DB Error

Help
TJ_DD
2010-10-31
2013-05-02
  • TJ_DD

    TJ_DD - 2010-10-31

    Hi,

    I'm working at my final thesis @ Dresden University of Technology and I want to use the XBRLAPI to process some XBRL-Files.
    I tried to fix the problem, but it seems I'm not able to fix it.

    If I run the LoadAllSECFilings.java I get the following message:

    "PANIC: fatal region error detected; run recovery
    XBRLAPI Exception: The BDB XML database environment could not be set up.
    Caused by: DB_RUNRECOVERY: Fatal error, run database recovery: PANIC: fatal region error detected; run recovery"

    Are there any ideas how I could fix the problem?

    Systemspecs: Win7 x64 and Java 6 U 22

    Thx,

    TJ

     
  • Geoffrey Shuetrim

    TJ

    I am not able to give you many suggestions based on the details provided above.  It does appear that the database is not set up correctly but that is about the extent to which I can diagnose the issues.  For what it is worth, when the bdbxml data store is constructed, it checks to see if there is an existing environment set up at the database location (the folder that will contain the database files).  If not, the environment is initialised.  If you look at the environment initialisation code in the bdbxml store implementation, that should tell you more about which particular part of the environment initialisation caused the PANIC.  We can then probably be more constructive about resolving the problem.

    Regards

    Geoff S

     
  • TJ_DD

    TJ_DD - 2010-11-01

    Hello,

    thx for ur answers.

    The problem is fixed. I dont know why, but it's gone. I reinstalled the BerkeleyDB into another directory, maybe there the DB can write (it was located in my user directory befor).

    But there's another problem now and maybe u how to fix it.

    "MapViewOfFile: Für diesen Befehl ist nicht genügend Speicher verfügbar.

    PANIC: Not enough space
    PANIC: DB_RUNRECOVERY: Fatal error, run database recovery
    unable to join the environment
    XBRLAPI Exception: The BDB XML database environment could not be set up.
    Caused by: DB_RUNRECOVERY: Fatal error, run database recovery: PANIC: DB_RUNRECOVERY: Fatal error, run database recovery
    unable to join the environment"

    The programm craches at this point at the StoreImpl.java: environment = new Environment(new File(locationName), environmentConfiguration); (line 187)

    I tried to run the loadAllSECFiling.java with the following arguments:

    -database C:\BerkeleyDB\container -container newContainer -cache C:\new

    Maybe I don't get the meaning of the value I have to set. I think the database argument is to set the installation directory of the DB, the container argument ist just a name and the cache argument is for the XBRL-files. Am I right?

    I think the DB is set up correctly now.

    Regards from Dresden

    Tobias

     
  • Geoffrey Shuetrim

    I have had other Windows users find that spaces in the path to the Oracle database can cause problems.  Maybe that is related to the fixing of the problem…

    Regarding the new problem, first up, make sure that you delete all database generated files from the directory that you are using as the database location.  If you have one set of environment files and try to open up a database container using a different environment configuration, you can run into problems.

    Next up, the message you are receiving actually looks like there is a problem with either disk space (unlikely) or memory available.  As far as I am aware, the database environment defines an amount of memory to be used for caching.  That is in addition to the Java heap space that you allocate when starting the JRE.  It may be that between those two, you are running yourself out of physical memory.

    More generally, these kinds of problems can often be dealt with better by those with Oracle Berkeley XML database expertise.  There is an excellent forum on the Oracle site that may well have information on the problems you are encountering.  Try using their search facility to hunt for fragments of the error messages you are encountering.

    The Forum URL is http://forums.oracle.com/forums/forum.jspa?forumID=274

     
  • TJ_DD

    TJ_DD - 2010-11-10

    Hi,

    I guess I solved the problem. I reinstalled the Database onto another HDD-partition. After testing some files like "__db.001" appeared. Since this moment the database is running well.
    The first error I described above comes up by errors in the database files "__db.001" and the second one by missing the database files.

    So far as good, but I guess u know what will come next. There is another problem.

    If I try to load some files into the container your Loadexapmle replies the following:

    "Using the default cache size.
    ERROR LoaderImpl.java 893  - C:/Users/Tobias/Documents/0000875357/bokf-20100930.xsd encountered a loading problem: F:\Eigene Dateien\Uni\Diplomarbeit\Eclipse_WS\xbrlapi\cache\C\null\null\-1\null\null\Users\Tobias\Documents\0000875357\bokf-20100930.xsd (Das System kann die angegebene Datei nicht finden)
    INFO  LoaderImpl.java 899  - Purged C:/Users/Tobias/Documents/0000875357/bokf-20100930.xsd from the data store.
    WARN  LoaderImpl.java 887  - Storing details of 1 documents that are yet to be loaded.
    WARN  LoaderImpl.java 551  - Some documents failed to load."

    What is the cache about? It stores information about my xbrl files in one file? But why the programm tries to seek the file in "..\C\null\null\-1\null\null\Users\Tobias\Documents.." Is it a problem caused by the entityResolver?

    The uri containes the right link to the file on my hdd, but after using the entityResolver ther are pathes which I dont know where they come from.

    Regards,

    Tobias

     
  • Geoffrey Shuetrim

    The cache is a local directory in which copies of all loaded XBRL documents are stored.  If a document is reloaded, then it is grabbed from the cache instead of the original URI.  The caching system converts the original URI to a directory that can be reverse engineered to get back the original URI.  Hence the bizarre paths for the files stored in the cache.  It does come with the benefit that you can fake putting some documents at the original URI by placing them in the right cache directory.

    My feeling is that you are using a windows path to the file to be loaded, instead of a file system URI.

    Take a look at http://blogs.msdn.com/b/ie/archive/2006/12/06/file-uris-in-windows.aspx for more information.  Bottom line is that you need to load data from a URI, not a file path.

    Try file:///C:/Users/Tobias/Documents/0000875357/bokf-20100930.xsd  I think that is right.

    Regards

    Geoff S

     
  • TJ_DD

    TJ_DD - 2010-11-11

    Hello,

    thank you very much, it works :). I can work on my project, now.

    Regards,

    Tobias

     

Log in to post a comment.

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks