Hide Forgot
Description of problem: Host get stuck starting VDSM because of permission on "/var/run/libvirt/libvirt-sock" Version-Release number of selected component (if applicable): VDSM-4.13.0-11 ovirt-engine 3.3 How reproducible: Steps to Reproduce: 1. Install new instance of Ovirt 2. Run engine-setup 3. Wait for it to complete Actual results: Installation completes but VDSM fails to start Expected results: installation completes without error Additional info:
Could you attach ovirt-host-deploy logs? Which is your underlying platform (Fedora/EL6)? Do you see any hints for the vdsm startup failure in /var/log/messages?
Created attachment 831841 [details] host-deploy-logs The host-deploy log is attached Here is a snip from the messages log: Dec 2 15:37:28 beetlejuice vdsm TaskManager.Task ERROR Task=`639d0236-2b89-4c95-81c4-d864987aa246`::Unexpected error#012Traceback (most recent call last):#012 File "/usr/share/vdsm/storage/task.py", line 857, in _run#012 return fn(*args, **kargs)#012 File "/usr/share/vdsm/logUtils.py", line 45, in wrapper#012 res = f(*args, **kwargs)#012 File "/usr/share/vdsm/storage/hsm.py", line 615, in getSpmStatus#012 pool = self.getPool(spUUID)#012 File "/usr/share/vdsm/storage/hsm.py", line 318, in getPool#012 raise se.StoragePoolUnknown(spUUID)#012StoragePoolUnknown: Unknown pool id, pool not connected: ('5849b030-626e-47cb-ad90-3ce782d831b3',) Dec 2 15:37:28 beetlejuice kernel: ata1: EH complete Dec 2 15:37:28 beetlejuice kernel: ata2: EH complete Dec 2 15:37:28 beetlejuice kernel: ata3: soft resetting link Dec 2 15:37:28 beetlejuice kernel: ata4: soft resetting link Dec 2 15:37:28 beetlejuice kernel: ata3.00: configured for UDMA/133 Dec 2 15:37:28 beetlejuice kernel: ata3: EH complete Dec 2 15:37:28 beetlejuice kernel: ata4.00: configured for UDMA/100 Dec 2 15:37:28 beetlejuice kernel: ata4: EH complete Dec 2 15:37:30 beetlejuice kernel: device-mapper: table: 253:3: multipath: error getting device Dec 2 15:37:30 beetlejuice kernel: device-mapper: ioctl: error adding target to table Dec 2 15:39:28 beetlejuice ntpd[2276]: 0.0.0.0 061c 0c clock_step +0.337626 s Dec 2 15:39:28 beetlejuice ntpd[2276]: 0.0.0.0 0615 05 clock_sync Dec 2 15:39:29 beetlejuice ntpd[2276]: 0.0.0.0 c618 08 no_sys_peer Dec 2 16:26:57 beetlejuice kernel: device vnet0 entered promiscuous mode Dec 2 16:26:57 beetlejuice kernel: ovirtmgmt: port 2(vnet0) entering forwarding state Dec 2 16:26:58 beetlejuice kernel: ip_tables: (C) 2000-2006 Netfilter Core Team Dec 2 16:26:58 beetlejuice kernel: ovirtmgmt: port 2(vnet0) entering disabled state Dec 2 16:26:58 beetlejuice kernel: device vnet0 left promiscuous mode Dec 2 16:26:58 beetlejuice kernel: ovirtmgmt: port 2(vnet0) entering disabled state Dec 2 16:26:58 beetlejuice vdsm vm.Vm ERROR vmId=`16e87c5f-ffbb-4ac8-a18f-41959347cd5e`::The vm start process failed#012Traceback (most recent call last):#012 File "/usr/share/vdsm/vm.py", line 2092, in _startUnderlyingVm#012 self._run()#012 File "/usr/share/vdsm/vm.py", line 2959, in _run#012 self._connection.createXML(domxml, flags),#012 File "/usr/lib64/python2.6/site-packages/vdsm/libvirtconnection.py", line 76, in wrapper#012 ret = f(*args, **kwargs)#012 File "/usr/lib64/python2.6/site-packages/libvirt.py", line 2665, in createXML#012 if ret is None:raise libvirtError('virDomainCreateXML() failed', conn=self)#012libvirtError: internal error ifname "vnet0" not in key map Dec 2 16:27:00 beetlejuice vdsm vm.Vm WARNING vmId=`16e87c5f-ffbb-4ac8-a18f-41959347cd5e`::trying to set state to Powering down when already Down
please attach also vdsm.log and /var/log/message to the bug. where did you see the permission error? during the start? is it reproducible? it should be: srwxrwx---. 1 root kvm 0 Dec 9 23:46 /var/run/libvirt/libvirt-sock can you copy the output of "ls -la /var/run/libvirt/libvirt-sock" ? Thanks
Hey Maurice James, Can you provide the information to help investigate the issue?
ping?
My emailed response didnt make it to the bug for some reason. I removed libvirt and qemu-kvm then reinstalled Ovirt. The errors went away. I would have loved to troubleshoot it to find the root cause but I didnt have time to dig through it. This bug can be closed. Thanks