[docker/cli] Docker CLI crashes when using with --no-trunc

### ROOT CAUSE The crash occurs because the system's `ulimit` for the maximum number of user processes (threads) is too low. The Docker CLI, when processing the `--no-trunc` flag, attempts to create additional threads, but the system limit prevents this, causing the runtime error. ### CODE FIX 1. **Increase System `ulimit`:** Temporarily increase the system's `ulimit` for the number of user processes. This is a system-level change and should be done cautiously. ```bash # Increase the limit to 4096 (adjust as needed) ulimit -u 4096 ``` Then retry the command. Note: This change is temporary and affects the current shell session. 2. **Modify Docker Service Configuration:** If this is a recurring issue, adjust the Docker service configuration to allow a higher `ulimit`. For example, on systems using `systemd`, edit the Docker service file (`/etc/docker/daemon.json`) to include: ```json { "default-ulimit": { "nproc": { "soft": 4096, "hard": 4096 } } } ``` Then restart Docker: ```bash sudo systemctl restart docker ``` 3. **Submit a Fix to Docker:** If the issue persists, the Docker team should investigate the code in the `service ps` command (specifically how `--no-trunc` is handled) to ensure it doesn't unnecessarily spawn threads. This requires patching the Docker CLI codebase. **Note:** Increasing `ulimit` is a workaround. The long-term solution requires the Docker team to optimize thread usage in the CLI.