Re: Router Choice

From: isabel dias (no email)
Date: Mon Nov 17 2008 - 14:07:05 EST

  • Next message: Ross Vandegrift: "Re: Catalyst 6500 High Switch Proc"

    Raymondo .....

    I guess you are bringing everyone together on the achieving resilience and "efficient" Load Balancing just in case the one path is temporarly unavailable ...... :-)

    *****commit/rollback
    CISCO IOS
    save a configuration to the NVRAM with the copy running-config startup -config command
    http://www-europe.cisco.com/en/US/docs/switches/wan/mgx/mgx_8850/software/mgx_r3/rpm/rpm_r1.1/configuration/guide/appc.html

    JUNIPER JUNOS
    You just have to send you syslog messages to /var/log/messages to a central point of management and avoid /dev/null :-)
    The best practice would be to follow the same in all Cisco devices Catos or IOS based ones.

    And again that is why management tools have their relevance and have cron jobs in place to keep the latest changes -accounting included)
    but yes you are right we can always "grep" something out and find what has changed.

    The commit concept is an interesting one and bring us back to the way they have compiled and allowed us "users" to fiddle around w/ the OS. Some versions not mentioned here
    require the word "comit" to be typed after a stop/start of some services-PID. Let us say a gracefull reestart of the process ......

    And yes the "old batle" ethernet vs ATM interfaces ....gosh ATM just had a "credit crunch" for the past years and ETHERNET standard knocked it down big time! Yes, the argument still standands ....if we want to take this further to the QoS/reliability ....I bet a lot of consultants would love an payed argument on this :-)

    As far as I am aware there are still a few interopability issues going on w/ some vendors and Eth-IEEE P802.3ba 40Gb/s, specially w/ having features available such as 802.1q/vlans etc ...but the card is there and available for everyone to work on ....

    And the world is moving to the 100 Gb Eth .....and so does IPv4 to IPv6.

    .//ID

    --- On Mon, 11/17/08, Raymond Macharia <> wrote:

