Hi!
Can I do upgrade from 3750 to 3750 Metro? If yes how?

Fri, 14, 2005 at 10:26 +0200, Asbjorn Hojmark - Lists wrote:
Can I do upgrade from 3750 to 3750 Metro? If yes how?

No.

The 3750ME has a small PXF router onboard, so it's simply not
the same hardware.

Is it official or unofficial information? i.e I didn't find about it
on cisco site

Thanks! And what about "PXF router onboard" ?

Fri, 14, 2005 at 19:03 +1000, Richard Gallagher wrote:
From:

The Enhanced Services ports support enhanced features such as
Hierarchical QoS and Traffic Shaping, intelligent 802.1Q tunneling, VLAN
translation, MPLS, and EoMPLS. These ports can serve as uplinks to metro
aggregation points, including the Cisco Catalyst 4500 and Catalyst 6500
series and the Cisco 7600 Series, and they provide greater intelligence
at the network edge.

The regular cat3750 does not have the same ES ports avialable to enable
these extra features.

HTH, Ric

Maxim Tuliuk wrote:
Fri, 14, 2005 at 10:26 +0200, Asbjorn Hojmark - Lists wrote:
Can I do upgrade from 3750 to 3750 Metro? If yes how?
>>No.
>>
>>The 3750ME has a small PXF router onboard, so it's simply not
>>the same hardware.
>
>Is it official or unofficial information? i.e I didn't find about it
>on cisco site

(2005-10-14 10:14 +0100), Ed Butler - RapidSwitch wrote:

To be clear on this, the 3750s are forwarding traffic 100% correctly with no
loss, however they're not dealing with traffic to themselves very well.
They're not stressed either, running ~10% CPU usage.

Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. cisco3750 81.6% 1227 0.4 0.5 0.3 2.7 0.3
2. linuxhost 0.0% 1227 0.2 0.2 0.1 16.2 1.7

Does anyone have any experience with a similar problem or situation, and can
advise how to deal with it?

3750 has quite agressive defences against the control-plane. You can give
it's control-plane 1.4Mpps of packets and your SPF/BGP will stay up and it will
forward packets normally.
So it might be that you have large amount of background traffic towards
the control-plane or then you might have packets stuck in hold-queue.

To investigate further please provide:
sh int | i Input queue: [^0]
show buffers old dump | in inputtime
sh ver | i uptime

Thanks,

Regards,

Ed Butler
RapidSwitch Ltd
DDI: 020 7106 0731

RapidSwitch Ltd, 5th Floor, Sovereign House, 227 Marsh Wall, London, E14 9SD

This email message is intended only for the addressee(s) and contains
information that may be confidential and/or copyright. If you are not the
intended recipient please notify the sender by reply email and immediately
delete this email. Use, disclosure or reproduction of this email by anyone
other than the intended recipient(s) is strictly prohibited. No
representation is made that this email or any attachments are free of
viruses. Virus scanning is recommended and is the responsibility of the
recipient.
Message
From: cisco-nsp-bounces (AT) puck (DOT) nether.net
[mailto:cisco-nsp-bounces (AT) puck (DOT) nether.net] Behalf Richard Gallagher
Sent: 14 2005 10:04
To: Maxim Tuliuk
Cc: Asbjorn Hojmark - Lists; cisco-nsp (AT) puck (DOT) nether.net
Subject: Re: [c-nsp] software upgrade 3750 to 3750 Metro

From:

186a00801eb820.html

The Enhanced Services ports support enhanced features such as Hierarchical
QoS and Traffic Shaping, intelligent 802.1Q tunneling, VLAN translation,
MPLS, and EoMPLS. These ports can serve as uplinks to metro aggregation
points, including the Cisco Catalyst 4500 and Catalyst 6500 series and the
Cisco 7600 Series, and they provide greater intelligence at the network
edge.

The regular cat3750 does not have the same ES ports avialable to enable
these extra features.

HTH, Ric

Maxim Tuliuk wrote:
Fri, 14, 2005 at 10:26 +0200, Asbjorn Hojmark - Lists wrote:
Can I do upgrade from 3750 to 3750 Metro? If yes how?
>No.
>>
>The 3750ME has a small PXF router onboard, so it's simply not the
>same hardware.

Is it official or unofficial information? i.e I didn't find about it
on cisco site

cisco-nsp mailing list cisco-nsp (AT) puck (DOT) nether.net

archive at

cisco-nsp mailing list cisco-nsp (AT) puck (DOT) nether.net

archive at

(2005-10-14 18:42 +0100), Ed Butler - RapidSwitch wrote:

c3750#sh int | i Input queue: [^0]
Input queue: 1/75/0/0 (size/max/drops/flushes); Total output drops: 0
Input queue: 1/75/0/0 (size/max/drops/flushes); Total output drops: 0

show buffers old dump | in inputtime
Gives no output

Good, there are no stuck packets.

c3750#sh int vlan 211 | i Input queue:
Input queue: 1/75/147760/0 (size/max/drops/flushes); Total output drops: 0

I am concerned by the significant number of drops it has had. These do not
seem to be incrementing in line with the 80% loss that 'should' be achieved.
In fact, they don't seem to increment at all whilst I sent 20k packets to
the interface of which 4k came back.

