Could NAT be an option?
Jean-Christophe Varaillon
------------
ALTEC Telecoms - NOC
14, Patmou, 151 23 Maroussi, Greece
Tel: +30 210 6872932
Fax: +30 210 6872904
E-mail: vajc(a)altectelecoms.gr
ICQ: 264-755-242
-----Original Message-----
From: swinog-bounces(a)lists.swinog.ch [mailto:swinog-bounces@lists.swinog.ch] On Behalf Of Steven.Glogger(a)swisscom.com
Sent: Thursday, March 22, 2007 12:12 AM
To: swinog(a)swinog.ch
Subject: [swinog] MPLS VRF source routing (inter-vrf routing)
hi all
i've got some nice cisco bugs / features / whatever.
some prerequisites:
- 2 VRFs: vrf blue and vrf red
- both vrfs have a different default route.
- a PPP session / user terminating in vrf blue
a specific route (10.0.1.0/29) is routed over static route (e.g. radius
avpair) over the ppp session (vrf blue).
this route is imported to vrf red by importing rd values and route-map filtering.
so the connectivity from the red vrf to the vrf blue is working (one way).
so, the goal (and this is the problem) is traffic souring that specific route should go back to vrf red.
how i thought would be the simplest way to do it: policy routing.
interface virtual-access123
ip policy route-map set-vrf-red
...
!
access-list 110 permit 10.0.1.0 0.0.0.7 any
route-map set-vrf-red permit 10
match ip address 110
set vrf red
!
would be the nicest way of doing this.
now the but: if you put the policy on the virtual-template / radius profile the session starts flapping (connect/disconnect/connect/disconnect....). so: not usable.
my other approach was:
interconnect vrf blue with vrf red by a vlan / interface.
assume on vrf blue: fastethernet0/0 with 11.0.0.1/30 connnected to vrf red with fastethernet0/1 with 11.0.0.2/30.
modifying the route map to:
route-map set-vrf-red permit 10
match ip address 110
set interface fastethernet0/0
set ip next-hop 11.0.0.2
!
this will stop the flapping (disconnect/connect/disconnect...) of the ppp session and the whole routing works as expected.... BUT: somewhen it stops working because of one thousand possible CEF bugs ;-(
i have to put "no ip route-cache cef" on the interconnection interface, then it works. some hours later (as already said) it stops working. when i do again "no ip route-cache cef" on the interface it works some other hours.
i've tried several IOS for the C7200series and the only half-way working version is the 12.4T (or even 12.3T).
so, now the big question to the community:
1) do you see any other working way doing source-routing from one vrf to another vrf?
(there's a vrf source routing command, but i think this will really not
scale)
2) do you have encountered the same CEF bugs? (i have seen them on 7206,
1841 and 2851 series routers)
how cisco tells me to do it:
http://www.cisco.com/en/US/products/sw/iosswrel/ps1838/products_feature_
guide09186a0080296409.html
i would be glad to get some input from you guys.
greetings
-steven
_______________________________________________
swinog mailing list
swinog(a)lists.swinog.ch
http://lists.swinog.ch/cgi-bin/mailman/listinfo/swinog
Disclaimer
The information in this e-mail and any attachments is confidential. It is intended solely for the attention and use of the named addressee(s). If you are not the intended recipient, or person responsible for delivering this information to the intended recipient, please notify the sender immediately. Unless you are the intended recipient or his/her representative you are not authorized to, and must not, read, copy, distribute, use or retain this message or any part of it. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses.
Matthias Leisi wrote:
> 194.67.23.0/24 does not equal the full set of *.mail.ru hostnames.
>
> Similarly, dnswl.org contains three /24s for uol.com.br (see
> http://www.dnswl.org/search.pl?s=3633). Now this is not a statement
> that uol.com.br is all nice and cosy, but it's a statement of the fact
> that the postmaster for uol.com.br told us that these are the ranges
> for the mailservers (and we verified that using eg senderbase.org).
>
> Since such ranges are usually not as trustworthy as /32s of
> well-respected mailserver operators, dnswl.org lists such ranges with
> a score of "none"; for all practical reasons, this should translate
> into "do not greylist, since there is most likely a legitimate
> mailserver at the other end who will retry anyway".
This sounds like a pretty good idea, but judging by the size of the
rbldnsd file, it's not very popular? Only 4317 entries.
/Per Jessen, Zürich
What about a "whitelist" website, every ISP could enter the IPs of his MTA's
and everybody could use for his whitelisting?
Radek Mrskos Email mrskos(a)volume.ch
volume.ch Tel: +41 43 534 40 24
Baechlerstr. 12 Mob: +41 79 219 68 66
CH-8802 Kilchberg Fax: +41 86079 2196 866
PGP:0x8CB69F6D URL: http://volume.ch
hi,
We use the email greylisting policy for incoming email for us and our
customers, and it appears that often the email from Sunrise network comes with
huge delays, from several hours to several days.
Did anyone else see such problems?
