[Planetlab-devel] Not fully booted

[Robert Soule]

OK, I see where the failing node, and a node that works deviate in their
log files:

Step: Chain booting node.
01:51:16(UTC) Updated /etc/planetlab/session
01:51:16(UTC) Updating configuration files.
01:51:16(UTC) chroot: cannot run command `/etc/init.d/conf_files': No such file
or directory
01:51:16(UTC)

Exception while running: Running chroot /tmp/mnt/sysimg /etc/init.d/conf_files s
tart --noscripts failed (rc=32512)
01:51:16(UTC)


One other thing I notices is that the succeeding node has :

23:58:58(UTC) Primary network as returned from PLC: {'network': None,
'ip': '128.122.140.154', 'dns1': None, 'hostname': None, 'netmask':
None, 'interface_tag_ids': [], 'interface_id': 1, 'broadcast': None,
'method': 'dhcp', 'mac': None, 'node_id': 1, 'gateway': None, 'dns2':
None, 'is_primary': True, 'type': 'ipv4', 'bwlimit': None}

And the failing node has:

01:51:08(UTC) Primary network as returned from PLC: {'network': '',
'ip': '128.122.140.158', 'dns1': '', 'hostname': 'nakika2.cs.nyu.edu',
'netmask': '', 'interface_tag_ids': [], 'interface_id': 2,
'broadcast': '', 'method': 'dhcp', 'mac': None, 'node_id': 2,
'gateway': '', 'dns2': '', 'is_primary': True, 'type': 'ipv4',
'bwlimit': 0}
01:51:08(UTC)


That is, one has "None" and one has blank default values. Would that
make a difference?

thank you for your help,
Robert



2009/6/5 Robert Soule <robert.soule at gmail.com>:
> Thanks for pointing me to the logs. They don't really give me an
> explanation though...
>
>
> Step: Validating node installation.
> 01:51:17(UTC) Checking for a custom kernel
> 01:51:17(UTC) Updated /etc/planetlab/node_id
> 01:51:17(UTC) Node installation appears to be ok
> 01:51:17(UTC)
> Error occurred!
> 01:51:17(UTC) BootManager finished at: Sat, 06 Jun 2009 01:51:17 +0000
> 01:51:17(UTC) Uploading logs to /boot/upload-bmlog.php
>
>
>
>
>
>
> On Fri, Jun 5, 2009 at 8:31 PM, Marc Fiuczynski<mef at cs.princeton.edu> wrote:
>> Hi Robert
>>
>> On the node in /tmp/bm.log is the log for the bootmanager. Take a peek at
>> the messages towards the end of this log file for an indication of the
>> failure.
>>
>> Marc
>>
>> Sent from my iPod
>>
>> On Jun 5, 2009, at 7:57 PM, Barış Metin <Talip-Baris.Metin at sophia.inria.fr>
>> wrote:
>>
>>> Hi Robert,
>>>
>>> I'm assuming that that's a node of a private plc. If the node was able
>>> to update it's boot status your plc will have the full log at
>>> /var/log/bm/. Also on the node the sources for the BootManager will be
>>> at /tmp/source. You can try to run BootManager.py, if needed with
>>> enabling breakpoints in utils.py, and investigate the problem further.
>>>
>>> regards,
>>>
>>> On Sat, Jun 6, 2009 at 1:40 AM, Robert Soule<robert.soule at gmail.com>
>>> wrote:
>>>>
>>>> Hello,
>>>>
>>>> I've set up a node to run off of a usb. When I bring the node up, it
>>>> seems to come up
>>>> fine, and I can log in to it using my plc's root password. However,
>>>> there is a message
>>>> that says "not fully booted yet", but no error messages explain why. From
>>>> my plc
>>>> web interface, I can see that the machines have communicated with
>>>> eachother,
>>>> but the status is listed as failed to boot. Can anyone tell me what
>>>> may be going wrong?
>>>>
>>>> thank you,
>>>> Robert
>>>>
>>>> _______________________________________________
>>>> Devel mailing list
>>>> Devel at lists.planet-lab.org
>>>> https://lists.planet-lab.org/mailman/listinfo/devel
>>>>
>>>
>>>
>>>
>>> --
>>> Barış
>>>
>>> _______________________________________________
>>> Devel mailing list
>>> Devel at lists.planet-lab.org
>>> https://lists.planet-lab.org/mailman/listinfo/devel
>>
>> _______________________________________________
>> Devel mailing list
>> Devel at lists.planet-lab.org
>> https://lists.planet-lab.org/mailman/listinfo/devel
>>
>