Ray dashboard failed to start when running train_solver_judge_flow example

[maksimstw]

Description

When I tried to run the train_solver_judge_flow example, the Ray dashboard failed to start and produced the following errors. The Ray instance still started, but the dashboard component crashed.

Error Logs

2025-10-26 19:45:22,084 ERROR services.py:1360 -- Failed to start the dashboard 
2025-10-26 19:45:22,084 ERROR services.py:1385 -- Error should be written to 'dashboard.log' or 'dashboard.err'. We are printing the last 20 lines for you. See 'https://docs.ray.io/en/master/ray-observability/user-guides/configure-logging.html#logging-directory-structure' to find where the log file is.
2025-10-26 19:45:22,085 ERROR services.py:1429 -- 
The last 20 lines of /tmp/ray/session_2025-10-26_19-44-57_751930_117461/logs/dashboard.log:
2025-10-26 19:45:05,945 INFO utils.py:338 -- Available modules: [<class 'ray.dashboard.modules.usage_stats.usage_stats_head.UsageStatsHead'>]
2025-10-26 19:45:05,945 INFO head.py:235 -- DashboardHeadModules to load: None.
2025-10-26 19:45:05,945 INFO head.py:238 -- Loading DashboardHeadModule: <class 'ray.dashboard.modules.usage_stats.usage_stats_head.UsageStatsHead'>.
2025-10-26 19:45:05,945 INFO head.py:242 -- Loaded 1 dashboard head modules: [<ray.dashboard.modules.usage_stats.usage_stats_head.UsageStatsHead object at 0x150676d3a020>].
2025-10-26 19:45:05,949 INFO head.py:296 -- Loaded 9 subprocess modules: [...]
2025-10-26 19:45:22,230 WARNING services.py:2155 -- WARNING: The object store is using /tmp instead of /dev/shm because /dev/shm has only 2147483648 bytes available. This will harm performance!
2025-10-26 19:48:08,709 E 117461 117461] core_worker_process.cc:223: Failed to register worker to Raylet: IOError: Failed to read data from the socket: End of file worker_id=01000000ffffffffffffffffffffffffffffffffffffffffffffffff

Steps to Reproduce

1. Run the solver_judge training example:

   bash train_solver_judge_flow.sh

2. Observe that the Ray dashboard fails to start with the above error.

[maksimstw]

The problem seems to be related to an I/O bottleneck on my machine.