Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
- Cisco Community
- :
- About Giuseppe Larosa
Online
Giuseppe Larosa
.jpg){kind=icon}
Hall of Fame Expert
Electronic Engineer in 1993
Since 1997 consultant in the networking for Telecom Italia Group and later for other service providers.
On July the 1st 2010 I have joined Cypress Consult as an Associate.
My new role is senior network consultant for a service provider customer.
I' m expected to provide technical support and knowledge transfer to customer's stuff.
My Specialties are:
In November 2012 I have joined Juniper and Cisco partner b! Spa part of CEIT group (owner Sandro Bordato)
ITIL Service Management Foundation v.3 since Nov 2008
Cisco CCIE Service Provider #14802 since June 23 2005
Cisco CCNP since Juanary 2002
Cisco CCNA since July 2000
Cisco CCNA Voice Enterprise Ed. since October 2009
Juniper JNCIS-M since December 2010
Juniper JNCIP-SP since November 16th 2012
Electronic Engineer in 1993
Since 1997 consultant in the networking for Telecom Italia Group and later for other service providers.
On July the 1st 2010 I have joined Cypress Consult as an Associate.
My new role is senior network consultant for a service provider customer.
I' m expected to provide technical support and knowledge transfer to customer's stu
Awards
Recent Badges
Stats
Member since
08-26-2003
15147
Posts
12217
Helpful
2250
Solutions
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-16-2019
12:56 PM
09-16-2019
12:56 PM
Hello,
check the effective IP MTU with
show ip interface vlan X
The system mtu 2248 may solve your ping issue between the C4510 in some platforms / IOS versions is needed to be able to use greater packets in the switch TCP/IP stack (the route processor)
Be aware that also L2 ports connecting to servers need an increased MTU to work ( end to end path between the two hosts).
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-16-2019
12:49 PM
09-16-2019
12:49 PM
Hello Paul,
post
show version
show version | inc image
As already noted you are likely running a software image that allows only to use IGMP snooping but not full multicast routing.
With these old images the image name tells the feature set ( it is part of the name).
Alternatively you may need to change the sdm template to another type (this requires reload to take effect) as you have a desktop SDM template. You may need to change it to another type more routing oriented ( I don't remember the exact name for your switch)
Hope to help
Giuseppe
... View more
09-16-2019
12:37 PM
Hello Frello,
there is no other way to do this.
Note:
if 2.2.2.2/32 was in area 0 , R1 would see it as an O route in area 0. It sees it as on O IA route. (inter area)
if 2.2.2.2/32 is moved to area 120 R1 would see it as an O route in area 120.
In your current setup R2 2.2.2.2/32 is not in area 0 and it is not in area 120.
Joseph's answer is correct.
When using a link state protocol like OSPF you are constrained by its rules.
Inter area updates and inter area traffic has to go via the backbone area.
I have learned by experience that distribute-list have only local effects : it can filter what routes are installed in the local routing table but it does not block LSA flooding.
Using distribute lists with OSPF when downstream OSPF routers are present can cause routing problems as they reiceve all the LSAs (this is required by OSPF rule that all routers in an area must have the same set of LSAs) and they can install the corresponding routes pointing to a device that is missing the route in its local routing table.
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-16-2019
12:20 PM
09-16-2019
12:20 PM
Hello ,
>> Any insight into this would be greatly appreciated. Do I need to enable large frames by some means?
Yes, however there are different ways to achieve this.
You can use a global command like
system mtu <size>
(Warning your OSPF adjacencies will flap so you need a maintenance window)
Or you can enable jumbo frames.
You need to ensure that all L2 interfaces in the end to end path are configured and support the increased MTU.
but, first of all you need to understand what MTU is needed by this application to work.
On the other hand you can find the involved hosts likely servers and check their MTU settings.
They have likely a non standard MTU like 9000 bytes and this can cause the issue more then the application itself.
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Switching
Giuseppe Larosa
in Switching
09-16-2019
12:11 PM
09-16-2019
12:11 PM
Hello,
as noted by Joseph the default duplex setting at 100 Mbps is half duplex not full duplex so the issues you have faced. You should have also seen late collisions as the switch side works in full duplex and the PC NIC in half duplex on the PC side.
I have seen in a customer network a better approach:
using in interface mode
speed auto 100
This allows to still negotiate with the other side and puts an upper limit to the usable speed.
