Hi all,

In ibm_zos_zosmf/roles/zmf_workflow_complete/tasks/main.yml

Lines 16 and 63:
workflow_host: "{{ inventory_hostname | default() }}"

This prevents me to override the workflow_host variable when using the role.
Should this not be workflow_host | default() with defaulting to inventory_hostname?

Kind regards,
Tobias

(https://github.com/IBM/ibm_zos_zosmf/blob/master/playbooks/sample_role_deploy_cics_application.yml)

sample_role_deploy_cics_application playbook fails with the below error

TASK [cics_wlp_install_app : Install CICS application - Write application binary to USS file] ****************************************************************
Wednesday 01 July 2020  01:24:32 -0400 (0:00:00.622)       0:02:34.006 ********
fatal: [zosmf1]: FAILED! => {"cache_control": "no-cache=\"set-cookie, set-cookie2\"", "changed": false, "connection": "Close", "content": "{\"category\":6,\"rc\":8,\"reason\":93651005,\"message\":\"File not found.\",\"stack\":\"  1 20506E58+000001D4 PosixFileControl::open(const char*)\\n  2 205053C8+00000472 PutFileServlet::doPut()\\n  3 2050C788+000000CE TsoServlet::service()\\n  4 205399F8+0000019E TsoServlet::run()\\n  5 20539350+000003A0 main\\n  6 0626DF90+0000127C CEEVROND\\n  7 1FCFD280+000000B4 EDCZHINV\\n  8 06103978+000001C6 CEEBBEXT\\n\",\"details\":[\"EDC5129I No such file or directory. (errno2=0x0594003D)\"]}", "content_language": "en-US", "content_length": "464", "content_type": "application/json; charset=UTF-8", "date": "Wed, 01 Jul 2020 05:24:59 GMT", "elapsed": 0, "expires": "Thu, 01 Dec 1994 16:00:00 GMT", "json": {"category": 6, "details": ["EDC5129I No such file or directory. (errno2=0x0594003D)"], "message": "File not found.", "rc": 8, "reason": 93651005, "stack": "  1 20506E58+000001D4 PosixFileControl::open(const char*)\n  2 205053C8+00000472 PutFileServlet::doPut()\n  3 2050C788+000000CE TsoServlet::service()\n  4 205399F8+0000019E TsoServlet::run()\n  5 20539350+000003A0 main\n  6 0626DF90+0000127C CEEVROND\n  7 1FCFD280+000000B4 EDCZHINV\n  8 06103978+000001C6 CEEBBEXT\n"}, "msg": "Status code was 404 and not [204]: HTTP Error 404: Not Found", "redirected": false, "set_cookie": "LtpaToken2=zdYQj/N7e5uP16tmx2nwig0nKYhJTXAKEdYjDPuejgx4DW4NSMz9JhVkuNLlnZ4RMObpG6OLnPZvxbIhlfP/ZNzETiq9uifFy4CMVS5/zQwYC/lbN8/Cs4bqwHI6eiiPPTm11O0fsztwvoIE0Ewk6ab1w8rWX2LvScYQWgAApBsVfZYry/GlTZKP73/ErxNjQsN4vusgAIpjgFNVURAIBNCSrzTQQYW81i3VfM32vmuPytJt+fb7atj2IWxNZwA/fyjy94YwwPuTNjBzE855x5dG+N14ri3O7NYg+4vLUOtsg2qoCUhDd3+IDW+YCF2s; Path=/; Secure; HttpOnly", "status": 404, "strict_transport_security": "max-age=31536000; includeSubDomains", "url": "https://xxxx:yyy/zosmf/restfiles/fs/workdir/DFHWLP/wlp/usr/servers/defaultServer/dropins/CloudTestServlet.war", "x_content_type_options": "nosniff", "x_ibm_txid": "tx000000000001A172", "x_powered_by": "Servlet/3.1", "x_xss_protection": "1; mode=block"}

Looks like the current logic is that if the workflow allows parallel execution, the ansible-playbook will default to parallel automation, and no way to override with the setting.

Looking for a parameter to specify what style of automation the playbook is to execute the workflow steps.

    - name: "zOSMF workflow"
      delegate_to: localhost
      ibm.ibm_zosmf.zmf_workflow:
        ...
        state: "started"
        action "start | startParallel" # <- example parameter
        workflow_owner: "xxxx"
        ...
        workflow_vars_file: "wf-var.properties "
      register: result

Good day team,

With the purpose of handling data (including job-log) containing Japanese characters, one of our clients is trying to set a combination of z/OSMF custom and standard headers (as mentioned below) while making use of various roles in ibm_zos_zosmf Ansible collection.

