Drag to reposition cover

Brocade ICX Series (cheap & powerful 10gbE/40gbE switching)

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

LodeRunner

Active Member
Apr 27, 2019
540
227
43
Thank You, you’re a star.

what’s the recommended method of upgrading a stack? Un-stack and do them one at a time?
ive got a pair of stacked 7250’s.
Since it's the same firmware as my 7450 stack, just update the active master and it will handle updating stack members and performing a rolling restart. Anything that's cross-chassis LAG'd will have no downtime.
 
  • Like
Reactions: Drewy

fohdeesha

Kaini Industries
Nov 20, 2016
2,728
3,076
113
33
fohdeesha.com
Since it's the same firmware as my 7450 stack, just update the active master and it will handle updating stack members and performing a rolling restart. Anything that's cross-chassis LAG'd will have no downtime.
indeed, you can follow the guide from just the UFI subsection to flash the new 8095d ufi image, it'll handle flashing it to all stack members ICX7250 / ICX7450 - Fohdeesha Docs
 

ZFSZealot

New Member
Aug 16, 2021
26
6
3
Does anyone have a scrap mainboard? I need two replacement push pins for the main routing IC heatsinks. I guess they failed with heat and age.

View attachment 9590
This is really digging up bones, sorry, this is one of the few mentions I've found for this happening, this is specific to the ICX6610.

With my first 6610, it was just the chip on the right side as viewed from the front, and it was obvious that it got hotter than the other one. I wonder if this is because that group of 24 ports got used and the other one didn't. Differences in heatsink color and PCB under the chip in question are pretty clear in the photos I'm posting. Chip on the right in the first pic is obviously mounted to the area on the left in the second pic, board was just flipped left to right:

IMG_0314Small.JPG
IMG_0315Small.JPG

This seems to have happened also to a lesser degree on the QSFP+ board, with the heatsinks on the full 40G port side being a little more brown than the black ones to their left.

Noted this as I was disassembling and cleaning dust out of the switch, primarily because I heard parts rattling around inside on receipt which turned out to be pieces of the heatsink pins and springs. Complete teardown was in order to make sure the pieces weren't going to short something.

I also found the same thing as mentioned further down from the quoted message, even though the pins are gone, the heatsink cannot be removed with any reasonable amount of twisting force.

Does anyone know exactly where/how to order the pins and springs? Unless these are glued well enough that the pins aren't really a concern? And just as a curiosity thing, why one chip lived life hotter than the other? Switch seems to work just fine as far as I've tested it.
 

Attachments

texteditor

New Member
Oct 8, 2019
15
4
3
Before I start bricking stuff, I'm looking to upgrade my 7150-C12 to 8090k from the fohdeesha default 8080. Guides/videos talk about usb upgrade, but before I brick it, they mention the security of NOT having a cfg installed when going to upgrade.

So... how do you do that? I have the ICX7150 and manifest files on my usb, but how else do I prepare the switch for the upgrade? Is there a doc that I didn't see about "simple" upgrade paths?

I for whatever reason could not get my 7150-c12 to take a TFTP'd image, but the USB method worked for me very well. I even thought I botched it at one point but was able to fix whatever I thought I messed up
 

nickf1227

Active Member
Sep 23, 2015
197
128
43
33
Since it's the same firmware as my 7450 stack, just update the active master and it will handle updating stack members and performing a rolling restart. Anything that's cross-chassis LAG'd will have no downtime.
Just make sure if you are pre8080 you go to 8080 first then to current.
 
  • Like
Reactions: Drewy

mmx

New Member
Dec 18, 2015
9
9
3
I love the revisions done to the guide!

Also, hack the planet! :D

EDIT: The serial number has 2 interpretations. Can I share my findings? :)
 
Last edited:
  • Haha
  • Like
Reactions: Yunia and ZFSZealot

mmx

New Member
Dec 18, 2015
9
9
3
Go for it although it looks like you noticed the LID hash already :p
First one I came up with was: "a beef box jk" lol

My second interpretation/guess was a word similar to your username, but wasn't able to make it work. :)
 

rootwyrm

Member
Mar 25, 2017
74
93
18
www.rootwyrm.com
I also found the same thing as mentioned further down from the quoted message, even though the pins are gone, the heatsink cannot be removed with any reasonable amount of twisting force.

