4 Replies Latest reply on Feb 24, 2016 5:56 AM by Vaidas Lauzeckas

    Failed to reap folderId

    Vaidas Lauzeckas

      Hi,

      i am having performance issue with Tableau server 9.2.4. My windows even log is full of tableau errors 4096 of such content:

      2016-02-14 11:37:35.104 +0200 staleFoldersReaperScheduler-1   ERROR : com.tableausoftware.tdfs.filestore.FileReconciliationService - Failed to reap folderId '{46AC13F4-7C13-4821-B6D6-7D33B589C344}' of type 'extract java.io.IOException: Unable to delete file: C:\ProgramData\Tableau\Tableau Server\data\tabsvc\dataengine\extract\bd\2f\{46AC13F4-7C13-4821-B6D6-7D33B589C344}\TDR_GLB_V1.tde at org.apache.commons.io.FileUtils.forceDelete(FileUtils.java:1919)

       

      What could be causing such errors?

       

      Some aspects about my environment:

      1. Running on VM 4 cores 16GB ram. Windows 2012 R2

      2. Extracts are refreshed every hour.

      3. There is postgresql database installed on same server which is used as data source

      4. I had same issue with 9.1.

      5. Tableau is running as "NT AUTHORITY\NetworkService"

      6. I have tried restart with --reset-coordinator, reinstall server nothing helps.

       

      any ideas would be helpful

      V.

        • 1. Re: Failed to reap folderId
          Toby Erkson

          You stated that you had the same issue with 9.1 so I take it you are running Tableau Server in a production environment, correct?  I ask because with such minimal server resources it looks more like a test or POC environment.

          How much free hard drive space do you have?

          • 2. Re: Failed to reap folderId
            Vaidas Lauzeckas

            Hi Tobi, thanks for being interested in this case.

            it is POC/PROD environment. There is only one user using the server and largest data mart is <1GB in space so i presume such specification should be enough. I clearly understand that server specifications are quite minimal, yet it does not explains for me why extract is not reaped (deleted). I hardly imagine situation where server load could forbid to delete the file.

            There is >~100GB free hard space available.

            • 3. Re: Failed to reap folderId
              Toby Erkson

              Thanks for the additional info.  It seems to me you have enough resources so my guess would be that isn't an issue.  Hmm, I don't know

              • 4. Re: Failed to reap folderId
                Vaidas Lauzeckas

                It could be that such behavior is normal. We do incremental extract every hour (new extract file is cretean on each extract). Reaper by default deletes extracts only older then 24 hours, so it could be that it is throwing alarm in case finds extract which yet can't be deleted. You can modify temp extract expire time by

                tabadmin set temp.file_expiry_s 21600

                 

                V.