Then I think we need to establish if you're forwarding packets in
hardware or not. 'show sdm prefer' are you exceeding any of these
values? 'show proc cpu' do you have high I/ load? If both answers
are 'no' then you actually might want to see what the router is processing
in CPU 'debug ip packet <>', I find it
best to log to syslog rather than terminal monitor, but YMMV. Some caution
should be excericed with 'debug ip packet'.

Regards,

Ed Butler
RapidSwitch Ltd
DDI: 020 7106 0731

RapidSwitch Ltd, 5th Floor, Sovereign House, 227 Marsh Wall, London, E14 9SD

This email message is intended only for the addressee(s) and contains
information that may be confidential and/or copyright. If you are not the
intended recipient please notify the sender by reply email and immediately
delete this email. Use, disclosure or reproduction of this email by anyone
other than the intended recipient(s) is strictly prohibited. No
representation is made that this email or any attachments are free of
viruses. Virus scanning is recommended and is the responsibility of the
recipient.
Message
From: cisco-nsp-bounces (AT) puck (DOT) nether.net
[mailto:cisco-nsp-bounces (AT) puck (DOT) nether.net] Behalf Saku Ytti
Sent: 14 2005 11:04
To: cisco-nsp (AT) puck (DOT) nether.net
Subject: Re: [c-nsp] 3750G packet loss

(2005-10-14 10:14 +0100), Ed Butler - RapidSwitch wrote:

To be clear on this, the 3750s are forwarding traffic 100% correctly
with no loss, however they're not dealing with traffic to themselves very
well.
They're not stressed either, running ~10% CPU usage.

Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. cisco3750 81.6% 1227 0.4 0.5 0.3 2.7 0.3
2. linuxhost 0.0% 1227 0.2 0.2 0.1 16.2 1.7

Does anyone have any experience with a similar problem or situation,
and can advise how to deal with it?

3750 has quite agressive defences against the control-plane. You can give
it's control-plane 1.4Mpps of packets and your SPF/BGP will stay up and it
will forward packets normally.
So it might be that you have large amount of background traffic towards the
control-plane or then you might have packets stuck in hold-queue.

To investigate further please provide:
sh int | i Input queue: [^0]
show buffers old dump | in inputtime
sh ver | i uptime

Thanks,

Regards,

Ed Butler
RapidSwitch Ltd
DDI: 020 7106 0731

RapidSwitch Ltd, 5th Floor, Sovereign House, 227 Marsh Wall, London,
E14 9SD

This email message is intended only for the addressee(s) and contains
information that may be confidential and/or copyright. If you are not
the intended recipient please notify the sender by reply email and
immediately delete this email. Use, disclosure or reproduction of this
email by anyone other than the intended recipient(s) is strictly
prohibited. No representation is made that this email or any
attachments are free of viruses. Virus scanning is recommended and is
the responsibility of the recipient.
Message
From: cisco-nsp-bounces (AT) puck (DOT) nether.net
[mailto:cisco-nsp-bounces (AT) puck (DOT) nether.net] Behalf Richard
Gallagher
Sent: 14 2005 10:04
To: Maxim Tuliuk
Cc: Asbjorn Hojmark - Lists; cisco-nsp (AT) puck (DOT) nether.net
Subject: Re: [c-nsp] software upgrade 3750 to 3750 Metro

From:

heet09
186a00801eb820.html

The Enhanced Services ports support enhanced features such as
Hierarchical QoS and Traffic Shaping, intelligent 802.1Q tunneling,
VLAN translation, MPLS, and EoMPLS. These ports can serve as uplinks
to metro aggregation points, including the Cisco Catalyst 4500 and
Catalyst 6500 series and the Cisco 7600 Series, and they provide
greater intelligence at the network edge.

The regular cat3750 does not have the same ES ports avialable to
enable these extra features.

HTH, Ric

Maxim Tuliuk wrote:
Fri, 14, 2005 at 10:26 +0200, Asbjorn Hojmark - Lists wrote:
Can I do upgrade from 3750 to 3750 Metro? If yes how?
>No.
>>
>The 3750ME has a small PXF router onboard, so it's simply not the
>same hardware.

Is it official or unofficial information? i.e I didn't find about it
on cisco site

cisco-nsp mailing list cisco-nsp (AT) puck (DOT) nether.net

archive at

cisco-nsp mailing list cisco-nsp (AT) puck (DOT) nether.net

archive at

(2005-10-14 13:04 +0300), Saku Ytti wrote:

Packets Pings
Host Loss% Snt Last Avg Best Wrst StDev
1. cisco3750 81.6% 1227 0.4 0.5 0.3 2.7 0.3
2. linuxhost 0.0% 1227 0.2 0.2 0.1 16.2 1.7

Does anyone have any experience with a similar problem or situation, and can
advise how to deal with it?

3750 has quite agressive defences against the control-plane. You can give
it's control-plane 1.4Mpps of packets and your SPF/BGP will stay up and it will
forward packets normally.

Heh, after reading Steinar's post, I reread your email and noticed that
it was act actually 120pps (I assumed 1pps:). Yes, indeed it's normal for
3750 to aggressively rate-limit generating icmp, punting etc. And
you'll love it, when you get your first DoS attack against it's control
plane, it'll just keep working (unless the DoS attack happens to be SPF
or BGP, which you should drop in ACL. There was possibility for the 3750
team to program ACL based on bgp neighbor statements, but they wanted
to conserve TCAM space for users and didn't do this automaticly.
BGP was not considered priority traffic at all, only SPF).