Brocade ICX Series (cheap & powerful 10gbE/40gbE switching)
- Thread starter fohdeesha
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Is anyone using opnsense with a 6610
I cannot for the life of me get it to work.
From a client PC I can ping the firewall - no drop packets
From the firewall I can't ping the client - 3/3 dropped packets
From the firewall I can ping the switch but I only get 1/3 packets.
My WAN and LAN have the correct Ethernet port assigned
I'm running 192.168.1.1/24 and I also configured a my interface as vlan 1
Edit:
It's now intermittently working. I can get Internet for a few seconds on certain clients and then it drops off again.
Client PCs see the DHCP server on 192.168.1.1
I cannot for the life of me get it to work.
From a client PC I can ping the firewall - no drop packets
From the firewall I can't ping the client - 3/3 dropped packets
From the firewall I can ping the switch but I only get 1/3 packets.
My WAN and LAN have the correct Ethernet port assigned
I'm running 192.168.1.1/24 and I also configured a my interface as vlan 1
Edit:
It's now intermittently working. I can get Internet for a few seconds on certain clients and then it drops off again.
Client PCs see the DHCP server on 192.168.1.1
Last edited:
I’ve used the 6610 for a few years with OPNsense. Now using a 7650. Never had problems, although the 7650 is now running as the core router and OPNsense is purely a gateway firewall. Drawing your network would make it easier to figure out what’s going on
Hi all, I've got a weird one with my 7450-48P. Updated to the latest firmwares on Fohdeesha's site.
It's powering an ubiquiti unifi 6 enterprise happily right up until I plug a desktop (not Poe) into a neighboring port. Then no PoE on any port works. I could watch the light go out on the AP as soon as I plugged the desktop in and start booting back up as soon as I removed it.
The same desktop/cable don't seem to be causing issues on my Nortel switch the Brocade is supposed to be replacing.
Any ideas?
Cheers,
Nate
It's powering an ubiquiti unifi 6 enterprise happily right up until I plug a desktop (not Poe) into a neighboring port. Then no PoE on any port works. I could watch the light go out on the AP as soon as I plugged the desktop in and start booting back up as soon as I removed it.
The same desktop/cable don't seem to be causing issues on my Nortel switch the Brocade is supposed to be replacing.
Any ideas?
Cheers,
Nate
If you plug a computer directly into the LAN port of the OPNSense firewall (ie unplug the switch from the OPNsense device and plug the computer into the port that the switch was plugged into), does the computer work normally and have access to the internet, etc?
If you do, then you know it is a switch setup issue (or possible a hardware issue). Are you sure the switch has been fully reset to default settings? If you followed the how-to in this thread, that should have been part of the process.
If not, then you know it is a OPNSense setup issue. How new are you to OPNsense (or pfSense)?
Last edited:
Would you mind clarifying this; what's set at 1.1/24 (should be just the lan in opnsense), and what's configured as vlan1? vlan1 is the default vlan, so no additional "configuring" should be needed.
Did you set that up in opnsense, or just in the 6610? For right now, vlans should only live on the switch. We'll get to vlans later, but they shouldn't be used in this config. Especially vlan1.
Anyone able to help with this?
Monitor>boot system flash primary
...
BOOT INFO: debug enabled!!!
platform type = 12
PCIE-1 LTSSM status: 22
PCIE Switch status: 0
..............................
Firmware integrity checksum passed
.......
Process [1] stopped by sds monitor.
$
Monitor>boot system flash primary
...
BOOT INFO: debug enabled!!!
platform type = 12
PCIE-1 LTSSM status: 22
PCIE Switch status: 0
..............................
Firmware integrity checksum passed
.......
Process [1] stopped by sds monitor.
$
I've ran out of 10gb ports on the 6610 I own so I ordered a 6650 but still require the POE functionality on the 6610 and then realized that the 6610 supports stacking but not MCT and the 6650 supports MCT but no stacking. I'll move my 10gb equipment over to the 6650 but those hosts will still need access to POE clients staying on the 6610. Can I simply use a 40gb QSFP DAC cable to connect the 6650 to the 6610? I'm not worried about high availability and my network is flat. No routing done (that I'm aware of) on the current 6610. Could I create a link aggregation between the two 40gb QSFP ports on the 6610 and two on the 6650 with DAC cables (80gb total--not sure i'll even need it but curious) and be good to go?
Followed the guide, got this, followed guide again, still get this.
What should I do?
Looking for some help with multicasting / mDNS.
I have tried everything and can't get this to work. I currently have vlans setup with working ACLs to segregate my normal clients and my IoT devices. However, if I place devices like my printer or speakers in the IoT network, I cannot cast to them from my "client" network.
I have tried every combination of ip multicast and pin settings on the brocade and still can't get it to work.
I have also tried running a VM with avahi. I can see all of my devices on the different interfaces in avahi but it isn't broadcasting them to other vlans.
