Hello dear Community,
I crashed Knime while I was generating an ORC archive of some data temporarily available on a db server. Data was extracted locally using a DB Query Reader Node (Labs) before being transformed in an ORC file.
Unfortunately, le reader node was in a loop, so that I got the following warning message when I reloaded the workflow:
WARN ORC Writer 0:1477 Can't continue loop as the workflow was restored with the loop being partially executed. Reset loop start and execute entire loop again.
Unfortunately, this was before the weekend, and now the data is not available on the DB server anymore.
Fortunately, I found in the knime-workspace that there was still a copy of the data at the output of the DB Query Reader Node. I found the following files:
- data.xml (file telling the data is in data.zip)
- data.zip (2Gb, this must be my data)
- spec.xml (describing the whole data structure)
Inside the zip, I found
Is there a simple way to reimport the data inside data.bin in another worflow to avoid losing it?
Using a specific node? A python snippet? Or anything else?
Thanks in advance for your help!
I have never done that but you could create a new workflow create a dummy table and use an appropriate node and save that in an executed state. Maybe the same kind of node that crashed.
Then close KNIME find the corresponding data.zip file and switch it with your file (always keep a copy). Then open KNIME and see if you could access the data. Could be worth a try.
Other than that people more familiar with the internal structure of knime files would have to weight in.
Thanks, @mlauber71, for your proposal. It just works! Thanks as well to @samuelVermeulen who had advised me the same solution.
I closed Knime during the files swap and used the same kind of node as the one I have taken the content from (I took all the files in the node directory, not just the files in the port_1 folder).
Nevertheless, it could be interesting to get from some Knimers some useful information about the way information is organized in the node folders of the workspace and eventually to have a more elegant way to re-read the content of these files.
Knime is a great software with a great community!
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.