Skip to main content
ExLibris
  • Subscribe by RSS
  • Ex Libris Knowledge Center

    p_manage_0: I/O error : file 'WORD4_HISTOGRAM'; Cannot create lock file

    • Article Type: General
    • Product: Aleph
    • Product Version: 17.01

    Description:
    We are getting this error trying to index our libraries:

    115 ISAM error: cannot create lock file.

    I found some information at http://docs.hp.com/en/B1031-90061/B1031-90061.pdf and used it to check to see if locking was available.

    Here is what the client returns, which looks good.
    root@lib2:/var/adm$ /usr/bin/rpcinfo -u penguin status
    program 100024 version 1 ready and waiting
    root@lib2:/var/adm$ /usr/bin/rpcinfo -u penguin nlockmgr
    program 100021 version 1 ready and waiting
    program 100021 version 2 ready and waiting
    program 100021 version 3 ready and waiting
    program 100021 version 4 ready and waiting

    However the NFS server is not supporting locking.
    $ /usr/bin/rpcinfo -u lib2.tosm.ttu.edu status
    rpcinfo: RPC: Program not registered
    program 100024 is not available
    $ /usr/bin/rpcinfo -u lib2.tosm.abc.edu nlockmgr
    rpcinfo: RPC: Program not registered
    program 100021 is not available
    $ rpcinfo -u lib2.tosm.abc.edu status
    rpcinfo: RPC: Program not registered
    The document says to start rpc.lock and rpc.statd, but they are not available on the NFS server.

    You all my be familiar with the issue since it is your COBOL application that is trying to lock files. Do you have any suggestions?



    Our application is quitting because it cannot lock a file it is trying to create. See notes below.

    Permissions changed to FULL CONTROL on /iws, maybe more. Please take a look. These files are on /iws/indexscratch.

    p_manage_01_c.log:I/O error : file 'WORD4_HISTOGRAM'
    p_manage_01_c.log:error code: 9/115 (ANS74), pc=0, call=1, seg=0

    Resolution:
    To make sure it is a problem with the /iws FS setting (enable lock), please change the link in abc01 scratch dir from “scratch -> /iws/indexscratch/” to “scratch -> /wrkspc/ttu01_scratch/”

    (you can do cleanup there – delete all from /wrkspc/ttu01_scratch/ dir) and rerun the indexes.

    If all runs ok, we know it is a problem with the /iws FS setting.


    • Article last edited: 10/8/2013