FAQ

How do I run a VSAM .odz batch pack from the CLI versus the Batch service?

Export batch settings from the Tables menu when CoreIQ mappings exist. One-shot: java -jar sqol-batch.jar path/to/pack.odz. Service mode: start with no ODZ argument (default port 8089) and POST packs; only one job runs at a time (409 if busy). The .odz includes a decrypted CoreIQ password — treat packs as secrets.

How do I generate UI, models, and OpenAPI from a Markdown spec?

Open the process in Process Designer → Spec, write or paste Markdown, then Generate. AI must be enabled and the Spec tab must have content. Generation can overwrite existing UI, models, and OpenAPI — confirm the warning. This is separate from Trail AI analysis on import.

How do I bootstrap a process from an industry template?

Create from Industry Template: pick industry (for example Banking, Insurance, Healthcare), category, then process. Create generates UI spec, OpenAPI, models, Markdown docs, and mock data immediately. Treat it as a starting point, then edit in Process Designer or refine with Generate from Spec.

When should I use MOCK, OL_HUB, or an external OpenAPI provider?

MOCK returns schema-based samples for frontend-first work. OLHUB calls live Hub / legacy-connected APIs. OPENAPI points at any external REST API from a spec (auth plus metadata cache). Flip the provider without leaving AppIQ; test the operation in Manage APIs first.

How does AtomIQ turn COBOL + BMS maps into a Hub terminal Screens Flow?

AtomIQ analyzes COBOL plus BMS map source and can upload a Hub terminal Screens Flow for OL Integration — a typed Hub method that automates a live 3270/5250 session. The in-product map preview is from BMS/DSPF source, not a live terminal. That upload artifact is a terminal Screens Flow, not a Screener or Next map Scree

How do I finish a spec in generator chat when it keeps asking open questions?

When the last reply has an Open questions list, use Copy all, Append to message, or Suggest (the model proposes answers). Send Final instructs the model to self-suggest defaults and omit Open questions. Reference files attach to the next send only (up to 12 files, 20 MB each, 48 MB total).

How do I play a UI flow, a service operation, or the 3270 terminal player?

The operation player runs a generated UI model or sends JSON to a logic operation. App Player (/workspace-player) lists processes and launches operations. The 3270-style terminal player needs workspace Legacy type (Mainframe / AS/400), imported BMS maps, and Support legacy on. That terminal player is converted-map play

How do I enable a plugin and run Setup versus workspace build versus operation Execute?

Enable the plugin on the workspace and fill connection values (or copy Settings → Plugin defaults). Run Setup once for scaffolding, then workspace actions for a full build, or Execute on an operation (With edits vs As imported when baselines differ). Download last run as a ZIP or View files in the artifact tree.

What do I need to drive live CICS/BMS maps over IBM MQ?

Set Global Settings: queue manager, channel, request queue, and reply queue. The host needs OLGATE SEND/RECEIVE and the target transaction started (for example AQ00). One remote session per process. Without those, the Interactive Remote Player has nowhere to send maps.

How does OAuth2 work with AS/400, mainframe RSE, or the mock provider?

Sign-in validates against the configured provider (IBM i security, RSE API, or mock that accepts any credentials). A JWT access token is issued and used on later API calls; refresh is automatic. Configure providers in auth-config.yaml. Mock is for demos — not production.

How do I apply a theme so generated UIs match our brand?

Open Themes for the application, create or duplicate a theme, edit CSS and images with live preview, then set default and activate. Default themes cannot be deleted. Process Designer UIs render under the active theme.

How do I expose DB2 tables as REST APIs from Populate APIs?

In Populate APIs, select tables from a DB2 schema. CoreIQ generates CRUD, list, and count REST endpoints, publishes them from the DB2 catalog, and serves them through the dynamic OpenAPI. Publish is required before callers can hit the endpoints.

How does Dataset CDC capture VSAM changes, and where do events go?

Dataset CDC watches VSAM insert, update, and delete, decodes records with catalog layouts, and publishes structured events to Kafka or HTTP. This is the generic VSAM CDC path, not the BMC AMI IAM-specific CDC path.

How do I page and filter dataset content (range, keys, JSON versus RAW)?

POST or GET /dataset-content with limit/skip, or range (1-based; range ignores skip/limit), or keyedPagination on unique-key KSDS. responseFormat RAW is a binary stream; JSON is records with id, position, and base64. Server-side filters need fields plus filterExpr. 409 means exclusive open by another process.

What does the AS/400 workbench cover versus OL Terminal or DataIQ?

The CoreIQ IBM i workbench is for SQL, jobs, spools, libraries, program calls, and FTP in the browser. Use OL Terminal for live 5250 emulation and DataIQ for query/schema/Hub data movement. The workbench is host operations, not a green-screen session and not a SQL console.

How do I browse VSAM datasets, attach copybooks, and publish dataset APIs?

Browse the dataset in the Datasets catalog, associate COBOL copybooks as layouts, then publish. Published layouts become callable dataset APIs. Browse alone does not expose an API — publish does.

How do I create a VSAM KSDS or ESDS cluster through the gateway?

POST /dataset with datasetName, vsamType (KSDS or ESDS), key length/position for KSDS, recordLength, and space. IDCAMS DEFINE CLUSTER runs as the caller’s RACF user and still requires the write surface. Success is 201. The caller must be allowed on the HLQ/volume/SMS policy.

How does BigID GRA consume CoreIQ Datasets, IMS-DB, and IMS-DLI?

CoreIQ exposes Datasets, IMS-DB (Universal JDBC), and IMS-DLI as BigID Generic REST Adapter data sources so BigID can enumerate and scan those catalogs. Wire GRA at the CoreIQ surface that already holds the approved layouts — do not point BigID at a raw host catalog.

How do shared business models show which operations use them?

Open /business-models and expand Used by for I/O role (Input / Output / Internal) and CRUD usage. Delete is blocked while any operation references the model. A Models section in the spec renders cards with a JSON tree and clickable cross-references.

How do I run AtomIQ as an IBM i / RPG platform instead of z/OS?

Switch the workspace to AS/400 platform mode. That surfaces JT400 import, DB400, DSPF map preview, and the RPG analysis pipeline, and hides z/OS-only paths such as CICS, RSE, and DB2. Use this when the estate is IBM i, not when you only have a few RPG members on a z/OS workspace.

Where do workspace roles come from, and why can’t I delete one?

Roles are collected from a Roles section in saved Draft specs — you do not create them on the catalog page. Operations show Primary / Secondary tags. Delete only after you remove every reference, save, then delete. Add or rename a role by editing the spec.

How do I generate regression tests when nobody documented the legacy operations?

Identify and save operations first, then use Regression Tests & Legacy Unit Testing. AtomIQ exercises the operations, records the answers, and writes a regression suite, unit tests, and a spec file you own. Use Operations & Domain Classification so tests follow business-process boundaries rather than program names.

How does Data Lineage trace a DB2 column or file field through COBOL?

Data Lineage follows a DB2 column or file field through COBOL host variables across programs and copybooks. Start from the field, not the program name, when you need impact analysis before a schema or copybook change.

How do I turn a successful query into a Hub API, ETL flow, or DB Poller?

Configure Hub URL and API key in DataIQ Admin and test connectivity first. From a successful SELECT you can create the query in a new Hub API project, a new DB module, or an existing DB module. Export Dataset → Create Hub ETL when you need a target connection and optional clear-existing-data. Export Dataset → Create Hu