Re: [Users] Use Filer for NFS storage

On Sun, Apr 15, 2012 at 04:57:15PM +0300, Dan Kenigsberg wrote: > On Fri, Apr 13, 2012 at 12:26:39PM -0700, Christian Hernandez wrote: > > Here is the log from the Host > > > > > > *Thread-1821::DEBUG::2012-04-13 > > 12:18:52,200::BindingXMLRPC::167::vds::(wrapper) [192.168.11.236] > > Thread-1821::ERROR::2012-04-13 > > 12:18:52,200::BindingXMLRPC::171::vds::(wrapper) Unexpected exception > > Traceback (most recent call last): > > File "/usr/share/vdsm/BindingXMLRPC.py", line 169, in wrapper > > return f(*args, **kwargs) > > File "/usr/share/vdsm/BindingXMLRPC.py", line 571, in > > poolValidateStorageServerConnection > > return pool.validateStorageServerConnection(domType, conList) > > File "/usr/share/vdsm/API.py", line 897, in > > validateStorageServerConnection > > return self._irs.validateStorageServerConnection(domainType, > > AttributeError: 'NoneType' object has no attribute > > 'validateStorageServerConnection' > > Thread-1822::DEBUG::2012-04-13 > > 12:18:52,333::BindingXMLRPC::167::vds::(wrapper) [192.168.11.236] > > Thread-1822::ERROR::2012-04-13 > > 12:18:52,334::BindingXMLRPC::171::vds::(wrapper) Unexpected exception > > Traceback (most recent call last): > > File "/usr/share/vdsm/BindingXMLRPC.py", line 169, in wrapper > > return f(*args, **kwargs) > > File "/usr/share/vdsm/BindingXMLRPC.py", line 491, in > > poolDisconnectStorageServer > > return pool.disconnectStorageServer(domType, conList) > > File "/usr/share/vdsm/API.py", line 823, in disconnectStorageServer > > return self._irs.disconnectStorageServer(domainType, self._UUID, > > AttributeError: 'NoneType' object has no attribute 'disconnectStorageServer' > > It seems like the interesting traceback should be further up - I > suppose self._irs failed initialization and kept its original None > value. Please scroll up and try to find out why this failed on Vdsm > startup. > > We have a FIXME in vdsm so that we report such failures better: > > vdsm/BindingXMLRPC.py: # XXX: Need another way to check if IRS init was okay > > Adam, could you take a further look into this? Have a look at http://gerrit.ovirt.org/3571 . This should handle the problem better by reporting a better error when storage was not initialized properly. -- Adam Litke <agl(a)us.ibm.com&gt; IBM Linux Technology Center