I think the switches were a mix of C3650 and older C3560.
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Switching
Giuseppe Larosa
in Switching
09-16-2019
12:01 PM
09-16-2019
12:01 PM
Hello Ramesh,
see the following document
https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst4500/XE3-6-0E/15-22E/configuration/guide/xe-360-config/pbroute.pdf
set ip next-hop recursive should be supported on Catalyst 4500 as reported in the link above and I guess on Catalyst 6500, 6800.
The set ip next-hop recursive may require a TCAM able to support it.
It is not so easy as normal PBR, multiple pointers to the possible next-hops need to be installed in TCAM in the FIB, and then CEF load balancing must choice one of them for each flow ( given a source IP and a destination IP.
address that match the route-map block/clause).
I'm afraid Access switches are not expected to support a feature like this.
Hope to help
Giuseppe
... View more
09-15-2019
10:20 AM
Hello Prashanma,
the key to achieve scalability when the number of Vlans is greater then 128 is to move to spanning-tree mode mst as noted by Joseph.
With MST you can provision how the IEEE whole range of Vlans 1-4094.
(Please note that modern SVL uses Vlan 4094 on the SVL link) is mapped to few MST instances even before you create the Vlan.
Because MST consider two switches to be in the same MST region if they agree on the following parameters:
MST region name ( up to 32 characters is a string)
MST revision number
Vlans to MST mapping.
Actually only the IST instance 0 BPDUs are sent out with additional sections for each MST instance.
In the BPDUs an MD5 checksum of the Vlans to MST instances mapping is actually sent.
So MST requires a different approach: when you need to create a new Vlan you need to keep an unused Vlan from the subset of Vlans that are mapped to the MST instance that has the desired topology ( root bridge and so on).
So you first choice the topology the new Vlan should have and then you can pick a Vlan-id in the correct sub-set.
Changing the Vlans to MST mapping causes the MST region to be partitioned.
To be noted VTP version 3 has a feature that allows to carry the Vlans to MST mapping that is very interesting and handy.
This is likely the reason why modern access switches have not improved the number of RSTP instances
Hope to help
Giuseppe
... View more
09-15-2019
05:06 AM
Hello Siva,
your understanding is correct.
the minimum size is 64 bytes because an 802.1Q tagged frame is STILL a standard ethernet frame.
This was more put in evidence when 802.1Q was compared with Cisco proprietary ISL. (30 bytes overhead 26 bytes header and 4 byte CRC, an ISL frame is NOT a standard ethernet frame but carries an ethernet frame in its payload using a tunnel encapsulation).
ISL does not support multiple Vlan tagging (like it is used in QinQ or other forms like provider backbone bridging ).
So this question was easier to answer many years ago when both 802.1Q and ISL were studied and compared.
Note: compliments for you questions they are becoming a challenge to answer.
Hope to help
Giuseppe
... View more
09-13-2019
12:44 AM
Hello Siva,
I agree with Reza.
I had a case where after a blackout one C6800 chassis member of a VSS was powered off for many hours.
The VSS had a four tengiga bundle versus another C6800 VSS, when the chassis came back up some ports have gone in error disable for UDLD ( If I remember correctly).
we enabled automatic recovery on the VSS that was affected by the blackout ( it was enabled on the other VSS).
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-12-2019
03:58 AM
09-12-2019
03:58 AM
Hello Paul,
IPv4 multicast routing needs additional commands to work:
a) ip multicast-routing in global config this makes the device a mcast router
B) on each SVI add
int vlan 2
ip pim sparse-dense mode
exit
int vlan 3
ip pim sparse-dense mode
exit
Create a loopback interface to make the switch an RP
int loop0
ip address 10.255.255.1 255.255.255.255
ip pim sparse-mode
exit
ip pim rp 10.255.255.1
Note:
if you have multiple devices the loop0 has to be created only on one f them and it needs to be advertised in a unicast routing protocol like OSPF or EIGRP.
Enabling PIM on an SVI enables also IGMP (version 2 by default)
IGMP deals with receivers (PCs)
PIM is needed between multicast capable routers on all links between them
Hope to help
Giuseppe
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-12-2019
03:11 AM
09-12-2019
03:11 AM
Hello ehsanpoureshagh,
thanks for your feedback.
