Cannot get DHCP Relay to work on HP5130

[YellowCar]

Hi all,

I've got 4 HP5130 switches and i've setup a VLAN30 on them and on the VLAN i've tagged in VLAN1 as well (i think) and i've set a management IP on VLAN30 of 10.10.10.10 and gone to the DHCP options and enabled DHCP Relay and pointed it to a Windows Server on VLAN1 (the native VLAN) which is on 192.168.2.0   (both are on subent 255.255.192.0) but i cannot pick up a DHCP IP when i tag a device (such as my laptop) as VLAN30 ? If i simply plug into one of the ports that has VLAN30 on it without an VLAN ID then i get the normal LAN DHCP IP of 192.168.2.x

It seems like it should just work but i can't figure out why not ?

These are the first HP switches we've had so it's slightly new to me.

Thanks in advance

[16again]

Post your config.
First try putting in manual IP addresses, and check IP connectivity

[dnat]

I see this thread ilde for about a year now. I encountered the issue with DHCP relay, too. Here some information - maybe someone can help?

My customer wants to add a new Access switch, a Comware HPE 5130 version 7.1.045, Release 3113P05, to the network. The Edge (Distribution) switches are HP 5406 version K.15.08.0013, so are the core switches. The core switches are interconnected with each other and connect to two HPE Synergy. The virtualised MS DHCP servers are running on the Synergy devices (see screenshot).

The switch is configured but unable to obtain DHCP addresses. A Wireshark output shows, that the client is broadcasting but is not getting a reply back from the DHCP server.  If the broadcast is not getting out of the switch or the reply is not getting back. In simple words: I don’t know where the the DHCP broadcast gets stuck.

On the other hand, I can ping all the resources in the different VLANs. From the DHCP server, I can ping the switch and access its Web GUI. So my assumption is that my DHCP configuration is faulty.

Ports 1-20 shall be user ports with VoIP phones, Port 21 shall be a printer port, port 22 is for the WLAN AP and ports 23, 24 are for future use. As uplink port 25 is configured as trunk port (10 GBit fibre).

Here some VLAN information:

We want  to discontinue the use of VLAN 1

 VLAN 10    Server             192.20.2.0 /24    DHCP servers are in this range
VLAN 40    VoIP                192.20.6.0 /24    
VLAN 50    DMZ                192.20.9.0 /24    
VLAN 60    Infrastructure  192.20.5.0 /24    
VLAN 70    WiFi internal    192.20.x.0 /24    
VLAN 90    Management   192.20.1.0 /24    Network devices reside here

Some of my configuration:

ip redirects enable

dhcp enable

dhcp relay client-information record

interface Vlan-interface60

 ip address 192.20.5.247 255.255.254.0

 dhcp select relay

 dhcp relay server-address 192.20.2.21

 dhcp relay server-address 192.20.2.22

interface GigabitEthernet1/0/1

 description User

 port link-type hybrid

 undo port hybrid vlan 1

 port hybrid vlan 10 40 50 tagged

 port hybrid vlan 60 untagged

 stp edged-port

 poe enable

interface Ten-GigabitEthernet1/0/25

 description Uplink to Edge

 port link-type trunk

 port trunk permit vlan 1 10 40 50 60 70 80 90

 ip route-static 0.0.0.0 0 192.20.1.254

A high-level diagram and the config file is attached

[dnat]

UPDATE questions:

-> would I have to amend the VLAN config and configure them as link-type port hybrid, too?

-> did I miss a statement for DHCP like dhcp select relay?

My customer wants to add a new Access switch, a Comware HPE 5130 version 7.1.045, Release 3113P05, to the network. The Edge (Distribution) switches are HP 5406 version K.15.08.0013, so are the core switches. The core switches are interconnected with each other and connect to two HPE Synergy. The virtualised MS DHCP servers are running on the Synergy devices (see screenshot).

The switch is configured but unable to obtain DHCP addresses. A Wireshark output shows, that the client is broadcasting but is not getting a reply back from the DHCP server.  If the broadcast is not getting out of the switch or the reply is not getting back. In simple words: I don’t know where the the DHCP broadcast gets stuck.

On the other hand, I can ping all the resources in the different VLANs. From the DHCP server, I can ping the switch and access its Web GUI. So my assumption is that my DHCP configuration is faulty.

Ports 1-20 shall be user ports with VoIP phones, Port 21 shall be a printer port, port 22 is for the WLAN AP and ports 23, 24 are for future use. As uplink port 25 is configured as trunk port (10 GBit fibre).

