Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
On the Cisco gear are you configuring "mtu" or "ip mtu" on the interface? I'm not sure which one Observium displays, but I'm sure a quick experiment on a spare switchport will answer your question :) .
Jacob Bisby
-----Original Message----- From: observium [mailto:observium-bounces@observium.org] On Behalf Of Jim Glassford Sent: Tuesday, 25 April 2017 10:21 PM To: observium@observium.org Subject: [Observium] Cisco switches showing wrong Packet Size
Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
_______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
We have Cisco 3850s with jumbo frames (that show correctly in Observium).
By chance, have you restarted your switches since configuring "system mtu 9216"? It doesn't take effect until a reload, but will still show you the new values in running config etc.
Cheers,
Michael
On 26 Apr 2017, at 12:21 am, Jim Glassford jmglass@iup.edu wrote:
Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Michael and Jacob,
Thanks for the replies, appreciated. The Cisco 3850s displaying correctly gives me hope might be able to get this going. Michael, does the 3850 show the 9216 as the MTU size when you do a show interface? With Jumbo working here, my interfaces still show 1500. I will have to try a few more things on the bench see if I can get it going.
On the Cisco models I am seeing the problem displaying the wrong MTU: N3KC3172TQ10GT running 7.0(3)I2(2d) 2960X running 15.2(2)E5 2960S running 15.0(2)SE10 The N3K are setup for Jumbo MTU using QoS classes, cuts from Cisco documentation for the N3K/:// A per-physical Ethernet interface MTU is no supported. Instead, the MTU is set according to the QoS classes. When you show the interface settings, a default MTU of 1500 is displayed for physical Ethernet interfaces. / On our 3K models:/ /policy-map type network-qos jumbo// // class type network-qos class-default// // mtu 9216// //system qos// // service-policy type network-qos jumbo// // exit
//On our 29xx models: /Can raise the jumbo MTU to a maximum of 9198 bytes on 2960S and 2960X switches/ /Switch(config)# system mtu jumbo 9198/ /Changes to the system jumbo MTU will not take effect until the next reload is done/
/Note that it is necessary to reload the switch in order for the new MTU to take effect./ /Note that the MTU setting does not appear in the running or startup configuration;/ /this is important to remember when replicating configurations or replacing switches./ /Remember to verify that the new jumbo MTU has taken effect after a reload:// //Switch# show system mtu// //System MTU size is 1500 bytes// //System Jumbo MTU size is 9000 bytes// //System Alternate MTU size is 1500 bytes/ /Routing MTU size is 1500 bytes/
thanks! jim
On 4/26/2017 6:24 AM, Michael wrote:
We have Cisco 3850s with jumbo frames (that show correctly in Observium).
By chance, have you restarted your switches since configuring "system mtu 9216"? It doesn't take effect until a reload, but will still show you the new values in running config etc.
Cheers,
Michael
On 26 Apr 2017, at 12:21 am, Jim Glassford jmglass@iup.edu wrote:
Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
On a 3850 that is running jumbos correctly, a sh int does display the increased MTU (9198 in our case). However, it uses a global paramter to set it, not per interface (or policy-maps). The 3850 is quite a bit different platform to the 2960 and other 3Ks.
Can you create a span port on your switch of the ingress port & another of the egress port to compare if jumbo frames are actually traversing the switch?
I think I saw a 2960 and a 3750 near my desk today, I will see if I can get some time tomorrow to fire them up and see what they say...
Cheers,
Michael
On 26 Apr 2017, at 10:05 pm, Jim Glassford jmglass@iup.edu wrote:
Michael and Jacob,
Thanks for the replies, appreciated. The Cisco 3850s displaying correctly gives me hope might be able to get this going. Michael, does the 3850 show the 9216 as the MTU size when you do a show interface? With Jumbo working here, my interfaces still show 1500. I will have to try a few more things on the bench see if I can get it going.
