Murder frontend problem

[Dan White]

On 06/05/15 16:44 +0200, Major Csaba wrote:
>Hi,
>
>Thanks for the quick answer.
>I managed to get further as I realized I missed a small piece from the 
>documentation. My fronted server and master update server is on the 
>same machine and I didn't configure the "mupdate_*" parameter. But as 
>I can see, the proxy still has to speak to mupdate when I would like 
>to create a new mailbox and the auth info is necessary even if they 
>are on the same host.
>
>So, it seems to be a misundersanding of the documentation which is not 
>so verbose :)
>I added the mupdate_* parameters (pointing to the host itself) and it 
>is working fine now.
>
>There is one more small question: why the proxied LMTP needs to have 
>"admins" permission on the backend? I thought the proxyservers setting 
>is for this, but LMTP doesn't work without adding my proxy user in the 
>"admins"...

On your backend, you should set 'lmtp_admins: murderproxy', rather than
specifying it as an admin, which limits its security impact.

With imap, the frontend proxy 'authenticates' as the user connecting to the
front end, which gains the permissions of the connecting user (on the
backend). E.g. you should see log entries on your backend with a successful
imap select which appears to be authenticating as the end user (e.g.
jsmith at domain.com).

lmtp may not proxy authenticate at all. If it does, you could specify
*that* user (e.g., the 'mail' account on your frontend) in the backend's
lmtp_admin, but I'm not sure that gains you much security wise.

Referencing syslog on the backend is the best way to flesh this out.

-- 
Dan White