Quanta LB6M (10GbE) -- Discussion

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

TheBloke

Active Member
Feb 23, 2017
200
40
28
44
Brighton, UK
Hi all

I'm going 10G on my workstation and server soon. In fact, ideally 2 x 10G.

My preferred UK used hardware vendor has a very tempting price on both the LB4m and LB6m.

At first I thought Lb4m, just to get 1 x 10G between workstation and server and inter-connected with 1G to the rest of the network. But now I'm really tempted by the LB6m, because I want to enable 20G between workstation (Windows 10) and server (Solaris 11.3). Hopefully automatically with LACP, but worst case just 2 x separate links, but connected to the rest of the network rather than host-to-host.

The issue I have is that this will be my first and only 'proper' switch. So I need it to do both 10G and 1G, ideally providing at least 8 x 1G plus 4 x 10G ports. Hence I fall between the Lb4m (48 x 1 + 2 x 10) and the Lb6m (4 x 1 + 24 x 10) - unless I can use adapters in the LB6's SFP ports.


EDIT: Actually, having thought about it some more and looked up the cost of 1G SFP RJ45 transceivers, I've realised that if I really do need 8+ 1G ports, I really should just buy both switches!



EDIT 2: and now having read the whole thread, I've realised that in fact 1G almost certainly won't work in the LB6M's SFP+ ports at all, and certainly not via RJ45. So it's a moot point, and I'm hoping to buy both Lb4M and LB6M to get both 1G and 10G coverage.
 
Last edited:

aj84

Member
Oct 28, 2015
60
27
18
Germany
I have two servers hooked up to my switch. When I try to ping the switch's management IP (192.168.100.2) from either server, I get not response. The # of packets sent ticks up on their switch port when pings are attempted, so I know its traveling through the right interface on the server to the switch. The switch also cannot ping either server IP. Both servers are on ports 0/1 and 0/3 respectively and their interfaces are tagged with VLAN 10 and static IPs are set.

Code:
[...]

network protocol none
network parms 192.168.1.4 255.255.255.0 192.168.1.1
vlan database
vlan 10,30,40
vlan routing 10 10
vlan routing 30 30
vlan routing 40 40
exit
[...]

interface 0/1
vlan pvid 10
vlan participation include 10,30,40
exit

interface 0/2
exit

interface 0/3
vlan pvid 10
vlan participation include 10,30,40
exit
[...]

interface 2/10
routing
ip address 192.168.100.4 255.255.255.0
exit

interface 2/30
routing
ip address 192.168.30.4 255.255.255.0
exit

interface 2/40
routing
ip address 192.168.40.4 255.255.255.0

exit
[...]
Not sure what's going wrong, but things I would check:
- why don't you ping 192.168.100.4 (you wrote ".2"!?) or 192.168.1.4
- is routing globally enabled (can't find it in your config)? (configure > ip routing > exit)
- if it still doesn't work try the following: disable routing on vlan 10 (= remove interface 2/10); then: change ip address of the management interface (to the removed one, e.g. 192.168.100.4) and put it into vlan 10 (enable > network mgmt_vlan 10 > network parms 192.168.100.4 255.255.255.0)
 

fan

New Member
Mar 11, 2017
4
1
3
46
I finally put my LB6M in production this weekend. I changed the 3 fans on the back of the unit with the Delta AFB0412LB fans. The same fans can be used in the power supply as well. Very quiet at 18.5 dBA.

Some of the specs:
Size : 40mmX40mmX15mm
RPM: 5000 RPM
Air Flow: 6.5 CFM
Noise Level: 18.5 dBA

Amazon.com: Cooling Fan-AFB0412LB-F00 40x40x15mm Case Fan, 3pin, RPM: 5000 RPM, Air Flow: 6.5 CFM, Noise Level: 18.5 dBA: Computers & Accessories

Because these are 3 pin and not 4 pin, I can no longer get the temperature and fan speed using the "show environment" command.

