Provisioning aborts before end of installation.

Bug #1012860 reported by Jean-Baptiste Lallement
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
UTAH
Won't Fix
Critical
Unassigned

Bug Description

I ran the command:
 sudo -i -u utah /usr/share/utah/examples/run_utah_tests.py -d -s precise -t server -a i386 /usr/share/utah/examples/master.run 'http://people.canonical.com/~max/max_test.run'

and after a while got the following output:
INFO: Starting the vm using vm-start
INFO: Using vm-wait to ensure VM is active
INFO: Uploading /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb from the host to /tmp on the VM
ERROR: Failed to copy /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb to machine
Exception: Failed to copy /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb to machine
INFO: Destroying vm using vm-remove

Excerpt from the log attached:
2012-06-13 22:53:06,014 utah-4-precise-server-i386 INFO: Starting the vm using vm-start
2012-06-13 22:53:06,014 utah-4-precise-server-i386 DEBUG: Running command: ['vm-start', '-v', '-w', 'utah-4-precise-server-i386']
2012-06-13 22:53:06,014 utah-4-precise-server-i386 DEBUG: Output follows:
2012-06-13 22:53:06,221 utah-4-precise-server-i386 DEBUG: erreur : Le domaine est déjà actif
2012-06-13 22:53:06,223 utah-4-precise-server-i386 DEBUG:
2012-06-13 22:53:06,224 utah-4-precise-server-i386 DEBUG:
2012-06-13 22:53:06,224 utah-4-precise-server-i386 INFO: Using vm-wait to ensure VM is active
2012-06-13 22:53:06,224 utah-4-precise-server-i386 DEBUG: Running command: ['vm-wait', 'utah-4-precise-server-i386', '300']
2012-06-13 22:53:06,224 utah-4-precise-server-i386 DEBUG: Output follows:
2012-06-13 22:58:20,292 utah-4-precise-server-i386 DEBUG: Waiting for 'utah-4-precise-server-i386' to come up: ............................................................................................................................................................................................................................................................................................................ timed out waiting for 'utah-4-precise-server-i386'
2012-06-13 22:58:20,310 utah-4-precise-server-i386 INFO: Uploading /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb from the host to /tmp on the VM
2012-06-13 22:58:20,310 utah-4-precise-server-i386 DEBUG: Running command: ['scp', '-B', '-o', 'StrictHostKeyChecking=no', '/usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb', 'utah-4-precise-server-i386:/tmp']
2012-06-13 22:58:20,310 utah-4-precise-server-i386 DEBUG: Output follows:
2012-06-13 22:58:21,019 utah-4-precise-server-i386 DEBUG: ssh: Could not resolve hostname utah-4-precise-server-i386: Name or service not known
2012-06-13 22:58:25,817 utah-4-precise-server-i386 DEBUG: Client deb path is /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb
2012-06-13 22:58:25,817 utah-4-precise-server-i386 ERROR: Failed to copy /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb to machine
2012-06-13 22:58:26,255 utah-4-precise-server-i386 DEBUG: Client deb path is /usr/share/utah/utah-client_0.2ubuntu311-0~295~quantal1_all.deb

Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :
Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :

Second try, same failure. It seems to cancel the installation while installation is in progress and not finished.

virsh list shows that the VM is running.

Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :

I confirm that installation is still running but UTAH aborted before the end.

summary: - ERROR: Failed to copy /usr/share/utah/utah-
- client_0.2ubuntu311-0~295~quantal1_all.deb to machine
+ Provisionning aborts before end of installation.
summary: - Provisionning aborts before end of installation.
+ Provisioning aborts before end of installation.
Gema Gomez (gema)
Changed in utah:
status: New → Confirmed
importance: Undecided → Critical
Revision history for this message
Max Brustkern (nuclearbob) wrote :

Based on discussion of this today I'm interested in whether DNS was setup correctly for libvirt. I think there is a problem with the setup we are using in Quantal, and Jean-Baptiste was going to send me information from his system so we can update the installation scripts to setup the DNS correctly.

Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :

Here is my /etc/resolv.conf

furthermore trying to resolv an hostname with host returns:
host: parse of /etc/resolv.conf failed

Revision history for this message
Max Brustkern (nuclearbob) wrote :

The latest verion in the dev branch will not write a resollvconf line if ipconfig fails for the bridge address, so it should avoid writing the bad line you saw. Unfortunately, that doesn't actually fix the problem resolving VMs by hostname. Could you send me the output of ipconfig on your system? When I crete a virtual machine and install utah, virbr0 does not appear. I'll try to get Quantal installed on an actual machine so I can look into this further, but if you have a chance, please send me the output of ifconfig on your system so I can see what interfaces are available.

Thanks,
Max

Revision history for this message
Jean-Baptiste Lallement (jibel) wrote :
Revision history for this message
Javier Collado (javier.collado) wrote :

Running the same command, I wasn't able to reproduce the problem.

In may /etc/resolv.conf I see:
nameserver 192.168.122.1 # Added by UTAH

Instead of:
nameserver # Added by UTAH

Let me know if there should be a way to reproduce the problem playing with the network configuration. Otherwise, it looks like this is no longer a problem.

Revision history for this message
Javier Collado (javier.collado) wrote :

@Jean-Baptiste

I'm wondering if the locale in your host machine is set to something different from English. In that case, the problem could be the same as in: bug #1030873

Revision history for this message
Javier Collado (javier.collado) wrote :

Since this bug has been opened for quite a while and the root cause hasn't been identified, I think I'm going to close it as "Won't Fix" and reopen it (or open a new one) if the same problem happens again.

Changed in utah:
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.