a month ago
Hello,
I'm using the N8N (w/ workers) template (
https://railway.com/template/r2SNX_) and managed to install two community nodes (n8n-nodes-globals & n8n-nodes-mcp) without issue.
But when I redeploy the primary service (n8n), the following error message appears next to both of the installed community nodes: "There is a problem with this package, try uninstalling it then reinstalling to resolve this issue".
Can you please advise what the issue could be?
Below is the deploy log (IDs have been removed):
Starting Container
Initializing n8n process
n8n ready on ::, port 5678
There are deprecations related to your environment variables. Please take the recommended actions to update your configuration:
- N8N_RUNNERS_ENABLED -> Running n8n without task runners is deprecated. Task runners will be turned on by default in a future version. Please set N8N_RUNNERS_ENABLED=true to enable task runners now and avoid potential issues in the future. Learn more: https://docs.n8n.io/hosting/configuration/task-runners/
- OFFLOAD_MANUAL_EXECUTIONS_TO_WORKERS -> Running manual executions in the main instance in scaling mode is deprecated. Manual executions will be routed to workers in a future version. Please set OFFLOAD_MANUAL_EXECUTIONS_TO_WORKERS=true to offload manual executions to workers and avoid potential issues in the future. Consider increasing memory available to workers and reducing memory available to main.
n8n detected that some packages are missing. For more information, visit https://docs.n8n.io/integrations/community-nodes/troubleshooting/
Version: 1.83.2
================================
Start Active Workflows:
================================
Unrecognized node type: n8n-nodes-mcp.mcpClient
Error: Unrecognized node type: n8n-nodes-mcp.mcpClient
at LoadNodesAndCredentials.getNode (/usr/local/lib/node_modules/n8n/dist/load-nodes-and-credentials.js:285:19)
at NodeTypes.getByNameAndVersion (/usr/local/lib/node_modules/n8n/dist/node-types.js:44:51)
at new Workflow (/usr/local/lib/node_modules/n8n/node_modules/n8n-workflow/dist/Workflow.js:56:39)
at ActiveWorkflowManager.add (/usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:370:24)
at ActiveWorkflowManager.activateWorkflow (/usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:300:45)
at processTicksAndRejections (node:internal/process/task_queues:95:5)
at /usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:289:17
at async Promise.all (index 0)
at ActiveWorkflowManager.addActiveWorkflows (/usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:291:13)
at ActiveWorkflowManager.init (/usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:92:9)
=> ERROR: Workflow "MCP Test" (ID: XXXXXXXXXXX) could not be activated on first try, keep on trying if not an auth issue
Unrecognized node type: n8n-nodes-mcp.mcpClient
Issue on initial workflow activation try of "MCP Test" (ID: OWsNaVPTIKW9tSYd) (startup)
- "Trello - Blog Creation" (ID: XXXXXXXXXXX))
=> Started
Try to activate workflow "MCP Test" (XXXXXXXXXXX)
Unrecognized node type: n8n-nodes-mcp.mcpClient
Error: Unrecognized node type: n8n-nodes-mcp.mcpClient
at LoadNodesAndCredentials.getNode (/usr/local/lib/node_modules/n8n/dist/load-nodes-and-credentials.js:285:19)
at NodeTypes.getByNameAndVersion (/usr/local/lib/node_modules/n8n/dist/node-types.js:44:51)
at new Workflow (/usr/local/lib/node_modules/n8n/node_modules/n8n-workflow/dist/Workflow.js:56:39)
at ActiveWorkflowManager.add (/usr/local/lib/node_modules/n8n/dist/active-workflow-manager.js:370:24)
at Timeout.retryFunction [as onTimeout] (/usr/local/lib/nodemodules/n8n/dist/active-workflow-manager.js:446:28)
at listOnTimeout (node:internal/timers:581:17)
at processTimers (node:internal/timers:519:7)
-> Activation of workflow "MCP Test" (XXXXXXXXXXX) did fail with error: "Unrecognized node type: n8n-nodes-mcp.mcpClient" | retry in 2 seconds
- "Backup your workflows to GitHub" (ID: XXXXXXXXXXX))
=> Started
Editor is now accessible via:
https://XXXXXXXXXXX.up.railway.app
Try to activate workflow "MCP Test" (XXXXXXXXXXX)
Unrecognized node type: n8n-nodes-mcp.mcpClient
9 Replies
a month ago
Attached are the screenshots of before and after reploy.
Attachments
a month ago
Hey!
This seems like an issue with your project/application. Unfortunately, we're unable to offer first-party support for issues unrelated to the Railway product or platform.
Other communities such as StackOverflow might be able to help you out further.
Status changed to Awaiting User Response railway[bot] • about 1 month ago
a month ago
Sure thing, totally understand.
I believe I need to persist the contents of the ~/.n8n/nodes directory (https://docs.n8n.io/integrations/community-nodes/troubleshooting/)
Can you please advise how I can persist a directory within Railway's configuration options?
Status changed to Awaiting Railway Response railway[bot] • about 1 month ago
a month ago
Hey there, unsure if this helps but you can configuring persistent storage for the ~/.n8n/nodes directory to ensure that the community nodes remain installed across deployments which might be possible with persistent volumes in Railway https://docs.railway.com/reference/volumes
Status changed to Awaiting User Response railway[bot] • 29 days ago
a month ago
Having the exact same issue.
Can you please tell exactly whre to find ~/.n8n/nodes directory so that we can configure persistent storage. A bit more context would be helpful.
Status changed to Awaiting Railway Response railway[bot] • 29 days ago
a month ago
Hey, unable to offer n8n specific advice here but I think if you create a volume and mount it to the ~/.n8n/nodes directory the service that this is referring to, that might do the trick. For more details on setting up a volume and mounting it to your service, you can refer to Railway's documentation on using volumes.
Status changed to Awaiting User Response railway[bot] • 29 days ago
a month ago
hi, did this work for you? what's the mount path b/c ~/.n8n/nodes just shows red
Status changed to Awaiting Railway Response railway[bot] • 26 days ago
bindingox
hi, did this work for you? what's the mount path b/c ~/.n8n/nodes just shows red
a month ago
No, I still can't get it working
flow-ai
No, I still can't get it working
a month ago
This works perfectly on local install but not cloud installs like railway so i'll stick to local