out of memory error?

    Hi,


    After installing the new version of conquest (version 1413) I decided to regenerate the index file. I'm using MySql ver 5.0.22 in the windows XP pro system. It has I G memory. MySql ini file is pattern after the suggestion in the manual. Our data is 1 terabyte approx 2mil images


    First did a dgate -v -r and got this error


    CTPACS] ***VR:ReAlloc out of memory allocating 1701273966 bytes
    [CTPACS] ***A fatal error occurred (out of memory) - closing server


    Tried to do a maintainance re-index of MAG0 using the GUI interface and still got the same error.


    Has anyone encountered the same problem?


    ajg




    I'm running the a regeneration of index with Conquest version 1412c. Will update if the same error occurs.

    The echo problem: could you switch on debug logging and show the request?


    The regenetion error may be caused by a malformed dicom file that silently was ignored before. Can you get me the file where the problem occurs?


    Marcel

    here is the log from Conquest after an echo
    PACS] set debug level from GUI
    [PACS] set debug log from GUI
    [PACS] ***Bufferspace: out of memory allocating 1061109499 bytes
    [PACS] ***A fatal error occurred (out of memory) - closing server
    *** Restarted dead server after error 7
    [PACS] DGATE (1.4.13, build Fri Nov 16 23:13:02 2007) is running as threaded server
    [PACS] Creating index on field 19 for H:\conquest\Data\dbase\DICOMImages.DBF with 476 records
    [PACS] Creating index on field 0 for H:\conquest\Data\dbase\DICOMPatients.DBF with 4 records
    [PACS] Creating index on field 13 for H:\conquest\Data\dbase\DICOMSeries.DBF with 52 records
    [PACS] Creating index on field 15 for H:\conquest\Data\dbase\DICOMStudies.DBF with 4 records
    [PACS] Creating index on field 0 for H:\conquest\Data\dbase\DICOMWorkList.DBF with 0 records
    [PACS] Index for H:\conquest\Data\dbase\DICOMPatients.DBF finished in 50 ms; total space = 10004 records
    [PACS] Index for H:\conquest\Data\dbase\DICOMSeries.DBF finished in 50 ms; total space = 48559 records
    [PACS] Index for H:\conquest\Data\dbase\DICOMStudies.DBF finished in 50 ms; total space = 10004 records
    [PACS] Index for H:\conquest\Data\dbase\DICOMWorkList.DBF finished in 50 ms; total space = 10000 records
    [PACS] Index for H:\conquest\Data\dbase\DICOMImages.DBF finished in 60 ms; total space = 444505 records
    [PACS] User interface test: local server is running!
    [PACS] set normal log from GUI

    Hi,


    You are correct. The cause of the fault is a malformed dicom image file. It triggers an out of memory fault with version 1413, and yes it is ignored in version 1412c. Removing the file from the storage data solved the problem. I will send the file to through your mail.


    Thanks for the tip.


    ajg

    Hi,
    [CONQUESTSRV1] UPACS THREAD 0: STARTED AT: Sun Dec 30 20:26:08 2007
    [CONQUESTSRV1] Calling Application Title : "Kpserver"
    [CONQUESTSRV1] Called Application Title : "CONQUESTSRV1"
    [CONQUESTSRV1] Application Context : "1.2.840.10008.3.1.1.1"
    [CONQUESTSRV1] Presentation Context 0 "1.2.840.10008.5.1.4.1.1.2"
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879224.7690.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879224.7691.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879224.7692.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879224.7693.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879226.7854.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879226.7855.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879226.7856.dcm
    [CONQUESTSRV1] [recompress]: recompressed with mode = un (strip=0)
    [CONQUESTSRV1] Written file: C:\KPacs\Imagebox\1.2.840.113704.1.111.2804.1167230217.9\1.2.840.113704.1.111.8136.1167230539.3\2.16.840.1.113786.1.5438.800.438879226.7857.dcm
    [CONQUESTSRV1] UPACS THREAD 0: ENDED AT: Sun Dec 30 20:26:10 2007
    [CONQUESTSRV1] UPACS THREAD 0: TOTAL RUNNING TIME: 2 SECONDS
    [CONQUESTSRV1]
    [CONQUESTSRV1] UPACS THREAD 1: STARTED AT: Sun Dec 30 20:26:11 2007
    [CONQUESTSRV1] Calling Application Title : "Kpserver"
    [CONQUESTSRV1] Called Application Title : "CONQUESTSRV1"
    [CONQUESTSRV1] Application Context : "1.2.840.10008.3.1.1.1"
    [CONQUESTSRV1] Presentation Context 0 "1.2.840.10008.5.1.4.1.1.2"
    [CONQUESTSRV1] ***VR:ReAlloc out of memory allocating 1081417728 bytes
    [CONQUESTSRV1] ***A fatal error occurred (out of memory) - closing server



    I tried to send the images from Kpacs and I got the above errors.
    Using Conquest, I regenerate Imagebox of Kpacs and I didn't have any problem.
    I sent the images from DICOMREADER 3.1 and I didn't have any problems.
    Is it a problem of transfer syntax in KPACS?
    I killed and restarted the server four times while the sending is still in progress in Kpacs, and Conquest started receiving images. although, I was missing the axial series.


    Any ideas?

    Hi,


    I scan the imagebox of kpacs (JPEG2000lossless) with DICOMReader 3.2b and sent it to Conquest with no problem.
    I tried to send the images from kpacs (jpeg2000lossless) to conquest with no success.
    So, that means that there is a problem when transferring the images (jpeg2000lossless) to conquest.

    Hi,


    Jpeg2000 is indeed not supported. The out of memory error may be related to a transfer syntax problem (it should not accept anything it doen not understand - try disabling LittleEndianExlicit, see top post).


    Marcel

    I think Conquest as you said does not support JPEG2000Lossless.
    I think DicomReader 3.2 receives JPEG2000lossless and uncompress it before it sends the images to Conquest and that is why it is successful.
    On the other hand, K-PACS does not decompress the images, but it send them as they are JPEG2000 lossless.
    So, are you thinking of adding this feature to Conquest that it will support JPEG2000 Lossless? or will it cause some problems?


    thanks

    I ran into this problem with images from an ATL HDI5000 which would cause the memory error and the server to shut down. Reverting back to 1412c seems to have solved the problem. I'll know in the next few days. This is an old ultrasound system so its adherence to DICOM 3.0 is probably questionable.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!

Register Yourself Login