node-red-nodes
node-red-nodes copied to clipboard
node-red
node-red-node-xmpp crash
Which node are you reporting an issue on?
node-red-node-xmpp
What are the steps to reproduce?
- configure everything
- connect to a MUC room with xmpp in and xmpp out
- shut down XMPP server
What happens?
- node-red service crashes entirely, or
- node-red service becomes unresponsive
What do you expect to happen?
- have the XMPP clients try to reconnect every 10 seconds or so, and...
- ...most importantly: dont let node-red crash!
Please tell us about your environment:
- [ ] Node-RED version: latest (2.1.5)
- [ ] Node.js version: 16.13.2
- [ ] Platform/OS: Ubuntu Server 20.04
- [ ] Browser: Firefox
- [ ] XMPP plugin: latest
Is there anything in the Node-RED log (the console output not the debug sidebar) when it crashes?
I've got this in syslog:
17:10:47 systemd[1]: Started LSB: Start/stop openfire jabber server. NOTE: here I stop OpenFire
Then the thing goes ballistic:
17:10:48 node-red[190463]: 13 Jan 17:10:48 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:48 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:48 node-red[190463]: 13 Jan 17:10:48 - [info] [xmpp in:NAS messages XMPP] XMPP Error: Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:48 node-red[190463]: 13 Jan 17:10:48 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:48 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:48 node-red[190463]: 13 Jan 17:10:48 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:48 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:49 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:49 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [info] [xmpp in:NAS messages XMPP] XMPP Error: Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [info] [xmpp out:XMPP transmit] Error: connect ECONNREFUSED HIDDEN IP:5222
17:10:49 node-red[190463]: at TCPConnectWrap.afterConnect [as oncomplete] (node:net:1161:16)
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [info] [xmpp-server:52fa80979f65b5c4] b5165f1cd028a58e sent a message to the xmpp server
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [red] Uncaught Exception:
17:10:49 node-red[190463]: 13 Jan 17:10:49 - [error] TypeError: Cannot read properties of null (reading 'write')
17:10:49 node-red[190463]: at /home/node-red/.node-red/node_modules/node-red-node-xmpp/node_modules/@xmpp/connection/index.js:333:19
17:10:49 node-red[190463]: at new Promise (
This is not always reproducible, sometimes it keeps spitting out the ECONNREFUSED at an incredible rate, which freezes Node-RED almost to a halt.
One thing I notice regarding reconnection, is that the default reconnect period is 1 second, while the connection timeout of the node is 60 seconds. This might cause at least part of the issues.
The reconnect timeout can be set in packages/reconnect (see https://github.com/xmppjs/xmpp.js/tree/main/packages/reconnect). The connection timeout is hardcoded in the node source code (while I don't know whether this value is actually supported, since I can't find it in the declarations of xmmpjs). I couldn't get a higher reconnect timeout to work by myself yet, though, it remains hammering the reconnect attempts.
xmpp node is crashing node-red I got this on the logs:
"Unknown error: StreamError: policy-violation - Too many unacked stanzas"
21 Aug 11:47:56 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:47:56 - [error] [xmpp out:39d00c59f7cace81] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:47:56 - [error] [xmpp out:39d00c59f7cace81] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:47:56 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:47:57 - [red] Uncaught Exception: 21 Aug 11:47:57 - [error] Error: write ECONNRESET at WriteWrap.onWriteComplete [as oncomplete] (node:internal/stream_base_commons:98:16)
After this the system crashes (running on docker).
I believe this started to happen right after a debug node was connected to the second output of a xmpp-in node.
This is a second crash error, just in case it brings more light:
21 Aug 11:57:03 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:57:03 - [error] [xmpp out:39d00c59f7cace81] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:57:03 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 21 Aug 11:57:04 - [red] Uncaught Exception: 21 Aug 11:57:04 - [error] Error [ERR_STREAM_DESTROYED]: Cannot call write after a stream was destroyed at new NodeError (node:internal/errors:371:5) at errorBuffer (node:internal/streams/writable:517:14) at afterWrite (node:internal/streams/writable:501:5) at onwrite (node:internal/streams/writable:477:7) at WriteWrap.onWriteComplete [as oncomplete] (node:internal/stream_base_commons:91:12)
Do you know how your server is configured as that error is due to a limit being exceeded on the server - eg see https://github.com/processone/ejabberd/issues/2615
No details on the xmpp server side, as it's a public server (at jabber.org). I saw the thread you added before posting the node-red crash here and guessed that's the root cause, but I raised the issue cause node-red will enter in a crash loop if the xmpp node is installed and is set to get connected to a server with this config. The weird thing is this happened after 24 hours, when I did some changes on the config of the node; so I guessed the stanza missconfig on the server is affecting the node-red xmpp client only sometimes.
can you paste the complete error stack ? and which version of the xmpp node are you using ? Thanks
xmpp node version 0.6.0
unfortunatelly I cannot get more logs than the ones I posted before, everytime I re-enable the xmpp node enabled, node-red crashes.
These are the lines I got on the last test some minutes ago, logs start when I click on "Deploy" after enabling the nodes:
24 Aug 14:38:42 - [info] Starting modified flows 24 Aug 14:38:42 - [info] Started modified flows 24 Aug 14:38:44 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 24 Aug 14:38:44 - [error] [xmpp out:39d00c59f7cace81] Unknown error: StreamError: policy-violation - Too many unacked stanzas 24 Aug 14:38:44 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 24 Aug 14:38:44 - [error] [xmpp in:7fef6170bdaf5d29] Unknown error: StreamError: policy-violation - Too many unacked stanzas 24 Aug 14:38:45 - [red] Uncaught Exception: 24 Aug 14:38:45 - [error] Error [ERR_STREAM_DESTROYED]: Cannot call write after a stream was destroyed at new NodeError (node:internal/errors:371:5) at errorBuffer (node:internal/streams/writable:517:14) at afterWrite (node:internal/streams/writable:501:5) at onwrite (node:internal/streams/writable:477:7) at WriteWrap.onWriteComplete [as oncomplete] (node:internal/stream_base_commons:91:12)
after this, node-red enters into an endless crash loop (on docker).
OK - hmm - we have event handlers in place for the xmpp object .on("error"... so not sure why it isn't catching them unless the underlying library isn't calling it correctly.