Please consider this matter closed (at least for me).

We ended up replacing our the HP DL180 gen 6 controller/Fuel servers with Dell r610s and the problem is now non-existent.

I'm not sure what the problem was related to however I can tell you I've received that same 'nailgun server not available' error in the Fuel UI a few times when attempting to save things to many nodes at once within our working MOS 9.2 install using Dell R610s. Reducing the number of nodes I'm modifying the config for or simply trying again works around the rare issue now.

I appreciate your help!


On 04/05/2017 02:35 PM, Evgeniy L wrote:
Given the fact that you can see settings tab with parameters, the error
"Nailgun server is unavailable.  Please check your connection and
contact your system administrator." most likely means that Nailgun
responded with 5xx error, which may happen because of some unhandled
exception.

1. We need to confirm that the problem is indeed in unhandled exception.
In "/var/log/nailgun/api.log" you should see an api request from your
browser and corresponding response code.

2. Without python traceback/logs it's hard to say what went wrong, as
far as I remember python traceback will be in "/var/log/nailgun/api.log"
file alongside with a request from your browser.

Thanks,

On Wed, Apr 5, 2017 at 9:29 AM, sil...@gmail.com
<mailto:sil...@gmail.com> <sil...@gmail.com <mailto:sil...@gmail.com>>
wrote:

    When I attempt to save any changes to a cloud environment I've
    created, such as within 'Setting' changing 'Hypervisor type' in the
    'Compute' setting menu from 'QEMU' to 'KVM', the save button circles
    for about 7-8secs and then a pop-up dialog box comes up that says:

    OpenStack Settings Configuration Error

    Nailgun server is unavailable.  Please check your connection and
    contact your system administrator.

    This happens while trying to use any version of Fuel greater than
    9.0 on a HP DL180 gen6 server (I also tried version 11 with the same
    results). I can use any version of Fuel without problems using
    identical settings on a Dell r610 server.

    I checked all of the logs in /var/log/nailgun and none of them
    indicate any errors that I can see.  There are also plenty of
    nailgun related processes running on the Fuel server.

    On 04/05/2017 10:14 AM, Evgeniy L wrote:

        Hi,

        Could you please clarify, do you receive this error in the UI?
        What do
        you mean by "save any changes"?
        I'm not sure if ranges can cause such behavior, please see logs
        "/var/log/nailgun/app.log" and "/var/log/nailgun/api.log" for any
        errors/python tracebacks.

        Thanks,

        On Mon, Apr 3, 2017 at 8:12 PM, sil...@gmail.com
        <mailto:sil...@gmail.com>
        <mailto:sil...@gmail.com <mailto:sil...@gmail.com>>
        <sil...@gmail.com <mailto:sil...@gmail.com>
        <mailto:sil...@gmail.com <mailto:sil...@gmail.com>>>
        wrote:

            Has anyone had a problem with Mirantis Fuel giving a
        'nailgund not
            available' error when attempting to save any changes to a cloud
            environment?  I'm encountering this problem after a fresh
        install of
            Mirantis Fuel 9.0 and then immediately upgrading to 9.2
        (didn't even
            login to the Fuel web interface prior to the upgrade).

            I've verified that nailgund is running and the network
        configured in
            'fuelmenu' during the initial 9.0 install is legit with Internet
            connectivity (172.16.0.0/22 <http://172.16.0.0/22>
        <http://172.16.0.0/22> network,
            172.16.3.250 is fuel server, 172.16.0.1 is the Internet
        gateway and
            172.16.1.1-172.16.2.254 is handed out via the fuel DHCP server).
            I've also verified that no other DHCP servers exist on this
        network,
            no errors or warnings in the 'fuelmenu' portion of the
        installation.

            I've installed this before using the same exact
        methods/options on
            some Dell hardware in a lab of mine however I'm having this
        problem
            on a production deployment using some older HP servers.  The
        only
            difference between the functional installation and this one
        is the
            IP ranges that were used and I set a root/admin password on the
            production setup (I left admin/root at default on the functional
            setup).  I have a suspicion that using the 172.16.0.0/22
        <http://172.16.0.0/22>
            <http://172.16.0.0/22> network as the admin/PXE network for fuel
            deployments might be the culprit as I noticed that the nailgund
            config file defines '172.16.0.0/16 <http://172.16.0.0/16>
        <http://172.16.0.0/16>' as
            'PUBLIC' in the template section as shown below:

            /etc/nailgun/settings.yaml

            "
            STATIC_DIR: "/usr/share/nailgun/static"
            TEMPLATE_DIR: "/usr/share/nailgun/static"
            NETWORK_POOLS:
              public:
                - "172.16.0.0/16 <http://172.16.0.0/16>
        <http://172.16.0.0/16>"
              private10:
                - "10.0.0.0/8 <http://10.0.0.0/8> <http://10.0.0.0/8>"
              private192:
                - "192.168.0.0/16 <http://192.168.0.0/16>
        <http://192.168.0.0/16>"

            NET_EXCLUDE:
              - "172.16.0.0/22 <http://172.16.0.0/22>
        <http://172.16.0.0/22>"

            ADMIN_NETWORK:
              cidr: "172.16.0.0/22 <http://172.16.0.0/22>
        <http://172.16.0.0/22>"
              netmask: "255.255.252.0"
              mac: "e4:11:5b:e6:90:3a"
              size: "1024"
              first: "172.16.1.1"
              last: "172.16.2.254"
              gateway: "172.16.3.250"

            VLANS_RANGE_START: "100"
            VLANS_RANGE_END: "1000"
            "

            I tried using Mirantis Fuel 9.0 without patching and it doesn't
            exhibit this same problem although this time around I left the
            admin/root passwords at default and did not set them as I
        did with
            the previous installations when I upgraded directly to 9.2
        so I'm
            not sure if that is factoring in here.

            Any ideas what the issue may be?

            Also, sorry if this is a repost, I sent this email many
        hours ago
            and didn't see it appear on the mailing list archives, not
        sure what
            the issue is or if I didn't wait long enough.

            silox

            _______________________________________________
            Mailing list:

        http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack>

        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack>>
            Post to     : openstack@lists.openstack.org
        <mailto:openstack@lists.openstack.org>
            <mailto:openstack@lists.openstack.org
        <mailto:openstack@lists.openstack.org>>
            Unsubscribe :

        http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack>

        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
        <http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack>>






_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Reply via email to