1:54 p.m.
----- Original Message -----
From: "Neil" <nwilson123(a)gmail.com>
To: "Haim Ateya" <hateya(a)redhat.com>
Cc: users(a)ovirt.org
Sent: Monday, October 22, 2012 4:33:11 PM
Subject: Re: [Users] (no subject)
Hi Haim,
On Mon, Oct 22, 2012 at 3:46 PM, Haim Ateya <hateya(a)redhat.com>
wrote:
> please attach installation logs from node under
> /tmp/{vdsm-bootstrap*).
I've checked all of the other logs and they are all the same, so I've
only included the latest one...
[root@node03 tmp]# cat vds_installer.100915.log
Mon, 22 Oct 2012 12:12:04 DEBUG **** Start VDS Installation ****
Mon, 22 Oct 2012 12:12:04 DEBUG get_id_line: read line CentOS
release 6.3 (Final).
Mon, 22 Oct 2012 12:12:04 DEBUG lsb_release: input line CentOS
release 6.3 (Final).
Mon, 22 Oct 2012 12:12:04 DEBUG lsb_release: return: CentOS.
Mon, 22 Oct 2012 12:12:04 DEBUG <BSTRAP component='INSTALLER'
status='OK' message='Test platform succeeded'/>
Mon, 22 Oct 2012 12:12:04 DEBUG <BSTRAP component='INSTALLER LIB'
status='OK' message='Install library already exists'/>
Mon, 22 Oct 2012 12:12:04 DEBUG trying to fetch vds_bootstrap.py
script cmd = '/usr/bin/curl -s -k -w %{http_code} -o
/tmp/vds_bootstrap_06a0a4ee-8935-4743-8c25-acc13971047b.py
http://backup.blabla.com/vds_bootstrap.py'
Mon, 22 Oct 2012 12:12:04 DEBUG <BSTRAP component='INSTALLER'
status='FAIL' message='vds_bootstrap.py download failed. Pathname
could not be resolved (verify computer/domain name).'/>
[root@node03 tmp]#
> also run the following command from your engine machine:
>
> psql -U postgres engine -c 'select * from vds_static;'
Output is attached as output.txt, shows all three hosts along with
the
correct ip's, node03 has no power management as of yet.
> please make sure 'node02.blabla.com' is accessible from host side
> (first try ping), later try with http request (simply issue wget
> http://node02.blabla.com:80/Components/vds/vds_bootstrap.py)
Both tests worked fine... node02 is one of my hosts, it used to be my
old engine but was changed, backup.blabla.com is my new working
engine.
so, the problem is that in your engine, VdcBootstrapUrl points to your old
engine address (node02.blabla.com:80) instead of your new one (backup.blabla.com), this
confuses the host as it tries to download
the script from the old engine, and fails since they are not there.
don't really know how you got to this state, but in order to resolve it, please apply
the following query:
psql -U postgres engine -c "update vdc_options set option_value =
'backup.blabla.com:80/Components/vds/' where option_name =
'VdcBootStrapUrl'"
[root@node03 ~]# wget
http://backup.blabla.com:80/Components/vds/vds_bootstrap.py
Connecting to backup.blabla.com|10.0.2.31|:80... connected.
HTTP request sent, awaiting response... 200 OK
Saving to: âvds_bootstrap.pyâ
2012-10-22 16:08:52 (72.6 MB/s) - âvds_bootstrap.pyâ
.
Please shout if you need any further info.
Thank you!
Neil Wilson.