if you've only gotten that far you can safely just reboot/unplug the switch and start over. it's only after the "erase" command under "replacing bootloader" that there's no turning back (at least not until you have a new bootloader copied)
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Not really, just point it to a folder and make sure you put the two brocade files in there. Then the switch should be able to access it. If you have Windows firewall on it might not be able to connect
I am happy to announced that this newbie has managed to flash successfully! Thanks again for the wonderful effort and clear instructions.
So I'm in a bit of a pickle...
I've been following this thread for a while, got a few LB6Ms for a LAB environment, and I'm having an issue with one after flashing it.
Basically, when I first flashed it and I was trying to setup the OOB and IB mgmt settings, it wasn't reachable via either method. I think it may have been a bad flash, so I went into the bootloader to re-flash back to stock and then back to Brocade. The issue I'm having remains. I plug in a cable to the MGMT1 port and set the ip (via ip address 192.168.1.49/24) and even though it says it is set, and the port light is on, and I have a connection, I cannot ping anything outside of the switch, and nothing else can ping the switch. The setup is very basic, a computer with Ubuntu and the TFTP server role, and the switch, both manually set IPs. So, because of this, the TFTP file pull times out and I can't do anything else.
I have done the t2 diag menu for ethernet, and that hasn't helped (boot, Monitor>t2, ethernet test > all tests > esc,esc > set IP > try to pull TFTP files (no luck).
The other two LB6M switches I have I did the same upgrade to (via the guide) and even reverted and etc once. No issues with them. When I set the IP in the bootloader (with the same net config) I can instantly ping both ways.
Any ideas or suggestions would be appreciated. If I have to buy a JTAG programmer and do that, is there a guide?
I've been following this thread for a while, got a few LB6Ms for a LAB environment, and I'm having an issue with one after flashing it.
Basically, when I first flashed it and I was trying to setup the OOB and IB mgmt settings, it wasn't reachable via either method. I think it may have been a bad flash, so I went into the bootloader to re-flash back to stock and then back to Brocade. The issue I'm having remains. I plug in a cable to the MGMT1 port and set the ip (via ip address 192.168.1.49/24) and even though it says it is set, and the port light is on, and I have a connection, I cannot ping anything outside of the switch, and nothing else can ping the switch. The setup is very basic, a computer with Ubuntu and the TFTP server role, and the switch, both manually set IPs. So, because of this, the TFTP file pull times out and I can't do anything else.
I have done the t2 diag menu for ethernet, and that hasn't helped (boot, Monitor>t2, ethernet test > all tests > esc,esc > set IP > try to pull TFTP files (no luck).
The other two LB6M switches I have I did the same upgrade to (via the guide) and even reverted and etc once. No issues with them. When I set the IP in the bootloader (with the same net config) I can instantly ping both ways.
Any ideas or suggestions would be appreciated. If I have to buy a JTAG programmer and do that, is there a guide?
Well that's certainly interesting. If you answer each question below we should hopefully be able to fix this quick
1. Do you have other flashed LB6M's plugged in while trying to set up and ping/use this broken LB6M? If so they all have the same mac address if you haven't changed it, which would explain these issues. You say that you couldn't even make contact using the IN band management, which tells me it's some kind of software or network issue
2. Stupid question, but the IP you're assigning the switch is in the same subnet as the tftp server right? and you're sure it's plugged into mgmt port 1, not 2?
3. In the brocade bootloader on the broken switch, run the below command. It should reboot back into the bootloader, now with a fresh valid MAC. Give it an IP and try pinging out to something in the same subnet again. If it still fails, try the t2 ethernet test, then try again.
set ether-address 748e.f8e2.d200
#pull power, then re-power. the "reset" command doesn't work in brocade BL so it can't be used
If it's not a MAC conflict with your other flashed switches, it could potentially be it using an invalid MAC. it tries to pull the chassis MAC address from EEPROM, when it sees the garbage data in there leftover from Quanta it should ignore it, but I guess it's possible that's not happening. The above command would resolve that
4. If all the above fails, try a different ethernet cable of a different length, plugged into a different switch/endpoint if you have one. If you're using a shorter cable, find a longer one, vice versa. If it IS your mgmt port refusing to work with the brocade fw EQ/clock values, changing the impedance/distance on the ethernet side has a slim chance of helping
Try all that and let me know what happens. Worst case it shouldn't come to JTAG. The units can be quite expensive with the right firmware on ebay. Another STH user completely bricked his, he shipped it to me and back so I could JTAG it, I think it cost him $40 total. I'd be willing to do the same for your chassis (if you're in the USA) if we can't get it figured out
Last edited:
I will give step 3 a try tomorrow. As for 1 and 2:
1 - Switch is only connected to the server running TFTP with no other LB6Ms connected. Mac address was change as part of the initial flash, though I erased the flash (it only has the bootloader now) so step 3 may change things (if perhaps the MAC is garbage).