>> not sure why, because virtual interfaces were using same physical link, had diffrent mac address, but as soon as i assigned manual mac address, OSPF came up
I had to this in the past on on C6500 at SVI level when building an EIGRP neighborship between two different SVIs in different VRFs going trough a transparent FW contxt in FWSM blade. But in my case SVIs were trying to use the same MAC address so we needed to change it on one of them.
Best Regards
Giuseppe
... View more
Created by
Giuseppe Larosa
in Switching
Giuseppe Larosa
in Switching
09-11-2019
12:16 AM
09-11-2019
12:16 AM
Hello Siva,
you have performed very interesting tests. Thanks for having shared your results.
As I have explained in the other thread the command vlan dot1q tag native applies for sure to user traffic frames and to those protocol frames that can support tagging.
I have provided there an example taken from real world experience of an interoperability issue caused by mismatch on tagging / not tagging IEEE RSTP BPDUs.
Your tests show that Cisco implementation will honor the tagging for both IEEE STP BPDU and Cisco PVST BDPU if the native Vlan is not the default one and vlan dot1q tag native is enabled.
However, when you changed the native Vlan back to 1 you see the IEEE STP BPDUs untagged and also Cisco PVST BPDUs untagged. This behaviour is in line with what I have seen on switches of other vendors when the native Vlan is the default one the IEEE STP BPDU are sent untagged for backward compatibility (or because so it is stated on IEEE standard).
Hope to help
Giuseppe
... View more
09-10-2019
11:48 PM
Hello Siva,
I can tell you what happened to me in job:
two devices one Juniper switch and one IBM switch part of an IBM blade server were not able to interoperate in RSTP, because the Juniper switch was sending the RSTP BPDUs untagged and the IBM switch was sending them tagged.
As a result of this when we added a second link between the two devices a bridging loop occurred, we removed one cable and then we made a packet capture on Juniper side and we had seen the difference between sent untagged IEEE RSTP BPDU and received tagged with vlan 1 RSTP BPDU sent by other switch. Changing the configuration on the IBM side solved this issue.
For IEEE standards the IEEE STP BPDUs should travel untagged as far as I know, this is why we say that it follows the native Vlan settings.
If you enable the Vlan dot1q tag native this command applies for sure to user traffic in native Vlan, to Cisco PVST+ / Rapid PVST, but some layer2 protocols may still travel untagged if required by their standard specifications.
However , this is implementation dependent and should be tested, but as I tried to explain with an example above tagging IEEE STP BPDUs can cause interoperability issues.
The modern trend of many vendors is to allow by default untagged traffic only for control protocols and to do not allow user traffic to travel untagged.
That in Cisco terms should be equivalent to have Vlan dot1q tag native enabled by default.
Edit:
You can find the 802.1D year 2004 specification here for free but you need to create an account as individual to access the content.
https://ieeexplore.ieee.org/browse/standards/number/ieee?queryText=802.1D%202004
Edit 2:
>> If we enable "Vlan dot1q tag native" answer B is also possible right?
After seeing your lab test results on the other thread I agree that answer B is acceptable, if native is 99 but the command for native Vlan tagging is enabled IEEE STP can travel tagged in Vlan 99.
If the native Vlan is 1 the default one the IEEE STP BPDU are sent always untagged regardless of native vlan tagging settings.
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
Giuseppe Larosa
in Routing
09-10-2019
12:35 AM
09-10-2019
12:35 AM
Hello ehsanpoureshagh,
I have used OSPF in multiple contexts on ASA with SW version 9.2.6 with good results
>>
DC1P-NETASAPR010/cntx1# sho ospf neighbor
Neighbor ID Pri State Dead Time Address Interface 192.168.78.23 1 FULL/BDR 0:00:35 192.168.65.91 hw_cloud 192.168.78.100 1 FULL/DR 0:00:36 192.168.68.62 wan_inet
both interfaces are in the same context cntx1 from the device prompt.
Note:
for successful installation of redistributed static routes in OSPF the static route next-hops must be advertised as internal routes using network commands and eventually passive interface.
The devices will look at the forwarding address field in the LSA type 5, if set to a value different from 0.0.0.0 (that would mean OSPF RID of ASBR node) they look for an internal route.
