Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 11 Dec 2017 11:34:10 -0800
From:      "Chris H" <portmaster@BSDforge.com>
To:        <portmaster@BSDforge.com>
Cc:        <freebsd-ports@freebsd.org>, <sgk@troutmask.apl.washington.edu>, "Adam Weinberger" <adamw@adamw.org>, "Matt Smith" <matt.xtaz@gmail.com>
Subject:   Re: Procmail Vulnerabilities check
Message-ID:  <00d53b391814cf11575da1e873839ae7@udns.ultimatedns.net>
In-Reply-To: <32da0142ef01d545aff61de3a3946d62@udns.ultimatedns.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 11 Dec 2017 08:39:02 -0800 <portmaster@BSDforge=2Ecom> said

> On Mon, 11 Dec 2017 11:10:32 +0000 "Matt Smith" <matt=2Extaz@gmail=2Ecom> sai=
d
>=20
> > On Dec 10 14:58, Chris H wrote:
> >>OK I'm puzzled a bit=2E FreeBSD' motto has always been:
> >>FreeBSD
> >>The power to serve!
> > >
> >>but many of the proposed, and recent changes/removals end up more like:
> >>FreeBSD
> >>I's castrated!
> >=20
> > The problem with software in the base is that it is *much* more=20
> > difficult to update to add new features or patch security issues=2E With =
a=20
> > port the software will be updated relatively quickly=2E And users can get=
=20
> > the benefits of that with a quick pkg upgrade=2E They might not update=20
> > their O/S for 6-12 months=2E
> >=20
> > In my opinion any software which is accessible to the internet should b=
e=20
> > patched and upgraded ASAP=2E It's for this reason that I've always=20
> > disabled things like OpenSSH/OpenSSL/ntpd etc in the base and used port=
=20
> > versions instead=2E
> I applaud that attitude=2E I couldn't agree more=2E For that same reason, I
> (not unlike you) have always excluded software that history has proven
> to pose security risks ( WITHOUT_BIND=3Dtrue ) for example=2E The same can =
also
> *easily* be said of OpenSSL=2E

[ excessive "jag" removed=2E sorry ]

> threat=2E
> In closing, and more to the point regarding Sendmail; Sendmail has a near=
ly
> impeccable security record in at the last decade=2E It provides a *secure*,
> more powerful, and more flexible MX on the cheap=2E I see little reason to
> consider it an attack vector=2E Which makes *security*, and it's related
> maintenance a pretty poor argument, for it's removal=2E
>=20
> --Chris
Let me attempt to make my point another way (and stay closer to topic)=2E
A user is able to accomplish more from sendmail in base, than with any
other MX port in base alone=2E
Sendmail provides OOB:
block by topic/portion of topic
block forged MX
block dynamic host(s)
all with the addition of one stanza, and (in the case "topic") the
addition of TOPIC_FILE
it also provides for some other measures that trip up, or otherwise
thwart spammer tactics;
delay (E)HELO
connection THROTTLING=2E As well as the ability to utilize block
list services, offered by third parties, or your own personal block
list=2E
Many of the other MX software in the ports tree provide a subset of
the shortlist I mentioned above=2E But none of them offer them all=2E
Given that the biggest concern, both security-wise, as well
nuisance-wise from anyone managing an internet facing MX service is
SPAM, and related threats=2E Wouldn't one be best served, if they had
the most options available to defend against such threats?
FWIW in ~5 months only having (ever) having sendmail from base,
without the addition of any additional "plugins"=2E
I was able to collect (and subsequently block)  ~9=2E9 million SPAM
sources=2E Not likely, but *actual* spam sources=2E
When I began life as a maintainer of ports=2E I was subsequently
required to subscribe to additional FreeBSD mailing lists, and
provide my/a email address along with the the ports I maintain=2E
As a result, my [that] address had a greater exposure to spammers=2E
In a short time, I found myself inundated with SPAM -- literally
*thousands* per day=2E My initial reaction was to curse the FreeBSD
ports/mailing-list management system, and those who were in charge=2E
But I decided against a knee-jerk reaction, and decided to give the
matter more thought, before making a decision=2E In the end, I decided
I wasn't going to allow myself to be a victim, but rather make the
whole matter a challenge, or puzzle that I would solve=2E In the end,
and with my current base version of sendmail, I now only receive
some 3-5 SPAM/week=2E That is a *remarkable* number, compared to my
initial experience, as the number of *actual* SPAM sources I've
been able to thwart=2E That 9=2E9 million number is not a *probable*
number, it's an actual figure, and in the end, I *always* get the
mail I want, and nearly *never* get the mail I don't=2E All with
sendmail from base, and without any external/third party services=2E
IMHO that makes a pretty strong argument for retaining Sendmail=2E If
I were an MX administrator=2E Would I not want all the options/help
I could get to defend myself against attack? This is the sort of
thing that makes FreeBSD the best choice for a Server Grade install=2E
It provides server grade applications in a Service oriented OS=2E

Yes=2E But if it's removed=2E Nothing stops you from installing it
from ports=2E

True=2E But if I'm selling a Server targeted OS=2E Don't I want to
advocate server grade services?

Thanks for listening -- I know it was long=2E

--Chris

> >=20
> > --=20
> > Matt
>=20
>=20
> _______________________________________________
> freebsd-ports@freebsd=2Eorg mailing list
> https://lists=2Efreebsd=2Eorg/mailman/listinfo/freebsd-ports
> To unsubscribe, send any mail to "freebsd-ports-unsubscribe@freebsd=2Eorg"





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?00d53b391814cf11575da1e873839ae7>