09-04-2019 05:49 AM - edited 09-04-2019 05:52 AM
Hello, colleagues.
I've a task of sending WakeOnLan magic packets from one server through routed network to remote subnets.
Curent configuration uses ip helper-address statment for every target network on router interface, where WOL server is located.
Router is ASR1002 with ADVENTERPRISE software, Version 15.5(3)S7b.
Problem is - maximum of 16 ip helper-address statements is alloved on this model, but I need more than 60.
Question is: is there any way to extend maximum number of ip helper-address statments per interface, or any way of sending routed WOL pacets withot this statemnt at all.
Solved! Go to Solution.
09-04-2019 08:12 AM
I am not prepared to address the possible issue of how many helper addresses are supported vs how many helper addresses is advisable. But I do have a suggestion to address the original question of how to deal with a limited number of helper addresses. If the number of destination subnets is greater than the number of helper addresses supported/advised then I suggest these steps:
1) divide the destination subnets into groups based on how many helper addresses you are comfortable using. for example if you need 60 destination addresses and are comfortable with using 15 helper addresses you could use 4 groups. If you think you want 4 helper addresses (as suggested by @balaji.bandi ) then you need 15 groups.
2) identify intermediate subnets between the subnet where the WOL is originated and the destination subnets. If you have 15 groups you will need 15 intermediate subnets and if you have 4 groups you will need 4 subnets. These intermediate subnets might be on the same router/switch where WOL is originated or they might be on devices on the paths to the destination subnets.
3) in the interface of the subnet where WOL is originated configured helper addresses, one for each of the intermediate subnets.
4) in the interface of the intermediate subnets configure helper addresses, one for each of the destination subnets of that group.
HTH
Rick
09-04-2019 07:05 AM
Not what is the use case to have 60 here.
To be honestly even though it say support 16, after 6 ip helper address, we see TTL expires, 4 is best to do.
09-04-2019 07:47 AM
09-04-2019 08:12 AM
I am not prepared to address the possible issue of how many helper addresses are supported vs how many helper addresses is advisable. But I do have a suggestion to address the original question of how to deal with a limited number of helper addresses. If the number of destination subnets is greater than the number of helper addresses supported/advised then I suggest these steps:
1) divide the destination subnets into groups based on how many helper addresses you are comfortable using. for example if you need 60 destination addresses and are comfortable with using 15 helper addresses you could use 4 groups. If you think you want 4 helper addresses (as suggested by @balaji.bandi ) then you need 15 groups.
2) identify intermediate subnets between the subnet where the WOL is originated and the destination subnets. If you have 15 groups you will need 15 intermediate subnets and if you have 4 groups you will need 4 subnets. These intermediate subnets might be on the same router/switch where WOL is originated or they might be on devices on the paths to the destination subnets.
3) in the interface of the subnet where WOL is originated configured helper addresses, one for each of the intermediate subnets.
4) in the interface of the intermediate subnets configure helper addresses, one for each of the destination subnets of that group.
HTH
Rick
09-05-2019 12:31 AM - edited 09-05-2019 12:32 AM
That's an interesting thought!
I do not realy want to go that way, it creates some unclear-hard-to-understand config.
I still have an option to create multiple WOL-servers.
But solution itself is cool, maybe I'll have to use it another time :)
09-04-2019 09:21 AM - edited 09-04-2019 09:37 AM
Hello eyazhuk,
I have read the document that you have provided
L3(config)#access-list 101 permit udp host 172.16.3.2 any eq 7
L3(config)#ip forward-protocol udp 7
On each "client facing" SVI
L3(config-if)#ip directed-broadcast 101
+
L3(config-if)#ip helper-address 172.16.3.2
The issue is on the SVI where the WOL server is connected where the suggested configuration is:
L3(config-if)#interface vlan 3 L3(config-if)#ip address 172.16.3.1 255.255.255.0 L3(config-if)#ip helper-address 172.16.2.255 L3(config-if)#ip helper-address 172.16.4.255
So you would like to have 60 ip helper-addresses here to be able to serve 60 different IP subnets.
But this is not possible as you have a limit of 16 helper-addresses.
These WOL packets are replicated on EACH ip helper-address statement.
The limit of 16 has been introduced because it protects the main CPU from load caused by excessive "COPIES" of each packet.
Here, I think that the Rick's suggestion is correct: to overcome the limit you need to split the servers in four servers connected to four different interfaces.
Each group of client WOL Vlans will point to a different server and each interface will have the corresponding ip helper-address using the corresponding directed broadcast IP subnets.
The server can be only one, using 4 NICs or even using a single NIC with Vlan tagging to create 4 different logical L3 interfaces.
I think that WOL generates very small amount of traffic so it should be safe to use 15 REPLICAS on a single interface.
Balaji's note is correct about other type of traffic that uses much more packets.
Hope to help
Giuseppe
09-05-2019 12:39 AM
09-05-2019 02:19 AM
09-05-2019 05:15 AM
Hello Leo,
it was a Catalyst 9500 or a Cat 3850 and the original poster needed a lot of ip helper-address for a customer built application, but Wake on LAN was not mentioned.
The limit was the same 16 ip helper-address per interface after an IOS XE upgrade.
Hope to help
Giuseppe
09-08-2019 11:59 AM
This has been an interesting discussion and there are several aspects of the ip helper-address and ip directed-broadcast to keep in mind. The ip helper-address is important on the interface/subnet where the server is located while ip directed-broadcast is important on interface/subnet where the clients are located. Some platforms do enforce limits on how many helper addresses can be configured on the interface. In this case I believe that my suggested solution is the most direct solution using some helper address commands on the originating interface, sending directed broadcast packets to intermediate interfaces/subnets which then send directed broadcasts to the ultimate destination subnets. There are other potential solutions which can be considered including having many WOL server on the original subnet each sending to a particular destination subnet. It may be possible that some WOL servers can be configured to send multiple WOL packets with one as a local broadcast (to be processed by the local helper-address) and other WOL packets send to remote directed broadcast addresses.
Thank you for marking this question as solved. This will help other participants in the community to identify discussions which have helpful information. This community is an excellent place to ask questions and to learn about networking. I hope to see you continue to be active in the community.
HTH
Rick
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: