hi,

my issue was related to bug in libvirtd.
it was found in core dump by libvirt team

"

I'll send a patch to upstream libvirt to fix this crash.  However it can take
a while to get it back to CentOS/RHEL.  The source of this crash is that you
have a "tun0" network interface without IP address and that interface is
checked before "ovirtmgmt" and it causes the crash.  You can workaround it
by removing the "tun0" interface if it doesn't have any IP address.

Pavel

"

workaround is working fine for me.

Regards,
Rafal Wojciechowski

W dniu 18.04.2017 o 16:55, Rafał Wojciechowski pisze:

hi,

I was unable to just remove glibc and install it again - I have reinstalled it and rebooted the machine but it was not fixed anything
Thanks anyway.

Regards,
Rafal Wojciechowski

W dniu 18.04.2017 o 10:53, Yanir Quinn pisze:
Hi Rafal
not sure it relates to your issue, but i experienced a similar issue with segfault (running on fedora 25) to resolve it i had to remove glibc packages and then install them again (maybe a same workaround for libvirt will do the job here)

Regards
Yanir Quinn

On Tue, Apr 18, 2017 at 9:32 AM, Francesco Romani <from...@redhat.com <mailto:from...@redhat.com>> wrote:



    On 04/18/2017 08:09 AM, Rafał Wojciechowski wrote:
    >
    > hello,
    >
    > I made comparison(+diff) between xml passing through vdsm which is
    > working and another one which cause libvirtd segfault
    >
    >
    
https://paste.fedoraproject.org/paste/eqpe8Byu2l-3SRdXc6LTLl5M1UNdIGYhyRLivL9gydE=
    
<https://paste.fedoraproject.org/paste/eqpe8Byu2l-3SRdXc6LTLl5M1UNdIGYhyRLivL9gydE=>
    >
    >
    > I am not sure if below setting are fine but I dont know how to
    change them
    >
    > <model heads="1" ram="65536" type="qxl" vgamem="16384"
    vram="8192" />
    > (I dont have so much ram and vgamem)
    >

    those are kibibytes though
    (https://libvirt.org/formatdomain.html#elementsVideo
    <https://libvirt.org/formatdomain.html#elementsVideo>), are pretty
    conservarvative settings
    >
    > <graphics autoport="yes" defaultMode="secure" passwd="*****"
    > passwdValidTo="1970-01-01T00:00:01" port="-1" tlsPort="-1"
    type="spice">
    > (ports with "-"? maybe it is fine because of autoport settings...)
    >

    Yes, "-1" means "autoallocation from libvirt".
    I don't see obvious issues in this XML, and, most importantly, one
    invalid XML should never cause libvirtd to segfault.

    I'd file a libvirt bug.


    --
    Francesco Romani
    Senior SW Eng., Virtualization R&D
    Red Hat
    IRC: fromani github: @fromanirh

    _______________________________________________
    Users mailing list
    Users@ovirt.org <mailto:Users@ovirt.org>
    http://lists.ovirt.org/mailman/listinfo/users
    <http://lists.ovirt.org/mailman/listinfo/users>





_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Reply via email to