            'Content-Type': 'text/plain;charset=UTF-8',
            'X-IBM-Intrdr-Class': 'H',
            'X-IBM-Intrdr-Recfm': 'F',
            'X-IBM-Intrdr-Lrecl': '80',
            'X-IBM-Intrdr-Mode': 'TEXT',
            'X-IBM-Intrdr-File-Encoding': 'IBM-939',

            'Content-Type': 'application/octet-stream;charset=UTF-8',
            'X-IBM-Intrdr-Class': 'H',
            'X-IBM-Intrdr-Recfm': 'F',
            'X-IBM-Intrdr-Lrecl': '80',
            'X-IBM-Intrdr-Mode': 'BINARY',
            'X-IBM-Intrdr-File-Encoding': 'IBM-939',

We understand that in case of role - zmf_job_complete, header values can be assigned using a combination of role variables: -

1. job_jcl_location
2. job_internal_reader_mode
3. job_internal_reader_class
4. job_internal_reader_recfm
5. job_internal_reader_lrecl

However, there is no role variable defined to set X-IBM-Intrdr-File-Encoding.

Mentioned below are the queries we have with regards to current functionality of ibm_zos_zosmf collection: -

1. Is there a way/ work-around to explicitly set the values of z/OSMF individual (custom, standard) headers inside Ansible ibm_zos_zosmf roles?
2. Is there a way to set custom header X-IBM-Intrdr-File-Encoding to a specific encoding scheme inside zmf_job_complete and zmf_job_query roles? This is similar to setting value for encoding parameter for zos_job_submit module. In our case, the value is IBM-939.

PS: Our client is able to handle data containing Japanese characters by making z/OSMF RESTful API calls through a Python script and looking forward to emulate the same behavior using Ansible ibm_zos_zosmf collection. If you need to look into the functioning of Python script, please let me know and I'll share the specifics.

Many thanks.

IBM System Automation for z/OS and IBM Z Netview must be notified so that they can monitor products/environements provisioned by Ansible.

Is it planned to add samples that illustrate how Ansible should notify IBM System Automation for z/OS and IBM Z Netview?

Hi all,

would it be possible to add a parameter to the role zmf_job_complete that triggers a delete of the job from the spool of the target system? Zowe CLI and hence the z/OSMF REST API support that (just a simple DELETE on the job ID I think).

This capability is missing in general from this collection, but would help me a lot.

Cheers,
Tobias

In CONTRIBUTING.md, the Legal section has links to the hyperledger project, which I do not think are intended.

how can i enter a job statement on 2 records in an ansible playbook with z/OSMF collection ? i have looked at https://www.ibm.com/support/knowledgecenter/SSLTBW_2.2.0/com.ibm.zos.v2r2.izua700/izuprog_API_WorkflowCreateNew.htm which says:
"For a workflow that submits a job, this property specifies the JOB statement JCL that is used in the job. This property can be null, or a list of JCL cards, each up to 72 characters long. Columns 1 and 2 of each record must be "//" or "/*" and the job name must be 1 to 8 characters."
The variations i have tried were not sucessful, like
workflow_job_statement: "//IZUWFJB JOB MSGCLASS=X,REGION=4M,NOTIFY=&SYSUID,
// MSGLEVEL=1"
The 2 records are appended on the same 1rst JOB CARD....
Could you post a sample how to achieve that, having a JOBCARD on 2 records with a continuation line.

Hello,

As an avid Ansible user (and Mainframe Sysprog), I am very excited that IBM is working on Ansible content to interface with the mainframe. We're big z/OSMF users so this code seems like it could be a great fit. However, I was a little confused about your code structure.

Why are most of the z/OSMF functions (like zmf_job_complete) presented as YAML roles instead of Python modules? Having all the logic built into YAML makes it more confusing and harder to maintain than having a separate Python Module for it (as discussed here: https://www.ansible.com/blog/make-your-ansible-playbooks-flexible-maintainable-and-scalable). I looked at IBM's other Ansible Content (z/OS + IMS) and most of that code is in the module format. Other companies also tend to use Modules for API-driven functions (VMWare, JIRA, etc.)

Are there any plans down the line to convert roles like zmf_job_complete to a Python Module instead?

Thanks!
Dakota Heyman

ansible version: 2.9.0
ibm_zos_zosmf version: 1.0.0

When I run the sample_role_cpm_provision.yml to create the software service, the instance can be provisioned, but I got following error in task log:

TASK [provision_software_service : Provision template] ************************************************************************************************************
fatal: [S0W1 -> localhost]: FAILED! => {"changed": false, "content": "", "elapsed": 31, "msg": "Status code was -1 and not [200]: Connection failure: The read operation timed out", "redirected": false, "status": -1, "url": "https://xx.xx.xx.xx:443/zosmf/provisioning/rest/1.0/psc/CICS55/actions/run"}