On the Cisco models I am seeing the problem displaying the wrong MTU: N3KC3172TQ10GT running 7.0(3)I2(2d) 2960X running 15.2(2)E5 2960S running 15.0(2)SE10 The N3K are setup for Jumbo MTU using QoS classes, cuts from Cisco documentation for the N3K: A per-physical Ethernet interface MTU is no supported. Instead, the MTU is set according to the QoS classes. When you show the interface settings, a default MTU of 1500 is displayed for physical Ethernet interfaces.
On our 3K models: policy-map type network-qos jumbo class type network-qos class-default mtu 9216 system qos service-policy type network-qos jumbo exit
On our 29xx models: Can raise the jumbo MTU to a maximum of 9198 bytes on 2960S and 2960X switches Switch(config)# system mtu jumbo 9198 Changes to the system jumbo MTU will not take effect until the next reload is done
Note that it is necessary to reload the switch in order for the new MTU to take effect. Note that the MTU setting does not appear in the running or startup configuration; this is important to remember when replicating configurations or replacing switches. Remember to verify that the new jumbo MTU has taken effect after a reload: Switch# show system mtu System MTU size is 1500 bytes System Jumbo MTU size is 9000 bytes System Alternate MTU size is 1500 bytes Routing MTU size is 1500 bytes
thanks! jim
On 4/26/2017 6:24 AM, Michael wrote:
We have Cisco 3850s with jumbo frames (that show correctly in Observium).
By chance, have you restarted your switches since configuring "system mtu 9216"? It doesn't take effect until a reload, but will still show you the new values in running config etc.
Cheers,
Michael
On 26 Apr 2017, at 12:21 am, Jim Glassford jmglass@iup.edu wrote:
Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
observium mailing list
observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list
observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi Michael,
I have had the span ports up a few times for our systems and database engineers over the years. Nothing like seeing it on the wire to verify. Do see the Jumbo frames on the interfaces. Speaking of verifying, going over this again I now see the Cisco 2960 models with Jumbo enable ~are~ showing the correct set MTU of 9198 on Observium. ( jeez, was looking at the wrong 2960 :-[ )
I am just having the problem with the Cisco N3K models. One Interface info showed below from the N3K for e1/18.
strtsgf1s# *show inter e1/18* Ethernet1/18 is up admin state is up, Dedicated Interface Hardware: 100/1000/10000 Ethernet, address: e4aa.5de7.4fd9 (bia e4aa.5de7.4fd9) Description: dbsp02-vlan326 *MTU 1500 bytes*, BW 10000000 Kbit, DLY 10 usec reliability 255/255, txload 1/255, rxload 1/255 Encapsulation ARPA, medium is broadcast Port mode is trunk full-duplex, 10 Gb/s Beacon is turned off Auto-Negotiation is turned on Input flow-control is off, output flow-control is off Auto-mdix is turned off Switchport monitor is off EtherType is 0x8100 EEE (efficient-ethernet) : n/a Last link flapped 6d04h Last clearing of "show interface" counters never 70 interface resets 30 seconds input rate 356008 bits/sec, 41 packets/sec 30 seconds output rate 589608 bits/sec, 65 packets/sec Load-Interval #2: 5 minute (300 seconds) input rate 849.26 Kbps, 26 pps; output rate 870.97 Kbps, 42 pps RX 861583892 unicast packets 148 multicast packets 626 broadcast packets 861584666 input packets 458349625933 bytes *230468923 jumbo packets * 0 storm suppression packets 0 runts 0 giants 0 CRC 0 no buffer 0 input error 0 short frame 0 overrun 0 underrun 0 ignored 0 watchdog 0 bad etype drop 0 bad proto drop 0 if down drop 0 input with dribble 0 input discard 0 Rx pause TX 12430790740 unicast packets 6734525 multicast packets 2119210 broadcast packets 12439644475 output packets 18512656967308 bytes *12031963701 jumbo packets* 0 output error 0 collision 0 deferred 0 late collision 0 lost carrier 0 no carrier 0 babble 0 output discard 0 Tx pause
thanks! jim
On 4/26/2017 9:04 AM, Michael wrote:
On a 3850 that is running jumbos correctly, a sh int does display the increased MTU (9198 in our case). However, it uses a global paramter to set it, not per interface (or policy-maps). The 3850 is quite a bit different platform to the 2960 and other 3Ks.
