openstack team mailing list archive

[Gui Maluf <guimalufb@xxxxxxxxx>]

I don't know if it is related to this hostname issue, but my nova-compute
on cloudcontroller can't initiate an iscsi volume in the same controller.

2012-12-14 14:13:46 DEBUG nova.utils
[req-98c1f3b8-ac10-4aa6-a806-f2dd18928af2 None None] Result was 255 from
(pid=5250) execute /usr/lib/python2.7/dist-packages/nova/utils.py:235
2012-12-14 14:13:46 CRITICAL nova [-] Unexpected error while running
command.
Command: sudo nova-rootwrap iscsiadm -m node -T
iqn.2010-10.org.openstack:volume-00000006 -p 200.131.6.250:3260 --rescan
Exit code: 255
2012-12-14 14:13:46 TRACE nova ProcessExecutionError: Unexpected error
while running command.
2012-12-14 14:13:46 TRACE nova Command: sudo nova-rootwrap iscsiadm -m node
-T iqn.2010-10.org.openstack:volume-00000006 -p 200.131.6.250:3260 --rescan
2012-12-14 14:13:46 TRACE nova Exit code: 255
2012-12-14 14:13:46 TRACE nova Stdout:
2012-12-14 14:13:46 TRACE nova Stderr: 'iscsiadm: No portal found.\n'

fulll nova-compute.log: http://paste.openstack.org/show/28109/


On Fri, Dec 14, 2012 at 1:45 PM, Gui Maluf <guimalufb@xxxxxxxxx> wrote:

> Razique, thanks for the tip.
> To find out which tables has the hostname set I did this:
> # for i in `mysql -uroot -p'PASSWORD' -e 'show tables;' nova | grep -v -e
> '+' -e 'T'`; do echo $i && mysql -uroot -p'PASSWORD' -e "select * from
> $i";' nova | grep OLD_HOSTNAME; done
> This will print many data with the name of the tables and after all
> occurrences of the HOSTNAME inside the tables.
>
> Then I've changed all host to the correct hostname
>
> # for i in {187..190} 193 195; do mysql -uroot -p'PASSWORD' -e "update
> TABLE set host=NEW_HOSTNAME where id =$i;" nova; done
>
> i.e: # for i in {187..190} 193 195; do mysql -uroot -p'PASSWORD' -e
> "update floating_ips set host='oxala' where id =$i;" nova; done
>
> Thanks man! it solved my a part of my issue.
>
>
>
>
> On Fri, Dec 14, 2012 at 12:59 PM, Razique Mahroua <
> razique.mahroua@xxxxxxxxx> wrote:
>
>> Check the nova database, the network tables and fixed_ip contains the
>> hostname.
>> Update these fields too and you should be ok :)
>> (Maybe your /etc/hosts files, keystone endpoints if you used hostnames)
>>
>> *Razique Mahroua** - **Nuage & Co*
>> razique.mahroua@xxxxxxxxx
>> Tel : +33 9 72 37 94 15
>>
>>
>> Le 14 déc. 2012 à 15:43, Gui Maluf <guimalufb@xxxxxxxxx> a écrit :
>>
>> Hey guys. I think I did a stupid thing.
>> My setup is 1 cloudcontroller + node and 2 node with FlatDHCP, mult_host,
>> net HA.
>>
>> Recently I've changed the hostname of my cloudcontroller; to make things
>> consistent I've updated the services table inside nova db to mark as
>> deleted=1 the lines where host=oldhostname.
>>
>> I dont know what happened because now nothing of the nova-network works.
>> It doesn't create br100 anymore, neither iptables rules of SNAT and DNAT
>>
>> The worse thing about that problem, is that this is  production setup.
>> All my instances from the cloudcontroller are unavaible cause of this.
>> Nodes are creating everything regularly
>>
>> I couldn't found nothing wrong on logs, no errors. Anyway, here are my
>> logs and setup files.
>>
>> controller nova.conf: http://paste.openstack.org/show/28095/
>> node nova.conf: http://paste.openstack.org/show/28096/
>> iptables -L -t nat + brctl show + virsh list:
>> http://paste.openstack.org/show/28098/
>>
>> Any other information please let me know!
>> I need to fix this ASAP
>>
>> --
>> *guilherme* \n
>> \t *maluf*
>>
>>  _______________________________________________
>> Mailing list: https://launchpad.net/~openstack
>> Post to     : openstack@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~openstack
>> More help   : https://help.launchpad.net/ListHelp
>>
>>
>>
>
>
> --
> *guilherme* \n
> \t *maluf*
>
>


-- 
*guilherme* \n
\t *maluf*