AppWorx, Dollar Universe and Sysload Community

 View Only

First Time Upgrading Applications Manager

  • 1.  First Time Upgrading Applications Manager

    Posted Aug 27, 2020 04:23 PM
    Does anyone know about this stuff?  I've upgraded the Automation Engine and the Local Agent, in our DEV environment, more or less without incident. I had trouble getting my head around locating a correct ojdbc6 file. Once I understood what CDINST.SH was trying to accomplish, I was able to get to the end and eventually startup the Automation Engine. The Remote Agents, one each predefined for our two "Banner by Ellucian" test environments, both failed to start during the routine. Both deliver a "did not start at /app/uc4/master/data/inst.pl line 5567." message.

    Next challenge was to run the client, which I couldn't do the established way (running a client.jnlp from the server). The install guide does explain about acquiring and preparing to use the RunClient.jar file. This approach should be just as easy to implement as our old one, and I was able to login to the familiar client software with its Backlog display.

    The install guide also indicates that any remote agents need to be upgraded as well. There seem to be contradictions, though. In one spot I'm told to run the Remote Agent install script awagentinstall to upgrade a Remote Agent. In another spot, I'm told that the upgrade script for Remote Agents is CDINST.SH. I can't find a script named awagentinstall anywhere, leaving CDINST.SH as my only option. When I run this, I am currently experiencing slowdowns and the periodic report of a client pipe timing out, and ultimately the following:

    executing awexe id_recrypt
    231 Error number from open pipe 2. /app/uc4/master/pipe/AWAPI_UC4DEVL_PIPE
    334 Check that the API server is running.
    Could not open server pipe.

    Check Oae upgrade
    231 Error number from open pipe 2. /app/uc4/master/pipe/AWAPI_UC4DEVL_PIPE
    334 Check that the API server is running.

    I looked at /app/uc4/master/pipe, and found this:

    $ cd /app/uc4/master/pipe
    $ ls -l
    total 0
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 10:37 AWAPI_retry_1_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 14:12 AWAPI_UC4DEVL_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 09:48 AWEXE_10875_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 09:01 AWEXE_19263_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 09:04 AWEXE_20313_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 14:12 AWEXE_22541_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 27 14:12 AWEXE_22598_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 20 12:37 AWEXE_26497_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 20 14:01 AWEXE_29715_PIPE
    prw-rw-rw-. 1 uc4 banner 0 Aug 20 10:22 AWEXE_4064_PIPE

    Comparing this directory to the same directory on our production UC4 server, all the aboves "files" that have 4- and 5-digit numbers embedded in the names are missing from production. Permissions are the same, though.

    What's the problem with AWAPI_UC4DEVL_PIPE?
    What is the meaning of "Error number from open pipe 2."? Is 2 the error number being reported from AWAPI_UC4DEVL_PIPE? Or is 2 the number of the pipe in question (it is second in the list)?
    What is an API server and how can I check whether it is running?

    All of this, by the way, happens before I am prompted for my choice to install the Automation Engine or to install a Remote Agent. I have gotten that far a couple times, but now this PIPE is getting in the way. When I do get to make that choice, what has happened is that the script terminates immediately, stating a failure but offering no reason why and not indicating a log file that may be referenced.

    Where can I find out more about this particular failure?

    ------------------------------
    Bob Fathergill
    ------------------------------