Guys from Sunrise, plase contact me directly for more details and
troubleshooting.
Such delays seen in email coming from
smtp-auth-be-01.sunrise.ch (mail-proxy-be-01.sunrise.ch [194.158.229.48])
smtp-auth-be-06.sunrise.ch (mx-auth.sunrise.ch [194.158.229.90])
cheers,
stan
044 732 9953
079 407 0224
--- "Kurt A. Schumacher" <Kurt.Schumacher(a)schumi.ch> wrote:
> Just a guess:
> The more retries to a certain SMTP destination, the longer the retry
> interval...leads to exponential delivery times.
no, it's quite sporadic: some emails are retried wihtin 10 minutes,
and some were delaying for hours.
> Pure greylisting is - just like blacklistsing - is useless or leads to
> unexpected behavior without a well-maintained and monitored whitelist for a
> corporate or ISP e-mail system.
it is actually a well-maintained whitelist, but of course sometimes
exceptions happen...
A customer is looking for a good (with experience and knowledge) lawyer
in CH concerning a .com domain dispute.
Any recommendations?
Thank you
Ralf Zenklusen
Thank you for the different suggestions I've received.
I'll pass them on...
Ralf Zenklusen
-----Ursprüngliche Nachricht-----
Von: Nik Hug [mailto:lists.nik@nts.ch]
Gesendet: Freitag, 23. März 2007 10:50
An: Ralf Zenklusen
Betreff: Re: [swinog] Lawyer recommendation
ralf.zenklusen(a)barinformatik.ch schrieb:
> A customer is looking for a good (with experience and knowledge)
lawyer
> in CH concerning a .com domain dispute.
>
> Any recommendations?
ev. Dr. Ursula Widmer ...
from widmerpartners-lawyers.ch
habe mal bei ihr studiert - nach mir war Sie auch zeitweise Antwalt von
Switch und hat nach mir verschiedene WIPO-Verfahren begleitet.
Gruss
Nik
>
> Thank you
> Ralf Zenklusen
>
>
>
> _______________________________________________
> swinog mailing list
> swinog(a)lists.swinog.ch
> http://lists.swinog.ch/cgi-bin/mailman/listinfo/swinog
hi all
i've got some nice cisco bugs / features / whatever.
some prerequisites:
- 2 VRFs: vrf blue and vrf red
- both vrfs have a different default route.
- a PPP session / user terminating in vrf blue
a specific route (10.0.1.0/29) is routed over static route (e.g. radius
avpair) over the ppp session (vrf blue).
this route is imported to vrf red by importing rd values and route-map
filtering.
so the connectivity from the red vrf to the vrf blue is working (one
way).
so, the goal (and this is the problem) is traffic souring that specific
route should go back to vrf red.
how i thought would be the simplest way to do it: policy routing.
interface virtual-access123
ip policy route-map set-vrf-red
...
!
access-list 110 permit 10.0.1.0 0.0.0.7 any
route-map set-vrf-red permit 10
match ip address 110
set vrf red
!
would be the nicest way of doing this.
now the but: if you put the policy on the virtual-template / radius
profile the session starts flapping
(connect/disconnect/connect/disconnect....). so: not usable.
my other approach was:
interconnect vrf blue with vrf red by a vlan / interface.
assume on vrf blue: fastethernet0/0 with 11.0.0.1/30 connnected to vrf
red with fastethernet0/1 with 11.0.0.2/30.
modifying the route map to:
route-map set-vrf-red permit 10
match ip address 110
set interface fastethernet0/0
set ip next-hop 11.0.0.2
!
this will stop the flapping (disconnect/connect/disconnect...) of the
ppp session and the whole routing works as expected.... BUT: somewhen it
stops working because of one thousand possible CEF bugs ;-(
i have to put "no ip route-cache cef" on the interconnection interface,
then it works. some hours later (as already said) it stops working. when
i do again "no ip route-cache cef" on the interface it works some other
hours.
i've tried several IOS for the C7200series and the only half-way working
version is the 12.4T (or even 12.3T).
so, now the big question to the community:
1) do you see any other working way doing source-routing from one vrf to
another vrf?
(there's a vrf source routing command, but i think this will really not
scale)
2) do you have encountered the same CEF bugs? (i have seen them on 7206,
1841 and 2851 series routers)
how cisco tells me to do it:
http://www.cisco.com/en/US/products/sw/iosswrel/ps1838/products_feature_
guide09186a0080296409.html
i would be glad to get some input from you guys.
greetings
-steven
The price seems to be a bit far out when considering that a normal
phone line (which includes the telephony infrastructure as well)
costs only CHF 25.25/Month. Reading the financial reports of Swisscom
from the last few years it doesn't seem Swisscom was losing any money
on the phone lines even when excluding any traffic minute charges.
http://www.swisscom.com/GHQ/content/Media/Medienmitteilungen/2007/20070320_…
Yet more work for BAKOM, Comcom and the Bundesgericht in the next
years.
--
Andre
