Skip navigation
Share:|

If you are doing or thinking about moving towards DevOps, there may be a conundrum you have to deal with. Some, perhaps many, of the enterprise tools that currently are prevalent in Ops, don’t fit well (I’ve been told I have a talent for understatement) into the Dev “as code” model. That world revolves  around code management systems like Git, automated build tools like Jenkins and configuration management tools like Chef or Puppet. Among those enterprise tools may be workload automation tools that manage your critical business applications.

 

On the other hand, the tools from the DevOps ecosystem that do fit well into the “as code” model, may be immature and basic (after all, building Rome was not the only project that took longer than a day) and may require developers to spend an inordinate amount of time scripting and integrating and instrumenting. That might be OK if they didn’t already have a day job or if they could miraculously develop sophisticated instrumentation that has taken the industry decades to evolve.

 

We believe the best approach is to take the leading solution for automating business applications, one that has already proven itself to be highly flexible and adaptable through decades of evolution from mainframe to distributed systems to big data to business-aware monitoring to self-service, and evolve it further to embrace and support the DevOps ecosystem.

 

That is the concept of Jobs-As-Code as implemented by Control-M. Take the best operational instrumentation for batch workflows, expose its functions via RESTful APIs, enable object definition using a familiar notation for construction (JSON), provide support for programmatic provisioning to support automated testing and generally make it simple to embed into whatever CI/CD tool chain you are already using. Make sure that all of the existing capabilities that have made this solution the leader over decades (the ONLY solution in the Leader Quadrant in Gartner’s Magic Quadrant since 2001) are fully accessible and dovetail seamlessly with the newly added automation capabilities and voilà. Done and dusted.

 

We’ll be showing Jobs-as-Code at the DevOps Enterprise Summit in San Francisco Nov. 7 – 9. Come by and take a look.


 

Share:|

-by Joe Goldberg, Solutions Marketing Consultant, BMC Software Inc.

The November 19th webinar titled “Improve Service Availability and Compliance with Workload Automation”, discussed how a recently released licensed add-on, BMC Control-M Workload Archiving, can help speed up troubleshooting and problem analysis, simplify auditing and compliance reporting and give application developers better insight into the operation of existing applications when changes are planned. The recorded event is available here.

 

There wasn't sufficient time to answer all the questions that were submitted and a few required some research. This blog provides answers to all of the questions including those that were already addressed during the webinar itself. If you have any additional quesitons not addressed here, please feel free to comment on this blog, contact me directly (joe_goldberg@bmc.com) or reach out to any person at BMC.

 

Question:     Any plans for archiving CM Server or agent PROCLOG files?

Answer: There are general plans to extend Workload Archiving to include other data that are logical extensions of historical information. However, there is nothing definite that can be stated at this time. Please note that if you wish to add other information, such as proclogs for example, you can create a Control-M job in which you invoke a standard utility such as cat/grep/copy, etc. any data you wish to stdout. If you then give that job a recognizable name (e.g. ProcLogDump), its output will be archived like any other job and you can locate such jobs based on your local convention. Watch BMC Communities for future postings with some samples.


Question: In the archive policy window you can set the max output size. If the output exceeds that size are you alerted? If not how would I know that all my output has been saved?

Answer: The current approach is that when data retrieval fails for any reason, an entry is created in the Archive DB but the data contains only the error message that was received by Workload Archiving.

 

Question: Would Arching collect audit data?

Answer: Currently, Workload Archiving collects only the Control-M Log and the job Output. See suggestions above for how you can process audit data today.

 

Question: How does this product work with the Control-D product?  Is this a piece of Control-D behind the scenes that is just for Control-M?

Answer: Workload Archiving is a completely new and separate solution that utilizes Control-M/Enterprise Manager facilities to retrieve its data. Control-D and Workload Archiving can coexist in the same environment. Typically, Control-D is used for managing business output such as invoices and customer statements, providing broad services such as decollation (splitting large reports into logical segments) bundling (collecting multiple reports into a single delivery object; either physically printed or electronically grouped), archiving and transformation from printer-specific data streams to electronic formats like text of PDF. Control-D is a mainframe-based solution and has its own user interfaces. Workload Archiving is aimed specifically at managing Control-M data that is used for problem analysis and compliance and is accessible from within Control-M user interfaces.

