Leave feedback
  • Question

    Integration with OpenText Archive - Livelink ECM

Enter a new topic
  • Basavaraj Patil Basavaraj Patil
    0 likes 8366 views

    Hi All,

    My requirement is to integrate StreamServe with OpenText Archive - Livelink ECM. Exact details is,

    1. Producing IXATTR & COMMANDS files along with PDF

    2. Transfer these to OpenText server on the network from where OpenText Pipeline component will pick up to archive

    Question is:

    1. How to use Livelink ECM connector for this purpose?

    2. Can Livelink ECM connector transfer the files across to a different network drive & takes care of any failovers?

    3. Is there a good documentation on this subject?

    Would appreciate if anyone helps me on this.

    Thanks,

    Basavaraj

    Saturday 01 May, 2010
  • Best Answer
    Stefan Cohen Stefan Cohen StreamServe Employee Administrator
    1 likes

    No worries. We are here to help. You didn't write what operating system the boxes are on, so I will try to explain in general terms.

    1. Yes, the connector can deliver the files to a network share. No, it doesn't use ftp. The data is written as if the share was a local file system. The operating system will take care of the network transfer (NFS, SAMBA, etc...). I guess the easiest way to do it is to mount the network share to a local drive or path (this depends on what operating system you are on). I haven't tried it myself but I suspect that you can also write directly to an UNC path. If the network or remote machine goes down, the StreamServer will log the issue, retain the job in the output queue and mark it as failed. Depending on your setup, this might involve retries (configured in queue settings). You can later resend the failed jobs when the network is up again.

    2. I can't say that there is an industry standard for this. Both of the options you list are valid and will work with StreamServe. You may choose the method that works best in your system. I would personally opt for writing directly to a redundant (clustered) network share rather than introducing a third component to take care of the network transfer.

    3. StreamServe does not provide documentation for the content of the COMMANDS and IXATTR files. Please refer to the OpenText documentation for your version of the system.

     

    Tuesday 04 May, 2010
  • Vyv Lomax Vyv Lomax Administrator
    0 likes

    Hello there,

    The Livelink ECM connector's purpose is to generate the IXATTR command file and associated PDF with unique name within a directory folder. The folder can be anywhere on the network that StreamServe has access to. There are a number of options available to tailor input as required.

    Open Text then polls this folder with its pipeline functionality and finally absorbs the file.

    Failover of the file transfer can be handled via external tools if necessary. Normally StreamServe generates the files locally (independant of network failures) and then either Livelink or another application picks up the files.

    Variables from your event and message can be inserted into the Runtime GUI for the connector quite simply. Formmatting of the directory structure and commands is taken care of automatically when in operation. You need to ask for precisely which metadata Livelink requires (as per customer requirements) as to be able to provide them in your project.

    It does not take more than a few attempts before you are perfectly up and running.

    Documentation exists within the Connectors.pdf documentation available opn this site:

    Downloads > Version > User Docuemtation > User Guides (PDF) > Connectors.pdf

    Good luck!

    //557

    Saturday 01 May, 2010
  • Basavaraj Patil Basavaraj Patil
    0 likes

    Hi,

    Thanks so much for that quick response. However, I need bit more details to really conclude some design here, please bear with me.

    1. StreamServe and Opentext are on two different physical boxes within same network. Are you saying, the Livelink ECM connector can deliver the files to any folder over the network? If yes, does that use FTP internally? If network breaks, what happens to the transfer?

    2. What is the regualr practice in the industry? Does the files gets produced locally and then some tool takes care of transferring to destination folder? OR a NFS mount is shared between StreamServe & OpenText to do the file transfer? Can we use FTP connector of StreamServe as a next step to transfer the files?

    3. Also, I am not finding the relevant documentation to populate the variables in COMMANDS & IXATTR files. Is there one?

    Unfortunately, Connectors.pdf do not provide details that I am asking above.

    Sorry, if I am asking for too much !

    Best Regards,

    Basavaraj

    Tuesday 04 May, 2010
  • Stefan Cohen Stefan Cohen StreamServe Employee Administrator
    1 likes

    No worries. We are here to help. You didn't write what operating system the boxes are on, so I will try to explain in general terms.

