One of the things we've been wanting to do is to be able to automate off of different highlighted messages without actually knowing what the message is. For example, we'd like to be able to create some generic automation when we receive a warning or alert message to allow, if nothing else, the automation team to know so we can contact the product owner. If there's something we can do to automate or expedite notification to the proper people, we want to try to do that.

We're missing the boat. We don't have a graphical display and we need something beyond just pretty PC pictures for the other departments. As a manager, I would love to have a dashboard that shows the problem and that the automation is taking care of it. The systems management and the computer operators would love to have a pretty graphic to display everything when it went down and to resolve the issue.

The graphical interfaces need to improved. It's good for a technician. However, when I have to go and talk to my management or a director or somebody about it, it is very hard to explain. It would be so much better if there was just a picture or a dashboard showing the problems of what is breaking or what is wrong, perhaps a type of roadmap during the initial load.

I just installed version 12.3, which actually already has a number of new features. It's going to make it easier for some of our new users.

I'm old-school, I learned it the old way. I actually want to go back and implement it and start doing things the new way. As I need to train people, it's going to make it easier for them and make the learning curve much shorter.

The new release addressed some shortcomings of the product. It was old-school. The learning curve was a little bit long in the beginning, but it's really come a long way.

Definitely having more web-based interface. Definitely be more mobile-open. More APIs, more open source to it.

I think, for me, I’d like to see more of its web base. The ISPF dialogue needs a little bit more features. However, the web base will be even more important for me.

It needs to have more compatibility/interfaces to other vendors’ products.

The only issue we have is when we try to integrate with common services. That's why it doesn't get a perfect rating. The OPS/MVS product itself is really good.

We have asked CA for things, including the ability to test multi-line WTOs. This is on their backlog. We have several of those that we would like to monitor and we can’t test our code before it goes in and actually runs.

Our last main complaint was having sysplex-wide variables (outside of OPSHFI), but this is now a reality.

If you look in the v12.2 command and function reference manual, there are sysplex variables. As in, their scope is sysplex-wide. If you update one on system A, it is accessible from system B.

Prior to sysplex vars, the only option was to use OPSHFI (also documented in that manual), which uses a dataset that is accessible from all systems. There are, however, still some restrictions with sysplex vars as far as which kinds of rules you can use them in, which is also the case for OPSHFI vars.

• SSM - when OPSMVS is started, there is a choice of either assuming all microhooks are up or all are down and working from there. It would be better if there was a feature that could discover the true state of every microhook.
• OPSLOG - merged OPSLOGs from multiple systems in a Sysplex, live, would help in problem determination.

There is always room for improvement and new features, and we like new features, so if CA delivers them, we are happy to use them.

Some of the commands are long. For example, to start a job, instead of just doing "s" plus space, to start a task, you have to type in "ssm", space, "sj", space, "job equals." That's how I do it. I use the commands. I know that you can go into a panel and kick things off in a different way. Maybe that would be easier.

The further expansion of the mainframe teamcenter interface. Also, the web services support is something that I am looking into.

Though, it currently meets my needs perfectly the way it is.

The GUI interface could be improved, but that is coming. They are working on that.

It does everything we need right now, I can't know of anything else we really need at the moment.

If I really had to put my finger on anything, mainframe tools are pretty old as far as the pricing models. Maybe that could be improved. Should I be able to buy whatever I want, and then I pay for what I use? Or maybe I don't pay anything for the product, and I then pay for the support. More of an open-source model.

I think some of their built-in functions are short on functionality. We have written homegrown REXX functions to perform many of these.

• There is nothing in place to ensure that your rules are consistent across all machines in an environment. I have worked at large (20+ LPARs) to gigantic (200+ LPARs) shops. The key to having good automation is to make sure all the rules and REXX execs are the same everywhere. Otherwise, you lose track of changes and differences very quickly. Obviously, some rules might only apply to one system, and we code them accordingly. (CA says that they are working on this currently, and I have seen a demo which looks very promising).
• To go with the previous topic, we have a function called PLEX that returns the current logical sysplex that the rule or REXX exec is running on. Currently, with the provided tools, you would have to code a long select and list each system.
• There is no function within OPS/MVS to open problem tickets to an outside product. Currently, our problem management tool is ServiceNow. We automatically open problem tickets and assign them to the appropriate group all automatically. This function I did demo for other companies at CA World in 2010. Some were very interested in what we have done.
• There is one built-in OPS/MVS function that actually does not work at all. If you are using OPSVALUE to retrieve global information cross-system, there are times that it is not possible to get the information. The reason is you have to pull the data one level higher than what is needed, and you are limited to a number (I do not remember exactly what the value is) that can be returned cross-system, and it is less than what is retrieved from your request. We wrote a function that will look up global information on the system where they reside and return only the info that is needed. This one I have opened a problem ticket on, and have posted it to the CA website.

These are the biggest issues. Some of these I have opened issues on, and others I have just posted online. So far, the only one that seems to have gotten attention is the first one.

I am not using all the stuff that I could use now, so I do not know. It is working very well for us.

There is just so much in there that I am not utilizing, though I might not need to utilize. It is just not knowing everything in the product.

I've seen a mobile app that allows you to get the system status from your phone. That's a really nice feature that I’d like to see.

I don't know of anything that is required to be added.

I'm in a sprint review with the next version, so I've seen some of the things that they're coming out with. It's going to very nice. I like the things they're coming up with.

The mainframe TSO side still displays green screens. Today, people are more into web and things like that. I would like to see that type of interface available for them.

Use something other than CCI for external communications.

• The interfaces to USS, CA UIM could be better.
  
• CA Common Services interfaces - namely Zdata Mover, Zmessage Service, Zdatacolletcor - are poorly documented.

One thing that comes to mind is the MQ interface. The last time we tried to use it, it seemed a little clunky.

Also, performing a stop/restart of OPS/MVS could be a little smoother too. Without special coding, some undesired tasks will be started and some necessary tasks will be stopped if we have to bounce OPS/MVS during the day.

It more or less does everything we need for it to do.

It’s a complex product. The initial installation is easy, but implementing various functions is a constant process.

I’m not aware of any complaints from my team; I hear nothing but good things. It was, however, a complex setup in the mainframe. Setup should be made much more straightforward.