On the C4500 you should use network commands in OSPF covering all the next-hops
>>
router ospf 21 vrf HW_CLOUDS router-id 192.168.78.23 redistribute static subnets network 192.168.78.23 0.0.0.0 area 0
network 192.168.65.88 0.0.0.7 area 0 ! ! ip route vrf HW_CLOUDS 10.140.0.0 255.255.0.0 192.168.192.10 ip route vrf HW_CLOUDS 10.140.0.0 255.255.0.0 192.168.192.14 5 ip route vrf HW_CLOUDS 10.150.0.0 255.255.0.0 192.168.192.2 ip route vrf HW_CLOUDS 10.150.0.0 255.255.0.0 192.168.192.6 5
I think you need a network statement to cover all the 192.168.192.x next-hops in vrf HW_CLOUDS.
This may be a general issue with OSPF not related to ASA or ASA with contexts.
Hope to help
Giuseppe
... View more
09-10-2019
12:10 AM
Hello Raj-toor,
yes in OSPF the interface command ip ospf priority 0 makes the device not eligible for DR or BDR role.
If all devices have this command on an SVI and the OSPF network type of broadcast (default ) is in use the devices will be stucked in two way state waiting for someone to become DR and BDR.
Removing that command solves the problem you can use an ip ospf priority 50 if you want to make a device the DR.
Cisco default values for OSPF priority is 1.
Hope to help
Giuseppe
... View more
Created by
Giuseppe Larosa
in Routing
09-16-2019
09-16-2019
Hello, check the effective IP MTU with show ip interface vlan X The
system mtu 2248 may solve your p...
09-16-2019
Hello Paul, post show version show version | inc image As already noted
you are likely running a sof...
09-16-2019
Hello Frello, there is no other way to do this. Note: if 2.2.2.2/32 was
in area 0 , R1 would see it ...
Created by
Giuseppe Larosa
in Routing
09-16-2019
09-16-2019
Hello , >> Any insight into this would be greatly appreciated. Do I need
to enable large frames by s...
Created by
Giuseppe Larosa
in Switching
09-16-2019
09-16-2019
Hello, as noted by Joseph the default duplex setting at 100 Mbps is half
duplex not full duplex so t...
Created by
Giuseppe Larosa
in Switching
09-16-2019
09-16-2019
Hello Ramesh, see the following document
https://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst...
09-15-2019
Hello Prashanma, the key to achieve scalability when the number of Vlans
is greater then 128 is to m...
09-15-2019
Hello Siva, your understanding is correct. the minimum size is 64 bytes
because an 802.1Q tagged fra...
09-13-2019
Hello Siva, I agree with Reza. I had a case where after a blackout one
C6800 chassis member of a VSS...
09-12-2019
Hello Paul, IPv4 multicast routing needs additional commands to work: a)
ip multicast-routing in glo...
Created by
Giuseppe Larosa
in Routing
09-12-2019
09-12-2019
Hello ehsanpoureshagh, thanks for your feedback. >> not sure why,
because virtual interfaces were us...
Created by
Giuseppe Larosa
in Switching
09-11-2019
09-11-2019
Hello Siva, you have performed very interesting tests. Thanks for having
shared your results. As I h...
09-10-2019
Hello Siva, I can tell you what happened to me in job: two devices one
Juniper switch and one IBM sw...
Created by
Giuseppe Larosa
in Routing
09-10-2019
09-10-2019
Hello ehsanpoureshagh, I have used OSPF in multiple contexts on ASA with
SW version 9.2.6 with good ...
Public Statistics
| Date Registered | 08-26-2003 02:42 AM |
| Date Last Visited | 09-16-2019 12:57 PM |
| Total Messages Posted | 15,147 |
| Total Helpful Votes Received | 12217 |
Helpful Votes Given To
| User | Helpful Count |
|---|---|
| 10 | |
| 104 | |
| 5 | |
| 5 | |
| 220 |
.jpg "VIP Expert"){kind=icon}
Helpful Votes From
Awards
Hall of Fame
Cisco Designated VIPs
2014 LAN WAN
Cisco Designated VIPs
2013 LAN WAN
Cisco Designated VIPs
2012 LAN
Cisco Designated VIPs
2011 LAN MPLS WAN
Community Spotlight Award
English Member's Choice Award, May 2019
Community Spotlight Award
English Community Questions Answered Award, May 2019
Community Spotlight Award
Member's Choice, June 2019