Does anyone have a similar setup that could walk me through it or at least show me how their config looks so I can have a start point?
Thanks
Here's what I'm using:
OPNsense on a proxmox VM
6450-48P in layer3 (router firmware)
Cloudkey Gen2 plus
Ubiquiti U6-Pro AP
I have tried everything and can't get this to work. I currently have vlans setup with working ACLs to segregate my normal clients and my IoT devices. However, if I place devices like my printer or speakers in the IoT network, I cannot cast to them from my "client" network.
I have tried every combination of ip multicast and pin settings on the brocade and still can't get it to work.
I have also tried running a VM with avahi. I can see all of my devices on the different interfaces in avahi but it isn't broadcasting them to other vlans.
Does anyone have a similar setup that could walk me through it or at least show me how their config looks so I can have a start point?
Thanks
Here's what I'm using:
OPNsense on a proxmox VM
6450-48P in layer3 (router firmware)
Cloudkey Gen2 plus
Ubiquiti U6-Pro AP
Has anyone replaced the power supply in a 7250-48P?
I have a 7250-48P with fried PoE. I'd like to remove the PoE board and the original power supply with PoE (first, to have a spare, second, to save energy).
Does anyone have closeup pics of the non-P power supply? Seems to be a Great Wall, but I cannot see voltage + current from the photos in the forum and the photos I have. It's probably 12V, but how many amps?
Also I'd have to measure the dimensions of the original PSU.
Meanwell EPP-150-12 might be a good replacement.
I have a 7250-48P with fried PoE. I'd like to remove the PoE board and the original power supply with PoE (first, to have a spare, second, to save energy).
Does anyone have closeup pics of the non-P power supply? Seems to be a Great Wall, but I cannot see voltage + current from the photos in the forum and the photos I have. It's probably 12V, but how many amps?
Also I'd have to measure the dimensions of the original PSU.
Meanwell EPP-150-12 might be a good replacement.
First off, a huge thank you to fohdeesha for your documentation and easy-to-use update package, getting this switch updated, reset and configured was straightforward.
I've got a 6610 that I've been installing at home, and part of the installation process was keeping the existing cheapo rosewill 1gig x 8-port switch in place as I moved things over port at a time. Every 5-10 minutes I would get a broadcast storm, duplicating broadcasts and reflecting them back to their origins. Mostly ARP requests but some other random traffic would happen as well. It'd be 20k packets or so at a pop then the 192.168.0/24 vlan would fault the ports (possibly on the linux side?) for a few minutes, then normal traffic would resume, then a broadcast storm, repeat forever.
Setup was core router <-- 40gbe on 1/2/6, tagged --> icx6610 <-- 1gig on 1/1/1 untagged --> rosewill -> random devices. One other server attached via 10gbe (1/3/1) on the 6610, most clients still connected to the old 1gbit switch at the time.
The fault is definitely with the cheap switch, as taking it out of the picture fixes it. However, it also doesn't occur when the cheap switch is software-bridged on via a 1gbe port on the server:
I.E: rosewill <-- 1gb --> core router <-- 40gbe --> icx6610.
(Linux bridge stack, no STP, just the 1g and 40g interfaces)
I'm trying to understand what went wrong. There's no loop, and while testing I was still seeing it with only one other device attached to the rosewill. There were two switches connected together before and no broadcast storms so it specifically is acting up when connected to the 6610.
It's "fixed" but I don't like fixes where it's unexplained. They have ways of coming back and biting you in the ass.
Any ideas for diagnosing the root cause? Ideally I'd like to have the 6610 catch and squelch this kind of thing incase some other device starts acting up. Once I've finished patching everybody over I can probably setup a test VLAN as a reproduction environment.
I've got a 6610 that I've been installing at home, and part of the installation process was keeping the existing cheapo rosewill 1gig x 8-port switch in place as I moved things over port at a time. Every 5-10 minutes I would get a broadcast storm, duplicating broadcasts and reflecting them back to their origins. Mostly ARP requests but some other random traffic would happen as well. It'd be 20k packets or so at a pop then the 192.168.0/24 vlan would fault the ports (possibly on the linux side?) for a few minutes, then normal traffic would resume, then a broadcast storm, repeat forever.
Setup was core router <-- 40gbe on 1/2/6, tagged --> icx6610 <-- 1gig on 1/1/1 untagged --> rosewill -> random devices. One other server attached via 10gbe (1/3/1) on the 6610, most clients still connected to the old 1gbit switch at the time.
The fault is definitely with the cheap switch, as taking it out of the picture fixes it. However, it also doesn't occur when the cheap switch is software-bridged on via a 1gbe port on the server:
I.E: rosewill <-- 1gb --> core router <-- 40gbe --> icx6610.
