Segmentation fault (error 6) in libvirtd when used with virt-manager (running remotely).
Setup:
Host Machine: alpine-xen-3.3.1 running Xen v4.6.0 and libvirt (verion
1.2.21)
Remote Machine: virt-manager v.1.2.1 on Fedora 22
Actions:
On ‘Remote Machine’ running virt-manager - Create new virtual machine.
configure different parameters. In my case:
(*) 1 vCPU
(*) 2 GB RAM
(*) ISO to boot from
(*) 6GB hard-disk space etc.
On the final screen click ‘Begin Installation’. Virt-Manager looses connection to the alpine-xen host with error message ‘libvirtError: internal error: client socket is closed’
On Host Machine:
/var/log/messages has the following entries (sanitized here):
Feb 26 21:59:58 kern.info kernel: [35597.103557] libvirtd
[4175]: segfault at 742c068f8000 ip 0000742c066f205b sp
0000742c068f4da8 error 6 in ld-musl-x86_64.so.1[742c066a1000+87000]
Feb 26 21:59:58 kern.alert kernel: [35597.103565] grsec:
From : Segmentation fault occurred at 0000742c068f8000 in
/usr/sbin/libvirtd[libvirtd:4175] uid/euid:0/0 gid/egid:0/0, parent
/bin/busybox[init:1] uid/euid:0/0 gid/egid:0/0
Feb 26 21:59:58 kern.alert kernel: [35597.103617] grsec:
From : denied resource overstep by requesting 4096 for
RLIMIT_CORE against limit 0 for /usr/sbin/libvirtd[libvirtd:4175]
uid/euid:0/0 gid/egid:0/0, parent /bin/busybox[init:1] uid/euid:0/0
gid/e
Feb 26 21:59:58 authpriv.info dropbear [4314]: Exit (root):
Disconnect received
/var/log/libvirt/libvirtd.log does not have any entries corresponding to these timestamps.
(from redmine: issue id 5193, created on 2016-02-26)