Hi,
On Wed, Nov 10, 2010 at 03:01:21PM -0700, John Neiberger wrote:
I'm just curious to hear your thoughts on OIR on this platform. Is
this something that you prefer to avoid? Do you have any OIR-related
horror stories you'd like to share?
On 6500/7600 (and 7200), we *never* had any issues.
Thanks for your answer,
This config nat all no ? if the 172.16.1.x want access to 172.16.2.xx,
it's not natted ?
only to the destination 172.16.10.x ?
(i am search to nat only to the 172.16.10 destination
thanks
Stephane
2010/11/10 Ziv Leyes z...@gilat.net:
You could use a 255 addresses
:- John == John Neiberger jneiber...@gmail.com writes:
I ran into a problem with an OIR last night on a 7609. I normally
don't like to do them. I usually prefer to power the router down
first, replace/add the card and then power it back up. It caused all
sorts of fun when it
Hi Stephen,
First, define the ip nat inside and ip nat outside on the interface, and the
implement the following command. Hope this helps. Thanks.
ip nat inside source static network 172.16.1.0 172.16.10.0 /24 no-alias
regards,
YapCH
http://itcertguides.blogspot.com/
-Original
On Thu, 11 Nov 2010, Gert Doering wrote:
On 6500/7600 (and 7200), we *never* had any issues.
We've had a few mishaps. Field engineers don't know exactly how to insert
the card properly so the bus stalls for a prolonged period of time
(remember that *every* time you insert or remove a blade
- Gert Doering g...@greenie.muc.de wrote:
Not that IOS for Sup720 would run on non-Cisco hardware, so pirating
doesn't really make *that* much sense...
But there *could* be someone out there downloading new IOS who doesn't have a
support contract! That's *literally* stealing food from
Hi,
On Thu, Nov 11, 2010 at 10:50:40AM +, Tim Franklin wrote:
they're not selling as many support contracts as they think they
should, and have introduced the New Improved Download Experience
and the IOS 15 nodelocked licence clusterfuck.
Welcome to the future...
15.0S, anyone? :-)
On 11/11/2010 04:44, Richard A Steenbergen wrote:
5 minutes? What boxes are YOU rebooting? :)
mmm, actually yeah, last reboot took 7m30s before link up - and that was
layer 2 only, not layer 3.
Once upon a time, a sup720 would reboot in less than 4 minutes, sigh.
Nick
Hi,
my experience is not sticky concerning only OIR, but may be interresting for
someone.
We had a 'new' (used, actually) 7606E, with Sup720-3CXL, X6724-SFP, and
X6704-10GE.
After boot, one of PSU's failed and X6704-10GE refused to boot because of
low power. Both PSU's were 2700W.
This was
But there *could* be someone out there downloading new IOS who doesn't
have a support contract! That's *literally* stealing food from the
mouths of Cisco coders!
In the same way as the music, movie and software industries decide
that
they're not selling as much as they think they should,
If you wanted you could break the port channel and run equal cost multipath
bgp... then you run bfd over each link..
I agree though. Hold timer is easier and will probably suit you fine..
On Nov 11, 2010 12:10 AM, John Elliot johnellio...@hotmail.com wrote:
I would have something like this
On 11/11/2010 10:50, Tim Franklin wrote:
But there *could* be someone out there downloading new IOS who doesn't
have a support contract! That's *literally* stealing food from the
mouths of Cisco coders!
There are legitimate ways of getting upgrades without a service contract in
place, the
Always, huh?
So it's 100% not possible ever for one or more memory locations to get
corrupted (due to a bad memory chip), and put an invalid pointer in an array,
and the IOS uses that pointer to access a memory area it's not supposed/allowed
to?
I'm not sure what perfect world you live in,
Maybe IPSLA?
On Thu, Nov 11, 2010 at 08:40, Chris Evans chrisccnpsp...@gmail.com wrote:
If you wanted you could break the port channel and run equal cost multipath
bgp... then you run bfd over each link..
I agree though. Hold timer is easier and will probably suit you fine..
On Nov 11, 2010
I'll second Gert - I've personally performed close to 100 OIRs on a
variety of 6500 chassis, and never had it cause a problem.
