This guide covers first-time Cloud Mac access, remote desktop, SSH, macOS, Xcode, CI Runner, storage expansion, and order processing. Follow the registration board in order to avoid changing the network, credentials, and build environment at the same time.
If the issue spans multiple areas, start with the first connection and rule out each possibility. Change only one variable at a time, and record the symptoms and time before and after the change.
Confirm the node region, host address, username, credential version, and currently permitted access method.
Remote desktopDistinguish between session failures, slow display updates, input issues, and client configuration problems.
SSHVerify the connection target, credential permissions, local key selection, and verbose connection output.
Xcode environmentPin the macOS, Xcode, command-line tools, and dependency versions before addressing project-level differences.
CI RunnerCheck the Runner's online status, task labels, concurrency settings, cache directories, and available space.
Storage & networkEstimate space based on the project, caches, assets, and artifacts, then validate network dependencies separately.
Do not start by reinstalling the client or resetting the environment. Confirm the registration details first, then determine whether the issue is with the credentials, node, local network, client, or specific session.
Reread the order ID, node region, host details, username, and credential-generation status from the portal. Do not continue using old tickets, screenshots, or historical connection details forwarded by team members.
Check whether the node record in the portal has entered an accessible state. While delivery registration is still in progress, keep the current record; do not recreate the connection configuration or rotate credentials repeatedly.
Test host resolution and the target port on both your current network and another trusted network. Record the test time, network type, and result; do not use a single latency figure as a complete diagnosis.
Confirm that the connection protocol, host field, port, username, and selected credentials are correctly aligned. For remote desktop issues, also distinguish session establishment, display updates, keyboard mapping, and local display settings.
Keep the original error text, occurrence time, reproduction steps, and client version. For SSH, provide redacted verbose connection output; for remote desktop, record the complete sequence from starting the connection to the appearance of the error.
System upgrades, Xcode changes, and dependency updates can alter several variables at once. Before troubleshooting, record the current environment and validate the basic toolchain with a minimal project.
Record the macOS version, full Xcode version, current command-line tools path, and project requirements. The toolchain reported by the graphical interface and command line should match.
Before updating the system, Xcode, package manager, or build dependencies, verify project compatibility, free disk space, and rollback materials. Do not switch versions midway through a critical running task.
Keep source code, dependency caches, build intermediates, test results, and final artifacts separate. This makes it easier to determine whether disk growth comes from project content, cache-hit changes, or build artifacts.
Use separate directories, locked dependencies, and environment-variable sets for different repositories, branches, and experimental tasks. When an issue occurs, reproduce it with the same commit and dependency versions.
Check job pickup, signing environments, caches, concurrency, and disk separately. Changing Runner configuration, dependencies, and cache policies in one task makes the result difficult to reproduce.
Check the process status, task labels, project authorization, queue assignment, and most recent heartbeat. Confirm which repository or organization the Runner is actually registered to.
Verify that the signing materials are complete, permissions are correct, and the script uses the expected path. Provide only redacted errors and material types in the ticket; never submit sensitive credentials.
Compare the cache key, dependency lockfiles, hit records, and working directory. First run one minimal task with caching disabled to confirm whether the issue actually comes from the cache layer.
Check the number of simultaneous tasks, shared directories, ports, simulator resources, and artifact paths. Assign separate workspaces to parallel tasks to prevent them from overwriting one another.
Measure source, dependencies, caches, intermediate files, logs, and artifact directories separately. Before cleanup, confirm that no task is still using the relevant directory and save any required artifacts.
Include source code, dependency caches, build intermediates, test data, assets, models, and final artifacts in the estimate. Expansion options address specific capacity or daisy-chaining needs; they do not replace project backups.
To change an existing order, open the portal and submit a ticket linked to the order record. Do not change storage paths or disconnect a device in use while a task is running.
Large asset batch processing, workflows that clearly depend on high-speed external storage, and long-running tasks that need a dedicated work disk. First confirm that the software, file system, and task paths can use the daisy-chained devices correctly.
Keep separate copies of project repositories, signing materials, models, assets, and build artifacts according to your team's policy. Expansion solves workspace capacity needs; it does not replace version control or backup procedures.
| Add-on | Daily | Weekly | Monthly | Quarterly |
|---|---|---|---|---|
| +1TB SSD | $2.6 | $7 | $12.9 | $35.1 |
| +2TB SSD | $5.2 | $14 | $25.8 | $70.2 |
| Thunderbolt 5 daisy chaining, per device | $1.9 | $5.2 | $9.7 | $26.4 |
A clear request tells support which order and region are involved, when the issue occurred, how to reproduce it, and what you expected to see. “I can't connect” or “the build failed” alone is usually not enough to begin troubleshooting.
Client version, macOS and Xcode versions, Runner labels, failed task ID, network type, and the last successful operation before the issue occurred.
Passwords, private keys, access tokens, original signing materials, payment credentials, complete environment-variable files, and unprocessed project source code.
The order total consists of the selected model, rental period, node region, and add-ons. Submit issues with existing orders through the portal so the ticket can be linked automatically to the account and node record.
Only USDT-TRC20 and Visa / Mastercard / Amex (via Stripe) are supported. All orders are settled in US dollars (USD); available gateways are determined by the portal.
The total consists of the base model, the daily, weekly, monthly, or quarterly rental period, the node region, and selected storage or Thunderbolt 5 daisy-chaining add-ons. Review each billing line before submitting.
A portal ticket can be linked to the order ID, node region, and delivery record, reducing back-and-forth checks. Include the occurrence time, reproduction steps, redacted logs, and expected result when submitting.
Review team permissions promptly, update the relevant credentials, and revoke access that is no longer needed. Never send unredacted credentials in email, chat messages, or tickets.
Submit order, connection, billing, or build issues through the portal first. For configuration reviews, bulk nodes, and business collaboration needs, contact the service team for further confirmation.