 

Question: I did not see a component for Archiving in the Configuration manager. How do you start and stop the archiving component?

Answer: Control-M Configuration Manager (CCM) Support for Workload Archiving will be provided in future releases (probably Control-M V9 subject to the standard roadmap disclaimers). The current version of Workload Archiving is delivered with its own start-up, status and shutdown scripts that can be run during server start-up or manually as required.

 

Question: If your Control-M products are installed on the Linux platform, since this is on a separate server/database, should it be installed on the same platform, or could it be installed on the windows platform for example?

Answer: Workload Archiving can be installed on any of its supported platforms regardless of the platform chosen for your Control-M infrastructure. Customers usually select platforms based on organizational familiarity or preference and Workload Archiving can accommodate such choices.

 

Question: How does it display if a job that has failed has output that is archived, as that appeared to be an option, but it wasn't evident on the screens displayed.

Answer: When building Archiving Retention Rules, you can specify whether that rule applies to jobs that end OK or NOTOK. This allows you to keep data from failed jobs longer (or shorter) than data from the same jobs that end successfully. Currently however, there is no indication in the job, which rule was used to process the data.

 

Question: When it comes to the Workload Archiving, is this an out of the box implemented solution or does it take time to implement Workload Archiving.

Answer: Workload Archiving was designed specifically to be very simple to implement. The installation takes only a few minutes and all that is required to start archiving output is the activation of at least one retention rule (a default is provided with the installation). That means you are archiving data within minutes of installation and for most environments, no changes are required to the current configuration. If you are using Control-M/Server security or have z/OS you will have to authorize a single user (ARCUSER). Of course this may vary for individual organizations as additional time may be required to determine which data for which applications should be retained and for how long. This is purely subjective and depends entirely on the needs of your organization.

 

Question: I did not see any new components during the demo in the CCM, will it show the connection to this tool?

Answer: Yes, it will be provided in future releases. See answer above

.

Question: When archiving is turned on, is the output only saved on the archiving server or is also on the agent?

Answer: Archiving does not affect the normal retention configured within Control-M. Both Log and Output continue to be retained according to their configuration settings. The data captured by Workload Archiving is a copy only.

 

Question: When you are implement archiving how much is workload archiving able to grab, is it able to grab for the previous 10 years or do you have to manually put that in the system

Answer: Currently, Workload Archiving captures only new data. There are no import capabilities to load previous data into the environment.

 

Question: When getting the SYSOUT and LOG from Z/OS System there is a default user which has to be defined in RACF. How can I have another User which will get my Sysout from z/OS?

Answer: Workload Archiving connects to the EM as user “ARCUSER” and that user must be defined on the z/OS side in order to retrieve data from the z/OS environment.

 

Question: Did you say it was ok to install Workload Archiving on same system as EM and CM/Server?

Answer: Yes, technically Workload Archiving can be installed on the same host as the Enterprise Manager or any other Control-M component. It is recommended that you consider the overall load on the resources of that host in order to determine if it will be acceptable to add Workload Archiving.

 

Question: Is the output / log data stored within the database or the file system?

Answer: The data is stored within the database.

 

Question: From a reseller standpoint what are the most important things to relay when talking about this product?

A Workload Archiving accelerates troubleshooting, reduces the cost of managing Log/Output data, simplifies auditing and gives application programmers greater insight into the execution patterns of existing jobs/applications. These are the primary areas where Workload Archiving adds value for workload automation users. Your customers will likely guide you in the direction(s) that are of greatest value to them.

 

Question: Can the Archiving database be created within the same database server as the EM database?  Can the database be Oracle, Postgre, or MS SQL?

A The current version of Workload Archiving defines its own database and is exclusively PostgreSQL.

 

Question: How is it licensed?

A BMC Control-M Workload Archiving is a priced add-on to Control-M Suite or Select. Please contact your BMC Sales Representative to discuss pricing details.

Filter Blog

By date: