VCS WARNING V-16-1-10367 Dump already in progress - Veritas Cluster Server

This is a discussion on VCS WARNING V-16-1-10367 Dump already in progress - Veritas Cluster Server ; Hi All, I am trying to save my VCS configuration to disk and I am getting the message above (subject of this POST) I have tried both from the command line and the JAVA GUI console. I have a general ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: VCS WARNING V-16-1-10367 Dump already in progress

  1. VCS WARNING V-16-1-10367 Dump already in progress


    Hi All,

    I am trying to save my VCS configuration to disk and I am getting the message
    above (subject of this POST)
    I have tried both from the command line and the JAVA GUI console.
    I have a general Solaris O/S upgrade coming up in a couple of days on all
    the systems in the cluster and would like to sort this out before I shutdown
    the servers so as not to loose my in-memory configuration which has some
    recent changes.
    Any ideas of how to force a save to disk? It's been like thsi for a couple
    of weeks or so. Thanks.

    root@Calldata # haconf -dump
    VCS WARNING V-16-1-10367 Dump already in progress
    root@Calldata # haconf -dump -makero
    VCS WARNING V-16-1-10367 Dump already in progress

  2. Re: VCS WARNING V-16-1-10367 Dump already in progress

    The issue is an OS one.

    I bet you have Solaris.

    There is IPC communications between the different parts of VCS. So,
    saving the configuration will open an IPC handle. The big issue here is
    that Solaris only allows a max. of 8 connections per process. (There is
    a variable called SOMAXCONN that can be set).

    Veritas has "fixed" the issue by making sure that not everything needs
    to have a handle open at all times (you most likely have VCS 3.5 ?)

    Just contact support and they will help you out.



    For the moment, what you need to do is to check the
    /etc/VRTSvcs/conf/config/main.cf to make sure that your latest changes
    has been saved and once you've shut down, you might need to go and
    remove a file /etc/VRTSvcs/conf/config/.stale before VCS will start up.



    Just for your info. The other Operatins Systems have a lot more
    SOMAXCONN connections. HP-UX has a default value of 4096, AIX has a
    default of 2048.

    Ayotunde Itayemi wrote:
    > Hi All,
    >
    > I am trying to save my VCS configuration to disk and I am getting the message
    > above (subject of this POST)
    > I have tried both from the command line and the JAVA GUI console.
    > I have a general Solaris O/S upgrade coming up in a couple of days on all
    > the systems in the cluster and would like to sort this out before I shutdown
    > the servers so as not to loose my in-memory configuration which has some
    > recent changes.
    > Any ideas of how to force a save to disk? It's been like thsi for a couple
    > of weeks or so. Thanks.
    >
    > root@Calldata # haconf -dump
    > VCS WARNING V-16-1-10367 Dump already in progress
    > root@Calldata # haconf -dump -makero
    > VCS WARNING V-16-1-10367 Dump already in progress


  3. Re: VCS WARNING V-16-1-10367 Dump already in progress


    "Ayotunde Itayemi" wrote:
    >
    >Hi All,
    >
    >I am trying to save my VCS configuration to disk and I am getting the message
    >above (subject of this POST)
    >I have tried both from the command line and the JAVA GUI console.
    >I have a general Solaris O/S upgrade coming up in a couple of days on all
    >the systems in the cluster and would like to sort this out before I shutdown
    >the servers so as not to loose my in-memory configuration which has some
    >recent changes.
    >Any ideas of how to force a save to disk? It's been like thsi for a couple
    >of weeks or so. Thanks.
    >
    >root@Calldata # haconf -dump
    >VCS WARNING V-16-1-10367 Dump already in progress
    >root@Calldata # haconf -dump -makero
    >VCS WARNING V-16-1-10367 Dump already in progress


    The configuration files are in /etc/VRTSvcs/conf/config and they are at least
    the main.cf and types.cf files. Actually everything in that directory with
    the cf extension should be backed up to another location. The haconf -dump
    and makero only means that you are going to modify the main.cf file not DUMP
    it to another
    location.

    From man haconf:
    -dump [-makero]
    Write the configuration to disk and remove the
    designation stale. Change the VCS mode to read-
    only. -makero changes the VCS mode to read-only.


+ Reply to Thread