Re: RADIUS fallback servers

From: Adam Brightwell <adam(dot)brightwell(at)crunchydata(dot)com>
To: Magnus Hagander <magnus(at)hagander(dot)net>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, Marko Tiikkaja <marko(at)joh(dot)to>
Subject: Re: RADIUS fallback servers
Date: 2017-03-06 15:24:41
Message-ID: CAE_9P=ibJWkswHupPrxs-v+CaQM8ZmPgaWcDEyODVaoCdPHqgQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

>> I wonder if removing the complexity of maintaining two separate lists
>> for the server and port would be a better/less complex approach. For
>> instance, why not go with a list of typical 'host:port' strings for
>> 'radiusservers'? If no port is specified, then simply use the default
>> for that specific host. Therefore, we would not have to worry about
>> keeping the two lists in sync. Thoughts?
>
>
> If we do that we should do it for all the parameters, no? So not just
> host:port, but something like host:port:secret:identifier? Mixing the two
> ways of doing it would be quite confusing I think.
>
> And I wonder if that format wouldn't get even more confusing if you for
> example want to use default ports, but non-default secrets.

Yes, I agree. Such a format would be more confusing and I certainly
wouldn't be in favor of it.

> I can see how it would probably be easier in some of the simple cases, but I
> wonder if it wouldn't make it worse in a lot of other cases.

Ultimately, I think that it would be better off in a separate
configuration file. Something to the effect of each line representing
a server, something like:

'<server> <port> <secret> <identifier>'

With 'radiusservers' simply being the path to that file and
'radiusserver', etc. would remain as is. Where only one or the other
could be provided, but not both. Though, that's perhaps would be
beyond the scope of this patch.

At any rate, I'm going to continue moving forward with testing this patch as is.

-Adam

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Jan Michálek 2017-03-06 15:39:13 Re: Other formats in pset like markdown, rst, mediawiki
Previous Message Aleksander Alekseev 2017-03-06 15:22:17 Re: Declarative partitioning optimization for large amount of partitions