[Spce-user] Updated EC2 AMIs for mr3.4.1

[George Mason]

> * George Mason [Thu Aug 21, 2014 at 06:09:44PM +0100]:
>> On 21/08/14 15:04, Michael Prokop wrote:
>>> on mr3.4.1 will provide the updated package ngcp-system-tools-ce
>>> version 0.12.1.4+0~mr3.4.1.4 which includes this change to address
>>> the automatic advertised address adjustment on EC2 systems.
>
>> Have just updated my EC2 instance, can see that ngcp-system-tools-ce has
>> upgraded to 3.4.1.4 but calls still drop after 30s. Unless I'm
>> misunderstanding, that update should have cured the elastic IP issue,
>> which should have resolved this?
>
> The ngcp-system-tools-ce update just ensures that the advertised IP
> configuration in network.yml is OK in the classic EC2 scenario. You
> can verify this on your own by checking whether the configuration in
> /etc/ngcp-config/network.yml matches the IP address reported via
> executing 'facter ec2_public_ipv4' on the command line.
Thanks - yes this has worked fine, the two match up.
>
>
> What's still missing is the fix to make sure that db-schema and
> cfg-schema work independent from the hostname (this is fixed in our
> "trunk" already but is waiting for an according release yet).
>
> But did you adjust the node name entries in the ngcp.cfg_schema and
> ngcp.db_schema tables so you're fully up2date with:
>
>   apt-get update && apt-get upgrade && \
>   ngcp-update-db-schema && ngcp-update-cfg-schema && \
>   ngcpcfg apply
>
> yet? Because if you've up2date db-schema and cfg-schema changes
> applied that should be working fine AFAIK, if not please let us
> know.
Unfortunately not working for me. I updated the node (to the EC2
internal hostname - in this case ip-10-237-180-106) in both db_schema
and cfg_schema and re-ran the update as above, but still receive the
following error on upgrade:

Applying revision script
/usr/share/ngcp-db-schema/db_scripts/base/0070_not_replicated.up: ERROR
1050 (42S01) at line 2: Table 'acc_backup' already exists
failed. :(