Does anyone know exactly where/how to order the pins and springs? Unless these are glued well enough that the pins aren't really a concern? And just as a curiosity thing, why one chip lived life hotter than the other? Switch seems to work just fine as far as I've tested it.
The heatsink being unable to be removed is not a good sign. It means the TIM has likely failed. These do not use paste or thin pads, they use special impregnated fabric which has limited adhesive properties. When it fails it can turn into the consistency of glue or epoxy. The only way to remove it is to heat things up (carefully) until it lets go again.
Do NOT use thermal paste, silicone, or closed cell pads as a replacement. This way baked devices lie, because the already marginal interface will be terrible. I recommend Fujipoly SARCON Form-In-Place SPG-25B or SPG-30 as an alternative, the ODM equivalent (superior) is Fujipoly SARCON NR-Tc, or Fujipoly Extremely Compressible PG25A or above.

The pushpin setup is almost always specific to the device, but they are a very common part. You can use a nut-and-bolt setup (USE LOCTITE! they will back out) in a pinch as long as you're extremely careful not to over-tighten. If you have to go metallic, you must insulate because the mounting holes have ground plane pads. It's almost impossible to find pushpins in anything less than qty 100, but you can try ATS at Digikey.
 
  • Like
Reactions: klui and fohdeesha

ZFSZealot

New Member
Aug 16, 2021
26
6
3
The heatsink being unable to be removed is not a good sign. It means the TIM has likely failed. These do not use paste or thin pads, they use special impregnated fabric which has limited adhesive properties. When it fails it can turn into the consistency of glue or epoxy. The only way to remove it is to heat things up (carefully) until it lets go again.
Do NOT use thermal paste, silicone, or closed cell pads as a replacement. This way baked devices lie, because the already marginal interface will be terrible. I recommend Fujipoly SARCON Form-In-Place SPG-25B or SPG-30 as an alternative, the ODM equivalent (superior) is Fujipoly SARCON NR-Tc, or Fujipoly Extremely Compressible PG25A or above.

The pushpin setup is almost always specific to the device, but they are a very common part. You can use a nut-and-bolt setup (USE LOCTITE! they will back out) in a pinch as long as you're extremely careful not to over-tighten. If you have to go metallic, you must insulate because the mounting holes have ground plane pads. It's almost impossible to find pushpins in anything less than qty 100, but you can try ATS at Digikey.
Oh wow. I'm assuming TIM is "thermal interface material"? If the device appears to still work properly, this is still something that has to be addressed, correct?

Is this a common issue with the 6610's or did I just get lucky?

Also, I had thought about using a bolt and nut setup, but with a nyloc nut and the springs that I was able to recover, is that a reasonable option?

Here are temperatures, it's idle but nothing seems too far out of line?

Code:
ICX6610-48_01#sh chassis
The stack unit 1 chassis info:

Power supply 1 (AC - Regular) present, status ok
        Model Number:   23-0000144-01
        Serial Number:  0VH    
        Firmware Ver:    B
Power supply 1 Fan Air Flow Direction:  Front to Back
Power supply 2 (AC - Regular) present, status ok
        Model Number:   23-0000144-01
        Serial Number:  04S    
        Firmware Ver:    B
Power supply 2 Fan Air Flow Direction:  Front to Back

Fan 1 ok, speed (auto): [[1]]<->2
Fan 2 ok, speed (auto): [[1]]<->2

Fan controlled temperature: 62.0 deg-C

Fan speed switching temperature thresholds:
                Speed 1: NM<----->78       deg-C
                Speed 2:       73<-----> 84 deg-C (shutdown)

Fan 1 Air Flow Direction:  Front to Back
Fan 2 Air Flow Direction:  Front to Back                        
MAC 1 Temperature Readings:
        Current temperature : 43.5 deg-C
MAC 2 Temperature Readings:
        Current temperature : 56.0 deg-C
CPU Temperature Readings:
        Current temperature : 62.5 deg-C
sensor A Temperature Readings:
        Current temperature : 46.5 deg-C
sensor B Temperature Readings:
        Current temperature : 49.0 deg-C
sensor C Temperature Readings:
        Current temperature : 40.5 deg-C
stacking card Temperature Readings:
        Current temperature : 48.0 deg-C
        Warning level.......: 81.0 deg-C
        Shutdown level......: 84.0 deg-C
 

Nnyan

Active Member
Mar 5, 2012
142
42
28
I’m thinking of getting a 6450 since I really don’t like junos on the 3300ex I just picked up. I’m not doing anything too fancy on it (some L3 routing and VLANs) but I prefer a workable UI. What’s the brocades like?