container-networking-method default doesn't allow trusty to start containers on aws cloud

I am having this issue , not sure what to do , the solution solved in the first moment to change the container networking , in the first deployment is solved, but int he second total new deployment not working. any glue?

Are you sating that you add a new model and the container-networking-method is not "local" as wanted?
There's a juju model-defaults command which allows you to set up default config for any newly created model. You can bootstrap with these default values, or you can set them after bootstrap, eg

$ juju model-defaults container-networking-method=local

then any new model will get that value.

See juju help model-defaults

When I used local in the first deployment it was working yes,,
it the second ... no and third ,no, I am not sure what to do ,

also some times 1 machine goes down and then you have issues to stand up
the juju client shows down , the juju real gui shows up, , this is very strange.
i used local and amazon got lost , I am not fully sure what to do ,

Some bundles shows like :
Failed to run: unsquashfs -f-d /var/lib/lxd/storage-pools/default/containers/jc573d0- XXXXXXXX
when tries to stand up specific containers..

in a secondary deployment isolated:

in juju shows all machines are up , was up but some with connectivity issues,

and eve JUJU GUI shows all machines are up,...
the JUJU client in cli shows:

that does not match
Machine State DNS Inst id Series AZ Message
9 down 3.19.30.8 i- bionic us-east-2c running
9/lxd/0 down 10.60.148.164 juju- -9-lxd-0 bionic us-east-2c Container started
9/lxd/1 down 10.60.148.33 juju-c -9-lxd-1 bionic us-east-2c Container started
9/lxd/2 down 10.60.148.187 juju-c 0-9-lxd-2 bionic us-east-2c Container started
9/lxd/3 down 10.60.148.224 juju-c5 -9-lxd-3 bionic us-east-2c Container started
10 down 18.189.178.213 i-0 e 0 bionic us-east-2a running
10/lxd/0 down 10.210.146.115 juju- -10-lxd-0 bionic us-east-2a Container started
10/lxd/1 down 10.210.146.90 juju -10-lxd-1 bionic us-east-2a Container started
10/lxd/2 down 10.210.146.46 juju 0-lxd-2 bionic us-east-2a Container started
11 down 3.15.24.147 i- 0c bionic us-east-2b running
11/lxd/0 down 10.160.119.170 juju-c d0-11-lxd-0 bionic us-east-2b Container started
11/lxd/1 down 10.160.119.98 juju-c 0-11-lxd-1 bionic us-east-2b Container started
11/lxd/2 down 10.160.119.119 juju-c -11-lxd-2 bionic us-east-2b Container started
12 down 18.222.18.179 i-01 f bionic us-east-2a running
12/lxd/0 down 10.51.106.79 juju- -12-lxd-0 bionic us-east-2a Container started
12/lxd/1 down 10.51.106.167 juju- -12-lxd-1 bionic us-east-2a Container started
12/lxd/2 down 10.51.106.37 juju- 12-lxd-2 bionic us-east-2a Container started

admin@maas:~$

Looking a bit further, using fan on AWS is necessary for container networking, so "local" is not something that's guaranteed to work.

There's some doc here on container networking, and there's a section at the end of that which talks about requirements on AWS. In particular, an account with VPC is required or else it won't work.

https://discourse.juju.is/t/fan-container-networking/1065

This bug has not been updated in 2 years, so we're marking it Low importance. If you believe this is incorrect, please update the importance.