There was a previous thread almost exactly like this, BTW - if you
feel like searching the archive. It was half-filled with OIR always
fails, I call it Online Insert and
What's the time length on the bus stall? Working on re working lots of
timers, hadn't thought of this. Something to add to the tests.
--chip
On Thu, Nov 11, 2010 at 9:55 AM, Geoffrey Pendery ge...@pendery.net wrote:
I'll second Gert - I've personally performed close to 100 OIRs on a
variety
On Thu, 11 Nov 2010, chip wrote:
What's the time length on the bus stall? Working on re working lots of
timers, hadn't thought of this. Something to add to the tests.
The bus is stalled all the time during the insertion. There is a few
millimeters of insertion length where the bus is
On Thu, Nov 11, 2010 at 8:16 AM, chip chip.g...@gmail.com wrote:
What's the time length on the bus stall? Working on re working lots of
timers, hadn't thought of this. Something to add to the tests.
--chip
In our case, I powered the card down, replaced it, then powered it
back up via the
It's not deterministic as it starts when first longest pin touches backplane
and ends when shortest pin connects. As a practical matter assume 100ms on the
low side and reboot on the high side. :)
Most protocol timers will be long enough that the low side is not a concern
exceptions being BFD
Mikael Abrahamsson wrote:
The bus is stalled all the time during the insertion. There is a few
millimeters of insertion length where the bus is stalled. If you're
rapid and firm in the insertion, you get a few tens of milliseconds of
stall. If you do it wrong and the car gets stuck in that
On Thu, Nov 11, 2010 at 10:58 AM, Kevin Loch kl...@kl.net wrote:
Mikael Abrahamsson wrote:
The bus is stalled all the time during the insertion. There is a few
millimeters of insertion length where the bus is stalled. If you're rapid
and firm in the insertion, you get a few tens of
If you were coming from SXF train and doing only a few basic tasks:
* Full BGP tables from several upstreams
* Several hundred SVI's (with counters)
* Basic netflow
1. Which would you prefer and why?
2. Any gotcha's when going from SXF - SXI?
Thanks in advance!
--
Randy
On 11/11/10 16:18, Randy McAnally wrote:
If you were coming from SXF train and doing only a few basic tasks:
* Full BGP tables from several upstreams
* Several hundred SVI's (with counters)
* Basic netflow
1. Which would you prefer and why?
SXI5; you're making such a big jump anyway that you
I have to configure a 4510R-E w/ dual 6E Sup's for a customer that already has
one and is looking for redundancy. I would like to fully populate it with the
WS-X4648-RJ45V-E blades but the Cisco configurator says that's not possible.
The WS-X4548-GB-RJ45V have an 8:1 oversubscription which I
I am getting this error when trying to bring up the second link on a
multilink ppp
Nov 11 16:55:35.110: Vi3 MLP: Request add link to bundle
Nov 11 16:55:35.110: Vi3 MLP: Adding link to bundle
Nov 11 16:55:35.110: Vi3 MLP: Computed frag size 59992 exceeds configured
value, changed to
2. Any gotcha's when going from SXF - SXI?
Can you be more specific?
I'm thinking boot code (getting stuck in rommon for some weird reason) and
config format changes -- for example, I don't want to be caught with my pants
down with half my SVI's missing after the reboot, ect.
Some other
Yes the LC bus is isolated but in async mode the BFD packet must still be sent
to the RP CPU and all BFD packets are generated from the RP CPU. 6500/7600 do
not support distributed BFD like CRS and GSR where LC CPU handles BFD.
I assume that you were considering scenario where BFD in echo mode
I think the issue is more complicated then just does it work or not. It is
dependent on how you have your 6500/7600 deployed.
