggestion on backing up OrientDB
from Walter Guerrero to everyone: @William - Is the Profiling on the Mainframe still useful? Yes, profiling on the mainframe is still useful
Nora Kennedy to everyone: I have a question about data reservation, how can I see what people have reserved as an admin? I currently only see my reservations
Debaditya Chatterjee to everyone: @Nora Kennedy. This has been requested by our customers. We are currently working on it.
Nora Kennedy to everyone: @Debabitya Thank you! Is removing the ownership dependency on Oracle DB for data requests also on the roadmap for data reservation?
Debaditya Chatterjee to everyone: @Nora Kennedy Can you please elaborate the scenario you are encountering with oracle DB ?
Nora Kennedy to everyone: In order to use an Oracle database for data reservation, it requires the connection profile user to be the owner of the DB
Debaditya Chatterjee to everyone: @Nora This helps to get access to data. I believe we have the request in our backlog
Debaditya Chatterjee to everyone: @Nora. Please email me with the question and I will get back to you
Walter Guerrero to everyone: @William Data profiling works based on the connectivity profiles, where you will use the TDM portal to setup the data profiling, which will be scanned and a heat map provided to you.
William Lindsay to everyone: @Walter Hi Walter, Nice to chat with you again. Last question I have is is there a best practice for reducing MIPS on the mainframe for Subsetting, Profiling and Masking. Seems to be a common pet peeve regarding cost. IE. There are many options to do subsetting, masking on the MF, some functions or methods (like using seedlists) might have preference in terms of performance). Some approaches about ordering functions, or the use of variables. More shops are becoming concerned with performance and cost. For masking is File Masking more efficient than DB masking, would it be work export, mask, import as opposed to DB direct update.
William Lindsay to everyone: @Walter, the MF profiling is what I'm asking. The Portal Profiling only supports the identification of PII. I'm looking for some of the capabilities for looking at the actual data and storing it in the REPO, so it can be used in Data Generation etc... and we can do deeper dive on understanding the data model. I dont think you can get that from Portal Profiling
Debaditya Chatterjee to everyone: @William Lindsey This is a great question. We will try out the export --> mask --> import suggestion and see how well it works. Thanks for the suggestion
Walter Guerrero to everyone: @William, in regards to the profiling, we are working on this at the present time.
Nora Kennedy to everyone: @Debaditya Thank you! One last question with the permissions for 4.3, is there a way to make someone an admin over the project without giving them access to user roles for the project? I tried removing the permissions specifically for that and it gave me an error for the group. Does a project "admin" require that user have access to security settings? I was able to remove other options, just not security.
Keith Puzey to everyone: @Nora can you please email me with more details on what you are trying to do with the security groups and i will look into this.
Nora Kennedy to everyone: @Keith Puzey Will do!
daniel to everyone: Hi is there any restrictions if we try to mask plain text files from different OS's? Like from Linux, Mac to Windows.
Walter Guerrero to everyone: @Daniel, as far as we know there are no restrictions, Did you happen to run into a scenario, where you encountered restrictions?
daniel to everyone: @Walter - Not yet, we are just starting to use the software. Thanks for the enlightment
William Lindsay to everyone: @walter There was a limition of the MF subsetting in that it supported only 50/tables at a time. The tables had to be chunked out and ordered manually into several steps. Has that been addressed. I nkow someone was looking into it.
Walter Guerrero to everyone: @William, I will check on that and I will let you know, send me a follow on email
William Lindsay to everyone: Just a suggestion, Datamaker is dependant on GTService. GTService needs to be configured in multiple places. Could the SETUP for DM, prompt for the GTService value, and update the configs.
Peter De Zutter to everyone: @William Lindsay, I like that idea. Would save a lot of time
Melanie Giuliani to everyone: @William - may be worth creating an idea on the communities on that! Here's the link: https://communities.ca.com/community/create-idea!input.jspa?containerID=3136&containerType=14
Debaditya Chatterjee to everyone: @William Lindsey I will explore the idea on GTService and find out if it is feasible.
William Lindsay to everyone: @peter send me a $1.00
Peter De Zutter to everyone: :)
Debaditya Chatterjee to everyone: :-)
William Lindsay to everyone: @melanie ok, will do
William Lindsay to everyone: last ??
Melanie Giuliani to everyone: Go ahead @William! We're still here!
William Lindsay to everyone: When doing profiling and using count/Tag. Is TDM using uncommited READ. for DB2 that is with UR. for Oracle, SQL Server something else, cant remember
Melanie Giuliani to everyone: @William - the team is looking into this. Give us a minute or two to get back to you.
Keith Puzey to everyone: @William with the PII profiling on DB2 we don’t use uncommitted Reads
William Lindsay to everyone: @Keith using this would reduce MIPS, since DB2 would do no locking
William Lindsay to everyone: @keith it would avoid resource collision
William Lindsay to everyone: for Oracle and SQL server also
Peter De Zutter to everyone: @Keith, a very valid point from William, not my first customer who is raising that point.
William Lindsay to everyone: Have to go, thanks for putting this together and for those of you I know, always nice. Say hello to everyone
Keith Puzey to everyone: @William we will add this to the backlog for when we start work on the DB2 mainframe profiling
William Lindsay to everyone: @keith thanks
Peter De Zutter to everyone: I know a couple of customers who are looking forward to that
Melanie Giuliani to everyone: @all - we're going to go ahead and wrap up for today. Thank you so much for joining! The transcript from today's chat will be posted to the community late