Can you create a span port on your switch of the ingress port & another of the egress port to compare if jumbo frames are actually traversing the switch?
I think I saw a 2960 and a 3750 near my desk today, I will see if I can get some time tomorrow to fire them up and see what they say...
Cheers,
Michael
On 26 Apr 2017, at 10:05 pm, Jim Glassford jmglass@iup.edu wrote:
Michael and Jacob,
Thanks for the replies, appreciated. The Cisco 3850s displaying correctly gives me hope might be able to get this going. Michael, does the 3850 show the 9216 as the MTU size when you do a show interface? With Jumbo working here, my interfaces still show 1500. I will have to try a few more things on the bench see if I can get it going.
On the Cisco models I am seeing the problem displaying the wrong MTU: N3KC3172TQ10GT running 7.0(3)I2(2d) 2960X running 15.2(2)E5 2960S running 15.0(2)SE10 The N3K are setup for Jumbo MTU using QoS classes, cuts from Cisco documentation for the N3K: A per-physical Ethernet interface MTU is no supported. Instead, the MTU is set according to the QoS classes. When you show the interface settings, a default MTU of 1500 is displayed for physical Ethernet interfaces.
On our 3K models: policy-map type network-qos jumbo class type network-qos class-default mtu 9216 system qos service-policy type network-qos jumbo exit
On our 29xx models: Can raise the jumbo MTU to a maximum of 9198 bytes on 2960S and 2960X switches Switch(config)# system mtu jumbo 9198 Changes to the system jumbo MTU will not take effect until the next reload is done
Note that it is necessary to reload the switch in order for the new MTU to take effect. Note that the MTU setting does not appear in the running or startup configuration; this is important to remember when replicating configurations or replacing switches. Remember to verify that the new jumbo MTU has taken effect after a reload: Switch# show system mtu System MTU size is 1500 bytes System Jumbo MTU size is 9000 bytes System Alternate MTU size is 1500 bytes Routing MTU size is 1500 bytes
thanks! jim
On 4/26/2017 6:24 AM, Michael wrote:
We have Cisco 3850s with jumbo frames (that show correctly in Observium).
By chance, have you restarted your switches since configuring "system mtu 9216"? It doesn't take effect until a reload, but will still show you the new values in running config etc.
Cheers,
Michael
On 26 Apr 2017, at 12:21 am, Jim Glassford jmglass@iup.edu wrote:
Hi,
No luck trying to review the Observium Archives see if this was asked before, getting a Not Found off the web link.
Running Observium 17.3.8438 (stable), question on MTU size on Cisco interfaces. On our HP ProCurve switches with Jumbo enabled, all displays correctly, see MTU as 9216 on the interfaces as expected. On our Cisco models with Jumbo enabled, MTU shows 1500 on Average Packet Size graph. I believe this is due to Cisco interfaces showing the MTU as 1500 even when Jumbo frames are enabled and are passing Jumbo on the switch interfaces. Seeing this on both Cisco models 2960s and 3K the Average Packet Size found to be 1500. Any chance someone have a work around for this Cisco problem, able to set the MTU up to 9216 for the interfaces?
Appreciated! Jim
observium mailing list
observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list
observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Am Wed, 26.04.2017 um 08:05 schrieb Jim Glassford jmglass@iup.edu:
N3KC3172TQ10GT running 7.0(3)I2(2d)
I don't think there is a way to get the 'real' MTU of a n3k and n5k except for parsing the QoS-config. Neither the command line 'sh interf' nor a SNMP query tells you the correct value.
But anyway: it's not possible to change it per interface so once set it's set for all.
/Steffen
participants (4)
-
Jacob Bisby -
Jim Glassford -
Michael -
Steffen Klemer