Jump to content
Nico_S

LSI SAS 9211 just killed my server

14 posts in this topic Last Reply

Recommended Posts

Is this just bad luck or do I miss something utterly obvious?

 

Mainboard: ASRock Z370 Extreme4

I installed a (the seller says) new LSI SAS 9211-8i 8-port card into the PCIe slot. The second I turned the machine on, sparks were flying out of the slot / card. Totally in shock - I never experienced something similar before - I pulled out the power cord. Luckily both brand new SSDs that were connected to the card are still working (checked on a Windows machine).  I haven't checked the HDDs that contain the actual data. Hopefully tomorrow the new mainboard will arrive...

 

Cheers,

Nico

IMG_20190712_132045_resized_20190712_073843949.jpg

IMG_20190712_132029_resized_20190712_073844532.jpg

Share this post


Link to post

OMG, in 9211 burning area, there usually have 2 fuse, but seems they don't work and make mainboard dead too.

Anyway you lost HBA and MB, shouldn't further.

Share this post


Link to post

Looks like the card was touching the metal surrounding the slot. 

Share this post


Link to post

I bet LSI never dreamed a motherboard manufacturer would be dumb enough to put a grounded shield where it could contact the card. I would be having a conversation with ASRock.

 

It looks to me like the electrically hot side of the surface mount fuse at F1 did indeed butt up directly against the metal shroud.

 

Good catch @StevenD!

Share this post


Link to post
Posted (edited)

By observision, shot circuit likely happen on slot metal cover, but I don't think so, the fuse or any components have enough clearances away from the cover.

 

Burning seems happen in front of fuse ( resistor worst ), so haven't avoid the incident .

 

Edited by Benson

Share this post


Link to post
21 minutes ago, Benson said:

the fuse or any components have enough clearances away from the cover.

Maybe the case angled the board towards that edge when tightened into the back panel?

 

I've seen people use too tall motherboard standoffs, or just poor manufacturing tolerances. Also, the typical card in that slot would have retention at the other end of the board with the "hockey stick" clip. This board only attaches at one end, allowing uneven insertion.

 

In any case, there was obviously zero clearance in an area that needed an air gap.

Share this post


Link to post
Posted (edited)
23 minutes ago, jonathanm said:

Also, the typical card in that slot would have retention at the other end of the board with the "hockey stick" clip. This board only attaches at one end, allowing uneven insertion. 

I agree this point much more.

 

If ref. PCIe pin assignment, once pin A3 B3 short with A4 B4, burnning will happen.

IMG_20190713_214958.thumb.jpg.383ed6f3d8955648580fef997613866f.jpg

 

Edited by Benson

Share this post


Link to post

Thanks everyone for your input. It very looks like the marked part of the card touched the metal of the PCI slot... but it's hard to tell afterwards.

Also I am very sure it was mounted correctly and vertically.

 

The good news is - the server is up and running again after a swap of the motherboard!

IMG_20190713_154209.jpg

Share this post


Link to post

Could you please take a picture of the burned card inserted into the burned motherboard? No need to mount it up in a case, just set the motherboard next to the edge of a box or the table so the PCIE card metal bracket can hang over the edge.

Share this post


Link to post

I'd send those pictures to both LSI and ASRock.

 

Somebody messed up. I can't find clear guidance on the physical specs as I'm not a PCI-SIG member, but there are only three possibilities I can see here.

 

1. ASRock ignored the specs when they added metal shielding where there wasn't supposed to be any conductive material present.

2. LSI ignored the specs by putting components in a "no go" area.

3. PCI-SIG hasn't actually codified one or the other of these specifications. I find that difficult to believe.

 

This kind of thing isn't supposed to be allowed to happen.

Share this post


Link to post
On 7/13/2019 at 5:53 PM, Nico_S said:

I will do that!

Thanks again everyone

Did you get any replies for this?

Share this post


Link to post

Hi there. So far it has been very frustrating.

I first called the seller (a quite big german online shop) and sent them the damaged motherboard. 2 weeks later the mainboard returned with a lettter saying they won't do anything, because I inserted the card improperly.

I then contacted the Asrock support - they told me to contact their RMA department. The RMA website says "please contact the seller". Which I did again.... They told me I probably have to pay for a repair when they send the mainboard on my behalf for a check / repair to Asrock's RMA.

What I haven't tried so far is contacting LSI...

Share this post


Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.