We are having an issue with L2 S1 preprocessing on SEN4CAP version 3.2.
For you information, we are using a local repository for raw images storage.
Currently all S1 preprocessing are systematically failing during Coregistration step with code [137,137,2147483647]
Please see the extract of the sen2agri log below.
Is it possible that this error is given due to an OutOfMemory exception. How much RAM do you have on the machine? Maybe you should consider adding some swap file if you cannot add more RAM to the machine.
We did follow your recommendations and increased the parameter processor.l2s1.min.memory from 8192 to 16384 and restarted sen2agri service afterwards.
Unfortunately, the L2A S1 preprocessing is still ending with failure [137,137,2147483647]. In the snapshot below, it fails during coregistration but we have observed other jobs failing with the same error code at another step.
When analysing the sen2agri logs, no other ERROR are shown. Nevertheless we noticed the [SEVERE] entry below appearing periodically in the log with " more or less " the same frequency of the L2S1 error. Could boths error be somehow connected?
Normally, the code 137 is given when an out of memory is given.
Could you try increasing the parameter processor.l2s1.min.memory from 8192 to 16384?
In the user interface this can be found in Configurations->S1 L2 Pre-processor → “Minimum free memory for a step”.
A sen2agri-services restart I think is recommended after changing this parameter.
We did follow your recommendations and increased the parameter processor.l2s1.min.memory from 8192 to 16384 and restarted sen2agri service afterwards.
Unfortunately, the L2A S1 preprocessing is still ending with failure [137,137,2147483647]. In the snapshot below, it fails during coregistration but we have observed other jobs failing with the same error code at another step.
When analysing the sen2agri logs, no other ERROR are shown. Nevertheless we noticed the [SEVERE] entry below appearing periodically in the log with " more or less " the same frequency of the L2S1 error. Could boths error be somehow connected?