IBM Z ChatOps v1.1.0 has been officially released on November 3rd, 2020. As a new member of z systems management offerings products, in what scenarios can Z ChatOps be used, how does it function when solving problems, and how can you interact with it? Let’s find out the answers in this post.
IBM Z ChatOps Use Case：
Let’s first take an example of traditional mainframe DevOps working procedure. When the DevOps team receives a system warning and finds that the CPU used percentage in the Z system is too high, the team may need to contact many different roles such as the operator, administrator, technical expert, or even notify other teams to help check the resource status information during troubleshooting. During the process, the team needs to make separate calls, write emails, open tickets to find the corresponding contact, and open multiple interfaces on multiple software platforms to operate. It is inevitable that the communication gets inconvenient, time-consuming and labor-intensive.
Now, IBM Z ChatOps can provide a lot of assistance to the DevOps team, making it more convenient for the team to collaborate. Let's take a look at how the DevOps team that has integrated with Z ChatOps will deal with the problems mentioned above.
First, when the problem of high CPU used percentage in the Z system occurs, IBM OMEGAMON will call the incident inbound web hook provided by Z ChatOps to send the incident to Z ChatOps. After receiving it, Z ChatOps will post the incident directly to the chat tool channel , for example, Slack channel, where DevOps team members are in so that they can check the detailed information of this incident.
Annette, an operator in the DevOps team, notices this incident from IBM OMEGAMON, and learns that the CPU used percentage of some job is too high, which may cause slow response of many other applications. It is necessary to find the corresponding automation application and stop it as soon as possible.
Annette then types and sends a command in the chat tool channel, “@chatbot lpar list status”, and gets the LPAR status information. She clicks the drill-down button “Show jobs of LPAR” and selects the target LPAR.
The chatbot responds with the jobs of the selected LPAR.
Annette finds out that it is the job BETA92CL whose CPU used percentage is very high, so she types the command “@chatbot application list status BETA* --dn 'KEYAPLEX INGXSGUG' --sn KEYA” to check the automation status reported by IBM Z System Automation for the corresponding automated applications. She realizes that the error is caused by BETA92CL application which needs to be stopped in IBM Service Management Unite (SMU).
Annette mentions the System Automation expert Michael in the chat tool channel and asks him to check and potentially restart the BETA92CL application.
Michael sees the message and reviews the previous chat history in the chat channel. He clicks the “Show resource details” button from the previous chatbot output visible in the chat channel to check for the detailed information of BETA92CL application. At first sight, the automation status for BETA92CL looks fine. He clicks the link to open it in the SMU dashboard to get further details.
The browser automatically opens the Automation Domain page in SMU and selects the BETA92CL application.
Michael opens the Properties of the BETA92CL application in SMU to look at some detailed automation status properties, looks for errors in the captured messages, and verifies the defined automation dependencies. The shown information confirms Annette’s assessment and it is fine to restart the application. Hence, he requests to stop the application from the context menu. He then right clicks the resource and selects “Copy as Link” in the menu and gets the launch-in-context link of BETA92CL application. After that, he pastes it in the chat tool channel.
The chatbot displays the detailed information of BETA92CL application in the channel.
Now, Annette sees the result and checks the LPAR jobs with the sentence “@chatbot show me all lpars” and selects the target LPAR in the drill-down menu of “Show jobs of LPAR”. The CPU used percentage of all jobs is back to normal. The problem is solved.
This is just a common scenario where Z ChatOps can be used. During troubleshooting, the DevOps team members can invite technical experts or management members to join at any time to follow up the problem-solving progress in real time. Newly joined members can view the historical chat records to keep abreast of the latest information status.
This collaboration model has greatly enhanced the DevOps collaboration technology and process, reduced communication costs, improved the timeliness of problem isolating and solving, and effectively saved manpower and resources.
Four Ways to Interact with IBM Z ChatOps
Z ChatOps provides four interaction methods, let's take a look one by one!
- Chat with the chatbot. Users can use Simple Formed Sentence, such as “@chatbot show me all sa-domains”, or commands, such as “@chatbot system list status –dn ‘TESTPLX INGXSGA0’” to ask for information in a chat.
- Mouse navigation. When users chat with the chatbot, most of the responses from the chatbot are returned with buttons, links or drill-down menus. Users can click the buttons or certain elements in the result with the mouse to get more detailed resource information.
- Automatically resolve URLs and message codes. Users can send launch-in-context links from Service Management Unite that reference automation domains, systems, applications, monitored sub-systems and more to their channels in the chat tool Slack or Mattermost. The chatbot will automatically recognize the link and display the detailed information of the target resource in the channel. With this feature, when users see that the domain, system or application status is abnormal in SMU and need senior colleagues to help find out the problem, they can share the launch-in-context link of the corresponding resource to the channel. Senior colleagues do not need to switch windows. Instead, they can point out the problem directly on the channel. Users can also send message ID into the channel, for example, ING001I. Z ChatOps will identify the message ID prefixed with ING or HAS, search for information related to this ID in the IBM Knowledge Center, and display the searched link to the channel.
- Inbound web hook. Z ChatOps provides an incident web hook (i.e., REST API endpoint) that other products can call to send incidents and warnings to Z ChatOps. The chatbot will send the received incidents or warnings to the designated chat channel, displaying detailed information about the incidents or warnings.
IBM Z ChatOps is now included in the following products and suites:
IBM Z Service Management Suite 1.7 or later
IBM Z Service Automation Suite 1.3 or later
IBM Z Monitoring Suite 1.2 or later
IBM Z System Automation 4.2
Welcome to visit the following page to learn more about IBM Z ChatOps!
IBM Z ChatOps doc site：http://ibm.biz/ZChatOpsDoc
Customer access portal：https://www.ibm.com/support/pages/node/710583
Demo video: https://mediacenter.ibm.com/media/Chatting%20with%20IBM%20Z%20ChatOps%201.1.0/1_96vwzgvy