Hopefully this will be helpful to others because I have gotten a lot of support on this forum.
I just changed fan to AFB0412LB. It is silent. But the sensors looks strange:
Code:
(FASTPATH Routing) #show environment

Temperature (Celsius).......................... 0
Fan Speed, RPM................................. 2147483647
Fan Duty Level................................. 0%
Temperature traps range: 0 to 45 degrees (Celsius)

Temperature Sensors:
Unit     Sensor  Temperature (Celsius)  State
----     ------  ---------------------  --------------
1        1       0                      Normal

Fans:
Unit   Fan     Type        Speed      Duty level State
----   ------  ----------  --------   ---------- --------------
1      1       Fixed       2147483647           0% Operational
1      2       Fixed       2147483647           0% Operational
1      3       Fixed       2147483647           0% Operational

Power supplies:
Unit     Power supply  Type          State
----     ------------  ----------    --------------
1        1             Removable     Failed
1        2             Removable     Failed
Temperature is 0!!!
 

alex1002

Member
Apr 9, 2013
519
19
18
Can someone please give me some insist on my multicast issues with VSAN? I would highly appreciate it.
 

alex1002

Member
Apr 9, 2013
519
19
18
So i have the following:

(switchname) (Config)# set igmp 10
(switchname) (Config)# set igmp querier 10
(switchname) (Config)# set igmp querier 10 address 10.1.0.100 <<-this is an address on my VSAN VLAN

Then "set igmp" on each interface

(switchname) #show igmpsnooping

Admin Mode..................................... Enable
Multicast Control Frame Count.................. 296952
IGMP Router-Alert check........................ Disabled
Interfaces Enabled for IGMP Snooping........... 0/1
0/2
0/3
0/4
0/5

VLANs enabled for IGMP snooping................ 10
I tried all these commands, but none work for me...
.

(FASTPATH Routing) (Config)#set igmp 20
^
% Invalid input detected at '^' marker.
 

aj84

Member
Oct 28, 2015
60
27
18
Germany
(The following is try-and-error - backup your config!)
I don't have any experience/knowledge with IGMP but have you tried something like the following? (I assume you want to enable it for a vlan.)
Code:
enable
vlan database
 set igmp 10                #10 = vlan-id
 set igmp querier 10 address 192.168.111.1               #10 = vlan-id, 192.168.111.1 = ip of whatever (Evan Battle says: IP of VSAN VLAN)
 exit
configure
 set igmp               # enable it switch-wide? ...is it necessary?
 exit
What do these say before and after above commands?
Code:
show igmpsnooping
show igmpsnooping querier
show igmpsnooping querier vlan 10               #10 = vlan-id
I'm just trying to help qithout using it myself...
 
  • Like
Reactions: alex1002

Terry Kennedy

Well-Known Member
Jun 25, 2015
1,140
594
113
New York City
www.glaver.org
I just changed fan to AFB0412LB. It is silent. But the sensors looks strange!
I keep trying to tell people this, but you can't just interchange fans based on dimensions and voltage.

AFB0412LB is a 2-wire fan for a fixed 12V application. It has no speed sensor and no supported way to vary the speed. AFB0412LB-F00 is a 3-wire fan for a fixed 12V application. It has a tachometer to report fan speed via open collector output and no supported way to vary the speed. AFB0412LB-R00 is a 3-wire fan for a fixed 12V application. It has a locked-rotor open collector output to report if the fan stops turning. It has no way to measure fan speed other than rotating / not rotating and no supported way to vary the speed. All are 5000 RPM fans.

There are a huge number of control / monitoring options. Here is a list from Nidec showing their thirty most common configuration options. As you read that document, you should note that you can't tell what option(s) a fan has just by the color of the wires.
 

fan

New Member
Mar 11, 2017
4
1
3
46
I keep trying to tell people this, but you can't just interchange fans based on dimensions and voltage.

AFB0412LB is a 2-wire fan for a fixed 12V application. It has no speed sensor and no supported way to vary the speed. AFB0412LB-F00 is a 3-wire fan for a fixed 12V application. It has a tachometer to report fan speed via open collector output and no supported way to vary the speed. AFB0412LB-R00 is a 3-wire fan for a fixed 12V application. It has a locked-rotor open collector output to report if the fan stops turning. It has no way to measure fan speed other than rotating / not rotating and no supported way to vary the speed. All are 5000 RPM fans.

