Sen4CAP 3.0 - Error message L2-S1 Pre-Processing


During pre-processing a get the following error. In one of the previous posts a reference was made to the available memory as a possible cause for this error (see message below from the l1_tile_history). I am using a VM with 128 GB of memory and plenty of storage.

Flavor Name hm.2xlarge
Flavor ID 26
Disk 384GB (extended to & TB)

How to solve this particular issue?

Best regards,

2022-04-20 12:05:13[SEVERE] Error executing step 2 (Coregistration) [codes=[137, 137, 137]];;2022-04-20 12:05:15[WARNING] Callback for product SEN4CAP_L2A_S1_V20190308T171559_20190314T171653_VH_015_COHE failed. Reason: java.lang.NullPointerException
at sun.nio.fs.UnixPath.normalizeAndCheck(
at sun.nio.fs.UnixPath.(
at sun.nio.fs.UnixFileSystem.getPath(
at java.nio.file.Paths.get(
at org.esa.sen4cap.preprocessing.Sentinel1Level2Worker.createProducts(
at org.esa.sen4cap.preprocessing.Sentinel1L2ProcessorV1.processProductPair(
at org.esa.sen4cap.preprocessing.Sentinel1L2ProcessorV1.processAcquisition(
at org.esa.sen4cap.scheduling.Sentinel1Level2Job.lambda$processSite$0(
at java.util.concurrent.Executors$
at java.util.concurrent.ThreadPoolExecutor.runWorker(
at java.util.concurrent.ThreadPoolExecutor$

Hello Marcel,

The exit codes from the Coregistration step (137) indicate that Docker did not have enough memory (maybe something else was executing at the same time? Multiple MAJA instances for example?). Do you constantly get this error?
Regardless, one thing you can do is to set the parameter “processor.l2s1.parallel.steps.enabled” to ‘false’. This will make the substeps to sequentially execute (as opposed in parallel by default), hence less memory needed.

Hope this helps,

Thank you Cosmin! Unfortunately it didn’t solve the problem. I’ve had similar issues with both version 2.0 and version 3.0. So the search continues.