TM1 Connection¶

OptimusPy uses TM1py to connect to TM1 / Planning Analytics. Connections live in config/config.ini, one section per instance, freely-named (the section name is what you reference as instance in cube configs and in the UI).

INI file format¶

[tm1srv01]
address=localhost
port=12354
user=admin
password=apple
ssl=False

[planning_sample]
address=localhost
port=12354
user=Admin
password=YXBwbGU=
decode_b64=True
ssl=True

Each [section] is independent. Add as many instances as you like (DEV, UAT, PROD, regional models, etc.).

On-premise (address / port)¶

[tm1srv01]
address=tm1.example.com
port=8010
user=admin
password=mypassword
ssl=true
async_requests_mode=true

Parameter	Description
`address`	Hostname or IP of the TM1 server
`port`	TM1 REST API port (often 8010 / 12354 / 5021)
`user`	TM1 user name
`password`	Plaintext or base64-encoded password
`ssl`	`true` for HTTPS, `false` for HTTP
`async_requests_mode`	`true` recommended for long-running queries

IBM Cloud / Planning Analytics on Cloud¶

[paac_prod]
base_url=https://your-tenant.planning-analytics.cloud.ibm.com/tm1/api/TM1 PROD
user=automation_user
password=your_password
namespace=LDAP
ssl=true
verify=true
async_requests_mode=true

Parameter	Description
`base_url`	Full PA-on-Cloud REST URL including the model name
`namespace`	Authentication namespace (commonly `LDAP`)
`verify`	`true` validates the TLS certificate (recommended)

Spaces in model names

base_url may contain spaces (e.g., …/tm1/api/TM1 PROD). Do not URL-encode them, because configparser reads the value as-is.

Encoded passwords (`decode_b64`)¶

To avoid storing plaintext passwords:

python -c "import base64; print(base64.b64encode(b'mypassword').decode())"

Set password to the encoded value and add decode_b64=True:

password=bXlwYXNzd29yZA==
decode_b64=True

Not encryption

Base64 is encoding, not encryption: anyone with the file can decode it. Use OS-level file permissions or a secrets manager for real protection.

Common parameters¶

Parameter	Default	Description
`session_context`	always `optimuspy`	Label that appears in `}StatsByActiveSession`. OptimusPy sets it itself, so a value in `config.ini` is ignored.
`async_requests_mode`	`false`	Recommended `true` for benchmarking large cubes
`connection_pool_size`	`10`	Connection pool size for parallel requests

Which config.ini is read¶

There are three ways to choose the file:

--config PATH on the command line, for one run: optimuspy scan --instance tm1srv01 --config /path/to/shared/config.ini, or optimuspy ui --config .... If the file doesn't exist, OptimusPy prints ERROR: config.ini not found: <path> and exits with code 1 before it connects.
A link from the Settings page. On Settings → TM1 Instances, give the path to a config.ini, or to the folder that holds one, and click Link to this file. OptimusPy then reads that file where it is, so edits made for RushTI or your own scripts show up without a second copy.
OptimusPy's own copy, config/config.ini in the folder you run from (for the bundle, the executable's folder). Create it from config/config.ini.example, or fill it with Copy into OptimusPy on the Settings page, which copies a file byte for byte, comments included.

The first that applies wins: --config, then the link, then config/config.ini.

The link is saved in config/settings.ini, so it's shared by the UI and the CLI: once you link a file in Settings, optimuspy optimize x.json without --config reads it too. If the linked file goes missing, the CLI prints ERROR: config.ini not found: <path> (linked from the UI's Settings page) and exits with code 1. The UI still starts, and Settings reports the missing file so you can link another one or switch back to config/config.ini.

OptimusPy doesn't edit a config.ini; the only file it writes is the copy Copy into OptimusPy makes. You edit the file in a text editor, or in the tool it belongs to. The Settings page shows each instance's fields read-only and has Test Connection.

Settings Page →