Here some VLAN information:

We want  to discontinue the use of VLAN 1

VLAN 10    Server    192.20.2.0 /24    DHCP servers are in this range
VLAN 40    VoIP    192.20.6.0 /24    
VLAN 50    DMZ    192.20.9.0 /24    
VLAN 60    Infrastructure    192.20.5.0 /24    
VLAN 70    WiFi internal    192.20.x.0 /24    
VLAN 90    Management    192.20.1.0 /24    Network devices reside here

Some of my configuration:

ip redirects enable
dhcp enable dhcp relay client-information record

interface Vlan-interface60  ip address 192.20.5.247 255.255.254.0  dhcp select relay  dhcp relay server-address 192.20.2.21  dhcp relay server-address 192.20.2.22

interface GigabitEthernet1/0/1
 description User  port link-type hybrid  undo port hybrid vlan 1  port hybrid vlan 10 40 50 tagged  port hybrid vlan 60 untagged  stp edged-port  poe enable

interface Ten-GigabitEthernet1/0/25
 description ULS-25-8  port link-type trunk  port trunk permit vlan 1 10 40 50 60 70 80 90

A high-level diagram is attached, full config below:


=~=~=~=~=~=~=~=~=~=~=~= PuTTY log 2017.11.23 16:20:20 =~=~=~=~=~=~=~=~=~=~=~=

#

 version 7.1.045, Release 3113P05

#

 sysname NewAccessSwitch

#

 irf mac-address persistent timer
 irf auto-update enable
 undo irf link-delay
 irf member 1 priority 1

#

 ip redirects enable

#

 dhcp enable
 dhcp relay client-information record

#

 lldp global enable

#

 password-recovery enable

#

vlan 1

#

vlan 10

 name Server

#

vlan 40

 name VoIP

#

vlan 50

 name DMZ

#

vlan 60

 name Infrastruktur

#

vlan 70

 name WLAN intern

#

vlan 80

 name Wireless intern

#

vlan 81

 name GPGAST

#

vlan 90

 name Management

#

stp region-configuration

               
 region-name gpag

 active region-configuration

#

interface NULL0

#

interface Vlan-interface1

 ip address 192.20.11.249 255.255.255.0

 undo dhcp select server

#

interface Vlan-interface60
 ip address 192.20.5.247 255.255.254.0
 dhcp select relay
 dhcp relay server-address 192.20.2.21
 dhcp relay server-address 192.20.2.22

#

interface Vlan-interface90

 ip address 192.20.1.249 255.255.255.0
 undo dhcp select server

#

interface GigabitEthernet1/0/1

 description User
 port link-type hybrid
 undo port hybrid vlan 1
 port hybrid vlan 10 40 50 tagged
 port hybrid vlan 60 untagged
 stp edged-port
 poe enable

#

#

interface GigabitEthernet1/0/21

 description Printer
 port link-type hybrid
 port hybrid vlan 1 60 untagged
 stp edged-port

#

interface GigabitEthernet1/0/22

 description WLAN AP
 port link-type hybrid
 port hybrid vlan 10 40 50 70 80 to 81 tagged
 port hybrid vlan 1 60 untagged
 stp edged-port

#

interface GigabitEthernet1/0/23

 port link-type hybrid
 port hybrid vlan 10 40 50 70 90 tagged
 port hybrid vlan 1 60 untagged
 stp edged-port

#

interface GigabitEthernet1/0/24

 port link-type hybrid
 port hybrid vlan 10 40 50 70 90 tagged
 port hybrid vlan 1 60 untagged
 stp edged-port

#

interface Ten-GigabitEthernet1/0/25

 description ULS-25-8
 port link-type trunk
 port trunk permit vlan 1 10 40 50 60 70 80 90

#

interface Ten-GigabitEthernet1/0/27

 description ULS-27-8
 port link-type trunk
 undo port trunk permit vlan 1
 port trunk permit vlan 10 40 50 60 70 80 90

#

interface Ten-GigabitEthernet1/0/28

 port link-type trunk
 undo port trunk permit vlan 1
 port trunk permit vlan 10 40 50 60 70 80 90
 shutdown

#
 ip route-static 0.0.0.0 0 192.20.1.254
#

domain system

#

 domain default enable system

#

user-group system

#
 ip http enable

 ip https enable
 
#

[dnat]

Meanwhile, I resolved the problem.

The core is also acting as "router on a stick", responsible for Intra-VLAN routing. On its VLAN 60, the ip helper addresses are defined already. What I needed is to make sure, that my L2 VLAN traffic is reaching the core and DHCP packets are not discarded. To make it work, I did the following:

1. configure user ports as port link-type access on VLAN 60, rather than port link-type hybrid.

2. Remove the VLAN60 interface

That was it. A minor thing with a big impact. All services are running now.