Terminal stays blank; several errors in logs

I’m running theia-full:latest and after not having used it for a while (the container was kept up to date by watchdog) I tried using it again today and noticed that when I try to open a terminal, the bottompane stays empty (with just a cursor).
The logs show errors like this:
terminal ERROR Error: ENOENT
at EventEmitter. (/home/theia/node_modules/@theia/process/lib/node/terminal-process.js:134:33)
at EventEmitter.emit (events.js:189:13)
at UnixTerminal.Terminal.emit (/home/theia/node_modules/@theia/node-pty/lib/terminal.js:77:42)
at /home/theia/node_modules/@theia/node-pty/lib/unixTerminal.js:98:23
at process._tickCallback (internal/process/next_tick.js:61:11)
terminal ERROR Terminal pid: 212 error: Error: ENOENT, closing it.
process WARN This process was not registered or was already unregistered. [ID: 4]
root ERROR Couldn’t resize terminal -1, because it doesn’t exist.
root ERROR Couldn’t resize terminal 4, because it doesn’t exist.

[original thread by Deleted]

@marc-dumais @paul-marechal Is there some timeout for terminal processes? maybe on the system level. Maybe OS closed these processes for some reasons?


Might be relevant info, but previously (sadly I don’t know when this stopped working) this worked on the same system. The relevant hosts are Ubuntu Server 18.04

ENOENT means that whatever was being spawned wasn’t found. If you can access the container filesystem before running it, you should try to add a console log in node_modules/@theia/process/lib/node/terminal-process.js in order to see the spawn options and check that the executable can be found in your environment. This is a weird issue. Trying to see if I get the same issue on my side.

Ok, of course we get ENOENT, it seems the terminals try to do spawn(’/bin/bash THEIA_DEFAULT_PLUGINS=local-dir:/home/theia/plugins’, []), this is not a valid call. I don’t know why yet.

Ok, was a typo in the dockerfile, will be fixed soon.


I love the support you guys give. Any idea what the ETA is for that fix, or should I try fixing it myself in my running container for the meantime?


Oh, I see you fixed it 16 hours ago, and that I merely had to pull the new image. Thanks so much!!!