    1. Yes, the connector can deliver the files to a network share. No, it doesn't use ftp. The data is written as if the share was a local file system. The operating system will take care of the network transfer (NFS, SAMBA, etc...). I guess the easiest way to do it is to mount the network share to a local drive or path (this depends on what operating system you are on). I haven't tried it myself but I suspect that you can also write directly to an UNC path. If the network or remote machine goes down, the StreamServer will log the issue, retain the job in the output queue and mark it as failed. Depending on your setup, this might involve retries (configured in queue settings). You can later resend the failed jobs when the network is up again.

    2. I can't say that there is an industry standard for this. Both of the options you list are valid and will work with StreamServe. You may choose the method that works best in your system. I would personally opt for writing directly to a redundant (clustered) network share rather than introducing a third component to take care of the network transfer.

    3. StreamServe does not provide documentation for the content of the COMMANDS and IXATTR files. Please refer to the OpenText documentation for your version of the system.

     

    Tuesday 04 May, 2010
  • Basavaraj Patil Basavaraj Patil
    0 likes

    This answers my question very closely. Thanks very much. I may comeback with very specific question a bit later in this regard if need be.

    With Best Regards,

    Basavaraj

    Wednesday 12 May, 2010
  • Jim Messa Jim Messa
    0 likes

    We actually use SAP, OpenText and StreamServe all together.  We receive an IDOC from SAP which contains information for several steps in the process plus all the data to render to PDF.  First we make a BAPI call to SAP and get a URI to store the document to in OpenText and then we do an HTTP Put to store the document in OpenText and we return to a BAPI call to update a link in a link table in SAP which allows the SAP user to clink on a link to a document (invoice for example) in SAP and retrieve the PDF that was "Put" there via http.

    I would talk with your OpenText administrator to see what options he can give you for storing the document into OpenText. 

    Wednesday 04 August, 2010
  • Andreas Persson Andreas Persson OpenText Employee StreamServe Employee
    0 likes

    I have another question about the StreamServe output.

    When streamserve creates the PDF and the metadata files, does it always create a new folder that these files are sent to?

    If I have the exact same output going twice to the OT folder..will it create 2 unique folders with the files in, or will it create one and then try to overwrite the files already there?

    thanks Andreas

    Friday 29 October, 2010
  • Stefan Cohen Stefan Cohen StreamServe Employee Administrator
    0 likes

    It is user configurable using the following setting:

    Max. no of documents in meta.

    The maximum number of documents to be stored in each meta document directory. When the maximum number of documents have been added, a new document directory is created.

    Please see the documentation for more info.

    Wednesday 03 November, 2010
  • Andreas Persson Andreas Persson OpenText Employee StreamServe Employee
    0 likes

    ok, thanks. so if we keep it to one then we will get them separated.

     

    Wednesday 03 November, 2010
  • Ralf Schroeder Ralf Schroeder
    0 likes
    Hi all, my question abaout IXOS outconnector: is it possible to set the IXOS Transfer directory name dynamic, or by an $VAR ? Thanks for your answers. Monday 13 December, 2010
  • Andreas Persson Andreas Persson OpenText Employee StreamServe Employee
    0 likes

    Hi, I am using this connector and I am having some problems configuring it.
    at the moment streamserve creates one directory for each PDF created. so if i run a batch of 60.000 documents I end up with 60.000 directories in my archive folder. and in each folder I hav another folder that contains the metadata files and one PDF.

    We are running on Solaris, if that makes a difference and streamserve SP4R2.

    I would like for streamserve to be able to create a directory in my archive folder and fill it up with documents to a certain number and then create a new directory and fill this directory up.

    Does anyone know how I can configure this?
    I have tried many different settings. the closest I have gotten when running a batch file with 10 documents in it is that streamserve create one directory, one IXATTR file with 10 NEWDOCS in it.
    BUT there is one COMMANDS file only referencing to one PDF document.
    BUt there is only one PDF file, where the documents have been appended (and we all know that this does not work).

    I am really hoping that someone know how to set this.

    thanks Andreas

    Friday 14 January, 2011
Next

 

Latest from the blogs