2- IP is the same as in the guide. The switch is set to 192.168.1.50/24 and the TFTP server that it is connected to (it's a VM on hyper-V with a dedicated physical NIC going to the switch) is set to 192.168.1.49. This config works with the other two switches just fine.
3- I didn't realize the RESET cmd didn't work. I thought it just caused the switch to halt/hard stop. Whoops. Otherwise, I'll give this a try as the MAC could be garbage.
4- I've tried different cables, even a brand new Cat6 one. I have not tried a different port, as it's the only one available to get to the TFTP server (that the other switches work with). I will however try on a different system just to see if I can ping.
Thanks for the quick reply. I'll follow up tomorrow with the tests.
third sentence in the guide: "The software itself requires no license or activation to fully function"
Thanks,. just was confused because a License was mentioned here.
from: https://www.servethehome.com/turbocharge-the-quanta-lb6m-with-brocade-turboiron/
yes I agree that's confusing wording, I'm assuming patrick put that there to protect himself and STH from any "legal repercussions"
(however after telling a Brocade SE at work about this, he laughed and couldn't care less, granted he doesn't speak for their legal department, but I'm assuming they couldn't care less about EOL firmware either)
(however after telling a Brocade SE at work about this, he laughed and couldn't care less, granted he doesn't speak for their legal department, but I'm assuming they couldn't care less about EOL firmware either)
So, resetting the MAC fixed it. Looks like whatever was going on was related to a bad/garbage MAC address. All 3 of my switches are flashed and working great so far on the Brocade V8 L3 firmware.
THANKS for the help!
Great news! And also interesting, never seen that happen before. I'll add it to the guide under emergency steps.
Be sure to use the same command to set the MACs on your other chassis so they don't all have the same, setting it to the original MAC (on the label on the right side of the switch) is probably best
After some debugging and digging, I'm officially changing the guide's recommendation to default to using the v7 image, not the v8 - even for non-critical applications.
It turns out the v8 image was basically a beta, I don't think it was ever even publicly released to customers for the TI (don't ask how I got it).
One of the major issues is it seems to very primitively respond to STP packets - even with STP disabled! this is why v8 LACP groups would flap for some people - if the other switch (or any other switch in the network) originated STP packets, the TI V8 firmware went nuts with LACP flapping.
the v7 image of course has none of these problems, it's the LTS "stable" codetrain for the TI, was patched and updated a good 15 months longer than the v8 image (all the way up to 2016), and is very well tested - haven't found a single issue yet.
Since there's no downside to running the v7 image (exact same featureset), and it's technically "newer" (latest commit date wise) I recommend everyone run it unless you explicitly need v8 for some reason. There weren't any syntax changes, so if you flash down to v7 from v8, your old config should continue working fine with no changes needed
will update the guide's text to reflect all this
It turns out the v8 image was basically a beta, I don't think it was ever even publicly released to customers for the TI (don't ask how I got it).
One of the major issues is it seems to very primitively respond to STP packets - even with STP disabled! this is why v8 LACP groups would flap for some people - if the other switch (or any other switch in the network) originated STP packets, the TI V8 firmware went nuts with LACP flapping.
the v7 image of course has none of these problems, it's the LTS "stable" codetrain for the TI, was patched and updated a good 15 months longer than the v8 image (all the way up to 2016), and is very well tested - haven't found a single issue yet.
Since there's no downside to running the v7 image (exact same featureset), and it's technically "newer" (latest commit date wise) I recommend everyone run it unless you explicitly need v8 for some reason. There weren't any syntax changes, so if you flash down to v7 from v8, your old config should continue working fine with no changes needed
will update the guide's text to reflect all this
Hi
I run md 0xfff80000 20, I get the different output:
=> md 0xfff80000 20 fff80000: 27051956 552d426f 6f742032 3030392e '..VU-Boot - Pastebin.com
Is it safe to flash?
I run md 0xfff80000 20, I get the different output:
=> md 0xfff80000 20 fff80000: 27051956 552d426f 6f742032 3030392e '..VU-Boot - Pastebin.com
Is it safe to flash?
Quite simple since the switch is already set up as brocade - just use the built in brocade image commands:
enable
copy tftp flash 192.168.1.8 v7image.bin primary
reload
It'll come back up with v7, and your config still intact. That said, the v8 image isn't HORRIBLE - if you're not having issues then you don't have to rush to change it or anything. I just realized it was more of a beta build than a long term release. That lacp bug is pretty bad, for example
for all you collectors out there, how about a......Dell Powerconnect TurboIron??
http://www.dell.com/downloads/global/products/pwcnt/en/switch-powerconnect-b-t124x-specsheet.pdf
and no, sadly they just ran the same brocade firmware we're already running, so no new software to try
http://www.dell.com/downloads/global/products/pwcnt/en/switch-powerconnect-b-t124x-specsheet.pdf
and no, sadly they just ran the same brocade firmware we're already running, so no new software to try