Some form of bus-stalls will occur with any OIR. They may be minor, they may
not and that comes down to how long it takes for the shared bus to re-stabilize
because
According to
http://www.cisco.com/en/US/prod/collateral/modules/ps2710/ps5494/product
_data_sheet0900aecd802109ea_ps4324_Products_Data_Sheet.html
It looks like 8-10 do not support E series line cards with a Sup6-E. So
scratch slots 8-10 as well as your supervisor slots in your
configuration -
As far as I remember 4610R+E chassis (the new1, which supports up to
48gb/slot) supports 46xx in all 8 slots
---
Nikita
2010/11/11 Mohlmaster, Jarod jarod.mohlmas...@redemtech.com
According to
http://www.cisco.com/en/US/prod/collateral/modules/ps2710/ps5494/product
On Thu, Nov 11, 2010 at 7:08 AM, Matlock, Kenneth L
matlo...@exempla.org wrote:
Always, huh?
So it's 100% not possible ever for one or more memory locations to get
corrupted (due to a bad memory chip), and put an invalid pointer in an array,
and the IOS uses that pointer to access a memory
Ken,
Rather than speculate, do you have an actual example of a crash that IOS
reported as 'SegV exception' that was caused by failed hardware?
People come here for answers, not speculation.
eninja
On Nov 11, 2010, at 6:08 AM, Matlock, Kenneth L matlo...@exempla.org wrote:
Always, huh?
This really just shot in the dark but you have CEF hash for ECMP and then
etherchannel hash for link selection. I wonder if it's a weird example of
polarization such that because CEF hash decided left now all traffic subject to
etherchannel hash for that port-channel will resolve to one link.
I think the point that myself and others were trying to put across was
rather than purely head towards it being a SW issue log a case with TAC and
get them to look into it.
On Thu, Nov 11, 2010 at 9:49 PM, Eninja eni...@gmail.com wrote:
Rather than speculate, do you have an actual example of a
Exactly!
The only way to determine 100% if it's a hardware problem or software problem
is to either go through TAC for a bug scrub, try a known good version for the
features you are using, or go through the code one line at a time and figure
out why the SegV happened.
I was just pointing
I really think Geoffrey is onto the true cause.I have never had a problem
when inserting cards confidently. It is also worth noting that while it does
stall the bus, DFC forwarded traffic is unaffected.We run 100% DFCs in all
of our 6500s and the only traffic I have seen dropped
Jarod,
This is exactly what I was looking for.
Thanks!
//LeBlanc
-Original Message-
From: Mohlmaster, Jarod [mailto:jarod.mohlmas...@redemtech.com]
Sent: Thursday, November 11, 2010 11:50 AM
To: Leblanc, Jason; cisco-nsp@puck.nether.net
Subject: RE: [c-nsp] 4510 Hardware Question
Ken, David,
Guess we are all on the same page afterall i.e Dominic should grab the
crashinfo files and contact his maintenance service provider TAC.
To close, if IOS crashes as a result of a 'SegV exception', the cause is a
software bug.
Unfortunately, any other diagnosis would be a
The customer already has a 4510R-E so I will not be able to mix and match.
Thanks,
//LeBlanc
From: Nikita Shirokov [mailto:ns.ha...@gmail.com]
Sent: Thursday, November 11, 2010 1:38 PM
To: Mohlmaster, Jarod
Cc: Leblanc, Jason; cisco-nsp@puck.nether.net
Subject: Re: [c-nsp] 4510 Hardware
Rather than speculate, do you have an actual example of a
crash that IOS reported as 'SegV exception' that was caused
by failed hardware?
Yes, I've had a failing VAM manifest itself with SegV crashes.
Eventually it died completely and wasn't recognized on boot;
once replaced, router was
Port channels are prone to uneven distribution. Assuming this isnt a bug The
sessions are probably being distributed evenly. The difference in throughput
could be one or two hi bw sessions. For example an entire company nat'ed to
one public IP. Per session hashing doesn't guarantee even
On Thursday, November 11, 2010 01:08:42 pm John Elliot
wrote:
Thanks everyone for the assistance - Upstream does
support bfd, but unfortunately I am running PortChan on
7200, which appears to not support bfd!(Bugger!)...so
looks like we will have to run with lower holdtime.
BFD on LACP
If I recall correctly this is only available on the bigger chassis' like the
4500/6500 and I assume newer models like the Nexus' (Nexii?)
On Fri, Nov 12, 2010 at 4:12 AM, Keegan Holley keegan.hol...@sungard.comwrote:
If it's supported you can configure tcp port as part of the hashing
43 matches
Mail list logo