    > From: Raymond Macharia <>
    > Subject: Re: Router Choice
    > To:
    > Date: Monday, November 17, 2008, 7:20 PM
    > Hello,I appreciate all your feedback. I have also recieved
    > more research
    > material from independent research institutes that give the
    > products thumbs
    > up.
    >
    > Best regards
    >
    > Raymond
    >
    > On Fri, Nov 14, 2008 at 2:30 PM, Paul Wall
    > <> wrote:
    >
    > > Whoa, excessive use of "!"...this isn't
    > IOS ICMP output.
    > >
    > > For those of you who want to have a chuckle, grep the
    > word "exit" on
    > > any of these fine 7750/7450 router configurations.
    > Seeing a router
    > > configuration that contains 10,000+ instances of the
    > word "exit" makes
    > > me recall the fine book FINAL EXIT. Seems like a poor
    > mans version of
    > > nesting with { }'s in JUNOS.
    > >
    > > Some of my gripes on the Timetra (whens the last time
    > Alcatel built
    > > something themselves instead of acquire it?) box are
    > that it really is
    > > catered to installs where Alcatel is running the
    > design side of the
    > > network as well. The CLI is somewhat non-intuitive for
    > IOS, IOS-XR or
    > > JUNOS operations staff. Here are some examples:
    > >
    > > Here in 2008, why are people buying boxes that do not
    > support
    > > candidate configuration or commit/rollback? The only
    > thing you can
    > > "commit" on the box is routing policy
    > changes. I thought this was a
    > > service provider box?
    > >
    > > For years (this might not be the case anymore), any
    > time you attempted
    > > to use the short-form of the "show" command
    > by typing "sh", you
    > > received a syntax error. This is because there were
    > two commands that
    > > began with sh: show and shell. The problem is that the
    > shell command
    > > prompts you for a password that only Alcatel knows
    > (and won't share
    > > with any customers that I'm aware of). So, if your
    > own customers cant
    > > run the command, why give users a headache?
    > >
    > > Its a router, why do I have to do "show router
    > route" to see a routing
    > > table entry? For years, you also had to suffix the
    > command "exact" on
    > > the end of every command as well.
    > >
    > > Pricing wise...they're way above other boxes that
    > you can find
    > > elsewhere that can do the jobs you need. Both the
    > Cisco 7600 and the
    > > Juniper MX line both have a way better CLI and employ
    > a knowledgeable
    > > staff of seasoned former service provider engineers.
    > Alcatel seems to
    > > be comprised of failed router startup guys from
    > Caspian or Chiaro.
    > > Feature wise, they're behind the curve when it
    > comes to competing with
    > > Cisco and Juniper. I think this is also shown in how
    > they name their
    > > software releases as "Feature Groups"
    > (telco-speak, anyone?).
    > >
    > > The main thing I want to speak to is that this box is
    > not made for
    > > your clueful IP operator. Alcatel is very insistent
    > that the customer
    > > use their UNIX/Windows NMS (I believe they call the
    > SAM) to interface
    > > with the routers. Sorry but...that might fly in
    > telcoland where
    > > executives ooh and ahh over point-and-click network
    > management, but I
    > > think most operators are going to find it a tad bit
    > useless.
    > >
    > > Sure, they do have NSR, but so did Avici. Does NSR
    > make up for the
    > > lack of features, high pricing and being stuck at
    > 20Gbps per slot?
    > > Yes, they do have 40Gbps per slot on the way, but who
    > doesn't support
    > > 40Gbps per slot today?
    > >
    > > Why bother stepping back a few years in development
    > when if you want a
    > > solid P core box, Foundry MLX/XMR, Juniper MX, Cisco
    > 7600s and CRS-1's
    > > are ready now and at prices that really aren't all
    > that bad. Oh yeah,
    > > you wont scratch the hell out of your finger nails
    > when removing the
    > > compact flash on those boxes.
    > >
    > > Drive slow, pinging 10(!!!!).
    > >
    > > On Wed, Nov 12, 2008 at 10:31 AM, devang patel
    > <> wrote:
    > > > I guess they have good lab in Plano, TX also!!!I
    > worked on the same
    > > routers
    > > > for IPTV deployment and really they are best!!!
    > > >
    > > >
    > > > regards
    > > > Devang Patel
    > > >
    > > > On Wed, Nov 12, 2008 at 8:43 AM, Dan Snyder
    > <> wrote:
    > > >
    > > >> I think that the 7750SR routers are great and
    > you won't be let down. We
    > > >> used to have an all Cisco network and I was
    > skeptical at first but they
    > > have
    > > >> been great.
    > > >>
    > > >> As for nss and nsr when we tested this by
    > failing a cpm we saw less than
    > > 50
    > > >> ms of traffic loss. I would see if you could
    > go to either California or
    > > >> Canada to one of ALUs labs and have it
    > demonstrated for you.
    > > >>
    > > >> hth,
    > > >> Dan
    > > >>
    > > >>
    > > >>
    > > >> Sent from my iPhone
    > > >>
    > > >>
    > > >> On Nov 12, 2008, at 7:40 AM, "Raymond
    > Macharia" <>
    > > >> wrote:
    > > >>
    > > >> Hello fellow nanogers,
    > > >>> I am a long time user of Cisco gear and
    > currently evaluating an
    > > >>> alternative
    > > >>> for my network expansion. currently the
    > one that looks like it will be
    > > >>> able
    > > >>> to do the job iare Alcatel-Lucent
    > 7710/7750 service routers.
    > > >>> I am looking for real life experience of
    > those who have used it and
    > > what I
    > > >>> may need to watch out for (if anything) I
    > have seen in some of their
    > > >>> documentation features like Non-stop
    > Services (NSS) and Non-stop
    > > Routing
    > > >>> (NSR). are these features real world
    > deployable.
    > > >>> oh, just to add I want to use the routers
    > as P routers in my IP/MPLS
    > > core
    > > >>>
    > > >>> Regards
    > > >>> --
    > > >>> Raymond Macharia
    > > >>>
    > > >>
    > > >>
    > > >
    > >
    >
    >
    >
    > --
    > Raymond Macharia

          


  • Next message: Ross Vandegrift: "Re: Catalyst 6500 High Switch Proc"





    Hosted Email Solutions

    Invaluement Anti-Spam DNSBLs



    Powered By FreeBSD   Powered By FreeBSD