(Linux bridge stack, no STP, just the 1g and 40g interfaces)
I'm trying to understand what went wrong. There's no loop, and while testing I was still seeing it with only one other device attached to the rosewill. There were two switches connected together before and no broadcast storms so it specifically is acting up when connected to the 6610.
It's "fixed" but I don't like fixes where it's unexplained. They have ways of coming back and biting you in the ass.
Any ideas for diagnosing the root cause? Ideally I'd like to have the 6610 catch and squelch this kind of thing incase some other device starts acting up. Once I've finished patching everybody over I can probably setup a test VLAN as a reproduction environment.
Code:
Current configuration:
ver 08.0.30uT7f3
stack unit 1
module 1 icx6610-24-port-management-module
module 2 icx6610-qsfp-10-port-160g-module
module 3 icx6610-8-port-10g-dual-mode-module
stack disable
vlan 1 name DEFAULT-VLAN by port
vlan 172 name NFS by port
tagged ethe 1/2/6 ethe 1/3/1
router-interface ve 1
vlan 192 name INTERNAL by port
tagged ethe 1/2/6 ethe 1/3/1
untagged ethe 1/1/1 to 1/1/23
router-interface ve 2
jumbo
hostname coreswitch
ip dhcp-client disable
ip dns server-address ....
ip route 0.0.0.0/0 ....
no telnet server
snmp-server community ..... ro
interface ethernet 1/2/6
port-name ROUTER_UPLINK
interface ethernet 1/3/1
port-name SERVER
speed-duplex 10G-full
interface ve 1
ip address ....
interface ve 2
ip address ....
endI just picked up an ICX6610 which should be arriving next week.
Reading through the threads I saw these two serial cables mentioned in this thread that people have used. Would someone be able to confirm if both of these will work, or if one is preferred over the other?
I know that they are only a few dollars different from each other, and I'm mostly concerned with damaging the equipment if I use the wrong cable. It seemed about 2,000 posts earlier (I haven't read the entire thread), people were damaging their management ports on the switches by using the wrong USB-C cable. I'm hoping to avoid this if possible.
Thanks in advance!
Cable Matters USB C to RJ45 Console Cable for Windows and MacBook Pro (Compatible Cisco Console Cable, Rollover Cable) with FTDI 6 Feet
and
USB Cisco Console Cable, USB to RJ45 Console Cable Compatible with Routers/Switch/Windows 7, 8,10 (12ft)
Reading through the threads I saw these two serial cables mentioned in this thread that people have used. Would someone be able to confirm if both of these will work, or if one is preferred over the other?
I know that they are only a few dollars different from each other, and I'm mostly concerned with damaging the equipment if I use the wrong cable. It seemed about 2,000 posts earlier (I haven't read the entire thread), people were damaging their management ports on the switches by using the wrong USB-C cable. I'm hoping to avoid this if possible.
Thanks in advance!
Cable Matters USB C to RJ45 Console Cable for Windows and MacBook Pro (Compatible Cisco Console Cable, Rollover Cable) with FTDI 6 Feet
and
USB Cisco Console Cable, USB to RJ45 Console Cable Compatible with Routers/Switch/Windows 7, 8,10 (12ft)
I think the issue with the cables were on 7k series devices that had USB-C form factor ports for console access that didn't use standard USB-C pinouts. I believe it would try and negotiate power over some of the pins and fry something.
As far as I know, any router serial cable with an RJ45 end for the switch and labeled as Cisco compatible will work for a 6k device. But I don't have a 6610 specifically, so I can't 100% confirm. I use this one with a 6450.
I have a 6610 and I use a normal Cisco cable. I try to use a system with a physical serial cable because I didn't like the fact that some older Prolific drivers get "deprecated" by Windows automatically. I suppose you can use any of these FTDI USB converters and it should work. Since their serial end is a RJ45, you can always use a coupler and ethernet cable to extend them. I do this for my Cisco DE-9 to RJ45.
I also have a 7150-C12P and the USB-C port only requires a normal USB-C cable. When you connect it to a PC it should install the driver. If it doesn't you can get it from Brocade / Ruckus 7150 USB Serial driver Windows | Software Downloads | Ruckus Wireless Support and Brocade / Ruckus 7150 USB Serial driver MAC | Software Downloads | Ruckus Wireless Support. More details at
community.ruckuswireless.com
I also have a 7150-C12P and the USB-C port only requires a normal USB-C cable. When you connect it to a PC it should install the driver. If it doesn't you can get it from Brocade / Ruckus 7150 USB Serial driver Windows | Software Downloads | Ruckus Wireless Support and Brocade / Ruckus 7150 USB Serial driver MAC | Software Downloads | Ruckus Wireless Support. More details at
How to console in the ICX-7150-12CP 'Ruckus Switch'
You will need one of the following cables: An USB console cable with Type-C connector (In one side you will have a Type-C USB and in the other side a normal USB) Or A serial console cable with an RJ-45 connector (In one side you will have a RJ-45 connector and in the other a normal USB)...