after a migration to new linux machine starting a workflow in batch mode always results in the error message "ERROR main BatchExecutor Workflow is locked by another KNIME instance".
But that's not the case. Therefore after half a day of frustrated trial and error fixing attempts my question:
How does KNIME "lock" a workflow and how can this be removed manually?
As the error message suggests I suppose the BatchExecutor (https://tech.knime.org/docs/api/index.html?org/knime/core/node/workflow/BatchExecutor.html) throws the exception but unfortunately I haven't been able to figure out how it checks for a "Lock" or how to remove that.
The workflow directory contains a file .knimeLock if it is in use and will be deleted if the workflow is closed. If you are sure that it's not in use any more and the file is still there, you can delete it.
Hi Thor,
thank you very much for your fast reply.
In general you are absolutely right with what you said and a test workflow I just created runs perfectly fine creating the .knimeLock file during execution and deleting it afterwards.
BUT unfortunately the original workspace resulting always in the above error message does NOT contain a .knimeLock file.
Nevertheless it throws this unfortunate error message. Is there anything else that triggers this message and why does the same workflow finishes perfectly in the GUI but not in batch mode on the command line? (Of course GUI is closed at this time)
Is your workspace residing on a network drive? Some network file system don't support file locking very well. Also please check if you have write permissions to the workflow directory.
Hi Thor,
thanks for your reply.
In my desperation I formatted the whole machine and installed everything new from the ground up.
At least I have a different error message now:
INFO main BatchExecutor ===== Executing workflow /home/knimeuser/knime-workspace/knime-export.zip =====
ERROR main FileWorkflowPersistor Unable to load node with ID suffix 5 into workflow, skipping it: InitialValue: value (=0.0) must be within the range [1…2].
ERROR main BatchExecutor Status: Error: Workflow_neu2 0 loaded with errors
Status: Error: Workflow_neu2 0
Status: Error: Document Creation 0:287
Status: Error: Unable to load node with ID suffix 5 into workflow, skipping it: InitialValue: value (=0.0) must be within the range [1…2].
ERROR main BatchExecutor IO error while loading the workflow
INFO main BatchExecutor ========= Workflow did not execute sucessfully ============
I really don’t know what could be the reason for that error. On a different linux machine with GUI the same process runs perfectly fine (even from the command line).
How can I fix or at least find more information about this error?
The process seems to fail immediately after reading the csv-file but I don’t know why. The same file with the same process works on a different machine…
@thor
I have a similar problem with version 4.4.1
Switched to a workspace in a filedrive (before had workspace on my local machine, which is not backed up with our company procedures, so wanted to make it safer by working in a location that IS backed up every day), and could NOT open workflows anymore due to the “workflow is locked by another user” error message.
When trying to hard delete the knimeLock file from the workflow directory (KNIME was NOT running at this point) and the restarting KNIME, the knimeLock file was generated again and the workflow could not be opened, giving the same error message.
I suggest this is a bug, if it is active after 5 years still?