There are a huge number of control / monitoring options. Here is a list from Nidec showing their thirty most common configuration options. As you read that document, you should note that you can't tell what option(s) a fan has just by the color of the wires.
Do you have any recommends to replace stock fan?

AFB0412LB is silent but after I power off switch and touch the main chip's heat sink, it was so hot! Obviously it was > 45 celsius.
 

Terry Kennedy

Well-Known Member
Jun 25, 2015
1,140
594
113
New York City
www.glaver.org
Do you have any recommends to replace stock fan?
I don't have one of these switches, so I don't have any specific advice. The manufacturer chose that specific fan to meet their performance requirements. Those requirements presumably include the ability to continue to operate with a single failed fan (so the cooling is probably over-engineered a bit) but also the expectation that the switch would operate in a datacenter (a more controlled environment than the average home). There's no "magic bullet" to get a quieter fan with identical performance specs - no fan manufacturer decides "we want to make a noisy fan". Quieting things down requires a fan with different performance specs - generally, larger fans and lower RPMs are quieter. If you're using the stock fan cutouts, then you have limited choices.

If I had one of these switches and had to make it quieter, I'd probably have some adapters 3D printed with a taper (a squared-off cone) so I could mount larger fans outside the case using the existing cutouts. That would allow me to move the same amount of air with a larger fan at a slower speed. That's the same idea as the original "TURBO-COOL" from PC Power & Cooling (then PC Cooling - Power came later). Of course, it would interfere with mounting equipment directly above or below it in a rack, but so does the "cut big holes in the top and put fans on them" solution.
 

fan

New Member
Mar 11, 2017
4
1
3
46
I don't have one of these switches, so I don't have any specific advice. The manufacturer chose that specific fan to meet their performance requirements. Those requirements presumably include the ability to continue to operate with a single failed fan (so the cooling is probably over-engineered a bit) but also the expectation that the switch would operate in a datacenter (a more controlled environment than the average home). There's no "magic bullet" to get a quieter fan with identical performance specs - no fan manufacturer decides "we want to make a noisy fan". Quieting things down requires a fan with different performance specs - generally, larger fans and lower RPMs are quieter. If you're using the stock fan cutouts, then you have limited choices.

If I had one of these switches and had to make it quieter, I'd probably have some adapters 3D printed with a taper (a squared-off cone) so I could mount larger fans outside the case using the existing cutouts. That would allow me to move the same amount of air with a larger fan at a slower speed. That's the same idea as the original "TURBO-COOL" from PC Power & Cooling (then PC Cooling - Power came later). Of course, it would interfere with mounting equipment directly above or below it in a rack, but so does the "cut big holes in the top and put fans on them" solution.
Thanks for the reply. I also plan to add 120mm PC fan...
 

HyperNet

New Member
Mar 19, 2017
5
0
1
Next I'm going to place a 120 MM fan in front of the power supplies and have a cardboard guide directing airflow right into their ends so those fans can be replaced with whisper quiet low CFM fans.
Added the 120mm fan in front of the power supplies. Although its not visible in this picture, the supply on the right is the one plugged in and it's top cover is off and its two fans have been removed. That eliminated the need to put a cardboard air chute. Plenty of air blows through the supply.

My switch is WHISPER QUIET and runs as cool as a cucumber. Once again, BE-CAREFUL about simply putting low CFM fans in. High heat will cause premature failure.


pw fan.JPG
 

Attachments

Andras

New Member
Mar 28, 2017
3
0
1
47
Hi,
Could someone help me?
I'm new to LB6M and to CLI setting of a switch.
I have managed to update the fw to 1.2.0.18.

I would like to clear the eventlog but the clear command doesn't recognize eventlog:

(QUANTALB6M) #clear eventlog
^
% Invalid input detected at '^' marker.

