This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.
You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

HP ProCurve Software Updates

I am doing an eval of NCM for my network of about 500 ProCurve switches. One thing I have not been able to resolve is having a successful reboot after a software upgrade. I am doing this via a simple script which is a scheduled job. The software upgrade works, but the reboot never occurs. I have been able to get the reboot to work in other scripts. For example, following a "max-vlans" or "console inactivity-timer", the reboot works fine.

In the case of the software upgrade, I suspect it is the delay during the time the image is transferred and uncompressed that causes the problem. Has anyone found a way to overcome this?

Thanks,

John

  • Hi John,

    have you checked the log file?

    It's found in {NCMInstallDir}\Jobs. The name:

    NCM-Jobs-Logging.PNG

    Regards,

    Jiri

  • Hi Jiri,

    Sure, it looks fine, but the reboot never occurs...

    5/21/2012 2:01:21 PM : Started TEST Update Software and Reload : Job-146293

    Execute Command Script on Devices

    1 devices selected

    NEW-Dana-DS1 (172.16.200.249):

    --------------

    copy tftp flash 172.16.35.9 K_15_04_0007.swi pri

    --------------

    --------------

    y

    --------------

    coptftp flash 172.16.35.9 K_15_04_0007.swi priThe PrimarOS Image will be deleted, continue [y/n]?  y

    00009K

    00019K

    00029K

    00039K

    00048K

    00058K

    00068K

    00078K

    00087K

    00097K

    00107K

    *********************************

    File Transfer Detail Omitted

    *********************************

    12128K

    12138K

    12148K

    12158K

    12167K

    12177K

    12187K

    --------------

    y

    --------------

    --------------

    reload

    --------------

    --------------

    y

    --------------

    --------------

    y

    --------------

    --------------

    end

    --------------

    ___________________________________________________________________________

    5/21/2012 2:02:29 PM : Completed TEST Update Software and Reload : Job-146293 ; Devices : 1 ; Errors : 0

    Execution time : 1 minutes, 8 seconds

    ___________________________________________________________________________

  • John,

    Would the obvious workaround (scheduling a separate job to reboot) work for you, or would you like me to try to find some solution offline?

    Regards,

    Jiri

  • I would be happier if there is a way to tell the script manager to wait 60 or 90 seconds, if that is possible, before running the reload command. If no other HP ProCurve user has been able to resolve this, I can certainly get by with scheduling two separate jobs until this can be remedied. However, this does create some obvious opportunities for errors if multiple nodes are involved.

  • Two ideas:

    1. Does the device itself have a command to wait?
    2. If not, you could try something with the RebootAt macro.

    Just ideas...

    Regards,

    Jiri

  • I'm testing using "reload after 00:00:10" to schedule a reload in 10 minutes. I'm using this in the script before the software image is transferred.

    This seems to work, although not all of my switches support this command syntax, so I'm back to 2 different jobs for those units.

    Thanks Jiri,

    John