Compression of already existing images not working

[Felipe Salazar]

Hello,

I'm trying to compress already stored images. I executed the following steps:

1) https://github.com/dcm4che/dcm4chee-arc-light/wiki/Compression-of-received-images

2) https://github.com/dcm4che/dcm4chee-arc-light/wiki/Compression-of-Images

The only entry that is shown in the log related to this is:

2023-03-21 16:32:18,940 INFO  [org.dcm4chee.arc.compress.impl.CompressionScheduler] (EE-ManagedScheduledExecutorService-default-Thread-10) start CompressionScheduler.execute()
2023-03-21 16:32:18,944 INFO  [org.dcm4chee.arc.compress.impl.CompressionScheduler] (EE-ManagedScheduledExecutorService-default-Thread-10) Found 0 Series scheduled for compression
2023-03-21 16:32:18,944 INFO  [org.dcm4chee.arc.compress.impl.CompressionScheduler] (EE-ManagedScheduledExecutorService-default-Thread-10) finished CompressionScheduler.execute()

The only related entry about my issue I've found is (hope is not my case because I have more than 48Tb stored here uncompressed):

https://github.com/dcm4che/dcm4chee-arc-light/issues/3957

Maybe is something I'm not considering in the configuration?

Thank you for your valuable time.

[Felipe Salazar]

Update,

Compression works when receiving images, but still not working with already stored images. At the moment, created another AET and started to transfer between internal AETs to achieve the goal using the API, and partitioning manually the number of studies/series, but would be great to identify how to use the native functions.

[pascu....@gmail.com]

Hello,

I have the same problem, it compresses new studies but not old ones that were archived before I activated the compression "Found 0 Series scheduled for compression" . Do you have any idea what it could be, Version 5.26 ?

Thanks.

[Jonathan Brooks]

I'm interested in getting this running on an archive that already has a lot of data on it, so would like to know a bit more about the problems.

One thing occurs to me - will the compression schedule hour=21-6 actually be understood by the code? Seems a bit unlikely. Could try 21-24|0-6?

Also I would imagine that the dayOfWeek format might not recognise a wild card, but instead require numbers separate by pipes (|) i.e. 0|1|2 etc?

Last thing - these probably have to be entered on separate lines i.e. press +Add to set another parameter.

Apologies if these options are exactly the right ones! I was looking for a reference and couldn't find one for the UI settings.

Best wishes,

Jon

[Felipe Salazar]

It's a bug, the only way I found to get it working is using another temporary archive to compress and return the compressed studies.