(QUANTALB6M) #

The second thing I would like to get help is that I would like to trunk the 4 1G ethernet ports (25-28) to connect it with a HP 2910.
I don't need any vlan,routing etc setting, the entire switch is fine with the default vlan.
Thx
 

sthsep

Member
Mar 7, 2016
72
10
8
Then just do something like this:


interface 0/25
addport 1/1
exit
interface 0/26
addport 1/1
exit
interface 0/27
addport 1/1
exit
interface 0/28
addport 1/1
exit

interface 1/1
description 'LAG'
no port-channel static
exit

And if it needs a vlan then just add the default vlan (vlan 1)
 
  • Like
Reactions: Andras

Andras

New Member
Mar 28, 2017
3
0
1
47
Then just do something like this:


interface 0/25
addport 1/1
exit
interface 0/26
addport 1/1
exit
interface 0/27
addport 1/1
exit
interface 0/28
addport 1/1
exit

interface 1/1
description 'LAG'
no port-channel static
exit

And if it needs a vlan then just add the default vlan (vlan 1)
Seems to be working... :)
I'll test it with iperf later.

About clearing the logs? Any idea?
I think that the orange led is blinking on the left of the switch because it has something in the log that should be cleared so the blinking would stop.
 

ajcorrea

New Member
Mar 31, 2017
2
0
1
40
I´m having issues with any SFP/SFP+ on LB6M !!!

None of these works....

running lastest version (final 18).

what´s de command to show info about sfp/sfp+ (vendor, rx/tx signal, etc).

Thanks


 

Terry Kennedy

Well-Known Member
Jun 25, 2015
1,140
594
113
New York City
www.glaver.org
what´s de command to show info about sfp/sfp+ (vendor, rx/tx signal, etc).
I don't know if this works specifically on the LB6M, but the generic Broadcom FASTPATH command is:
Code:
switch#show fiber-ports optical-transceiver

                                    Output    Input
Port      Temp  Voltage  Current     Power    Power   TX     LOS
           [C]   [Volt]     [mA]     [dBm]    [dBm]   Fault
--------  ----  -------  -------   -------  -------   -----  ---
Te1/0/23  34.6    3.316      7.2    -1.917   -2.151   No     No
Te1/0/24  36.9    3.272      7.4    -1.963   -2.742   No     No

 Temp - Internally measured transceiver temperatures.
 Voltage - Internally measured supply voltage.
 Current - Measured TX bias current.
 Output Power - Measured optical output power relative to 1mW.
 Input Power - Measured optical power received relative to 1mW.
 TX Fault - Transmitter fault.
 LOS - Loss of signal.
I don't know if there's a way to get an actual dump of the raw data (like Cisco's "show idprom interface <slot/port>").
 

gr8ape

Member
Jun 15, 2015
42
4
8
I´m having issues with any SFP/SFP+ on LB6M !!!

None of these work
Most of those are SFP not SFP+. You have a few SFP+ that at LR not SR so you would need a good chunk of cable as my understanding is they can fail with small runs.

What you want are SFP+ SR adapters.


Sent from my iPhone using Tapatalk
 
  • Like
Reactions: Andras

Terry Kennedy

Well-Known Member
Jun 25, 2015
1,140
594
113
New York City
www.glaver.org
Most of those are SFP not SFP+.
That is indeed an issue. Since the picture included some 10GbE parts and the OP said "none of them worked", it is unlikely that it is the only issue.
You have a few SFP+ that at LR not SR so you would need a good chunk of cable as my understanding is they can fail with small runs.
I use LR exclusively when wiring my POPs, other than a few SR when requested by specific customers and some ZR on my dark fiber DWDM links. In addition to many handoffs from carriers, I have hundreds of LR jumpers ranging from 1m to 5m connecting equipment and I've never run into a "too hot" signal.

This was definitely not the case on older optics - I routinely had to add attenuators to OC-3 circuits, and measuring the signal level and picking the right attenuator was a mandatory part of the turn-up process. But I've never had this on SFP / SFP+ optics.