Recovering from failed install - [SOLVED]

[Piers <pi...@rowan.id.au>]

For the benefit of others:

I have been installing, wiping and reinstalling ACS for a couple of days 
now.

I get the impression that an initial install created network 
configurations that were incompatible with re-installation. An indicator 
of this was the output line in

cloudstack-setup-managment

# I dont use 169.254.0.1
Detected local IP address as 169.254.0.1, will use as cluster management 
server node IP[ OK ]

# X.X.X.X is my real IP
Detected local IP address as X.X.X.X, will use as cluster management 
server node IP[ OK ]


The only way things worked was removing the network settings by 
commenting out the bridge line in ifcfg-eth0 and manually deleting the 
bridges:

[root@vhost2 images]# brctl show
bridge name    bridge id        STP enabled    interfaces
cloud0        8000.000000000000    no
cloudbr0        8000.003048fc6bb4    no        eth0
virbr0        8000.000000000000    yes


[root@vhost2 images]# brctl delbr cloud0
[root@vhost2 images]# brctl delbr cloudbr0
[root@vhost2 images]# brctl show


bridge name    bridge id        STP enabled    interfaces
virbr0        8000.000000000000    yes


[root@vhost2 images]# service network restart

I know have a working web UI and the SSVM nodes started and all looks good.

My advice to anyone having trouble is:

- Make sure you wipe everything before starting
- set keepcache=1 in yum.conf  (so you dont keep downloading the same 
packages over and over)
- down load the image using wget and us the -f option to decompress it 
/usr/share/cloudstack-common/scripts/storage/secondary/cloud-install-sys-tmplt
- purge the SQL databases
- check and remove the network interfaces - NOTE: I use two NIC's so 
that if I kill the public IP I can connect from the next server in the rack

Hopefully this helps some else / feel free to correct or criticise 
anything I have said in error

Cheers

Piers

Re: Recovering from failed install - [SOLVED]

[ilya musayev <il...@gmail.com>]

Thanks for posting back your solution.

On 8/27/14, 3:14 PM, Piers wrote:
> For the benefit of others:
>
> I have been installing, wiping and reinstalling ACS for a couple of 
> days now.
>
> I get the impression that an initial install created network 
> configurations that were incompatible with re-installation. An 
> indicator of this was the output line in
>
> cloudstack-setup-managment
>
> # I dont use 169.254.0.1
> Detected local IP address as 169.254.0.1, will use as cluster 
> management server node IP[ OK ]
>
> # X.X.X.X is my real IP
> Detected local IP address as X.X.X.X, will use as cluster management 
> server node IP[ OK ]
>
>
> The only way things worked was removing the network settings by 
> commenting out the bridge line in ifcfg-eth0 and manually deleting the 
> bridges:
>
> [root@vhost2 images]# brctl show
> bridge name    bridge id        STP enabled    interfaces
> cloud0        8000.000000000000    no
> cloudbr0        8000.003048fc6bb4    no        eth0
> virbr0        8000.000000000000    yes
>
>
> [root@vhost2 images]# brctl delbr cloud0
> [root@vhost2 images]# brctl delbr cloudbr0
> [root@vhost2 images]# brctl show
>
>
> bridge name    bridge id        STP enabled    interfaces
> virbr0        8000.000000000000    yes
>
>
> [root@vhost2 images]# service network restart
>
> I know have a working web UI and the SSVM nodes started and all looks 
> good.
>
> My advice to anyone having trouble is:
>
> - Make sure you wipe everything before starting
> - set keepcache=1 in yum.conf  (so you dont keep downloading the same 
> packages over and over)
> - down load the image using wget and us the -f option to decompress it 
> /usr/share/cloudstack-common/scripts/storage/secondary/cloud-install-sys-tmplt
> - purge the SQL databases
> - check and remove the network interfaces - NOTE: I use two NIC's so 
> that if I kill the public IP I can connect from the next server in the 
> rack
>
> Hopefully this helps some else / feel free to correct or criticise 
> anything I have said in error
>
> Cheers
>
> Piers