LSI 9285-8e and Supermicro SC837E26-RJBOD1 duplicate enclosure ID and slot numbers

Question

I have 3 bootstrap columns and the problem is: 1. column: column do not 'grow' when the text inside is getting bigger, so the text doesn't stop where the column ends. 2. column: when the text is not just a single word, but for example 2 words (xxx xxx) the column doubles

How can I change that so the column which is a circle stays the same and the text stays INSIDE the column? Thanks!

Html:

<div class="row">
  <div id="one" class="col-md-1">xxxxxxxxxxxxxxxxxxxxxxxxx</div>
  <div id="two"class="col-md-1">xxxxxxxxxxxx xxxxxxxxxx</div>
  <div id="three"class="col-md-1">xxxx</div>
</div>

Css

.row {
font-size:18px;
}
#one{
cursor:pointer;
background-color:#00b3b3;
border-radius:50%;
margin-left:150px;
width:240px;
hight:50px;
line-height:250px;
text-align:center;
}
 #two{
 cursor:pointer;
 background-color:#00b3b3;
 border-radius:50%;
 margin-left:100px;
 width:240px;
 hight:50px;
 line-height:250px;
 text-align:center;
}

#three{
cursor:pointer;
background-color:#00b3b3;
border-radius:50%;
margin-left:100px;
width:240px;
hight:50px;
line-height:250px;
text-align:center;
}

Solution

We managed to finally fix this issue. The ultimate cause and fix? It seems a manufacturing process error caused some JBODs shipped from Supermicro to come with a default burned in logical ID (the 0000007F). This address is actually supposed to match the SAS address by default.

To fix this issue, we had to run a tool called ExpanderXtools Lite (ftp://supermicro.com/utility/ExpanderXtools_Lite/). You run the SMC binary and will get a X window popup (you'll either need X installed, or if you are not running X on your servers like us, a local X server on your laptop being forwarded through SSH). In the SMC program you select the COM menu and click on inband.

Now, you can go to the WWN menu and select WWN. A new popup will show your JBOD primary and secondary (if you have the E26 model) controllers. You need to update both controllers at the same time before closing the window. After updated and the window is closed, power off the array for a bit and power back on. Use the SMC binary again to verify the logical address shows correctly.

The biggest pain point of this was having to power down the arrays. It might be possible to do this online and do a rescan with your RAID card. But better to play it safe. Port IDs for disks will change. For us, our LSI card was able to pick up the arrays after the change. Your mileage may vary.

Comments

[Admin]

Can you start the rebuild from the controller during POST as a work-around?

[Admin]

Tom, I added some MegaCLI output to outline the issue. Mike, nope! The machine is in heavy production and this is obviously the last thing I want to try...

[Admin]

Adding this information as it may be relevant. I believe the chip used in the Supermicro enclosures is lsi.com/products/storagecomponents/Pages/LSISAS2x28.aspx. I don't actually see a single 28 port PHY. So I am guessing this could be the chip used in both the E26 and E16 variants. If this is true, I don't see why the distinction between E26 or E16 should cause the problem. But this is very helpful information none the less! I am still continuing to solve this issue as it just recently caused a major headache for us.

[Spencer Rohan]

height is spelled wrong on css

[Spencer Rohan]

What are you trying to achieve? Do you want a specific height on each of these circles?

[Andy Shinn]

Is there a way for MegaCLI to rebuild an array using something other than slot and enclosure? The biggest issue here is that an array with a degraded drive in the second chassis cannot be rebuilt because it always selects slot 11 on chassis ID 36,

[Jan Marek]

@AndyShinn Did you try only remove this drive and give it back after a minute? It is a operation, which should initiate rebuild process...

[Philip]

(First time seeing this Q, sorry I couldn't help earlier) SuperMicro is somewhat infamous for screwing up Logical IDs, MACs, and any sort of unique ID. Fortunately LSI makes tools to renumber the screwed up devices, so it's usually a quick fix.

[Felipe Vergara]

I have exactly the same problem! Even asked question here, but in my case xflash throws double buffering error in glibc and cannot change anything. What OS/distro and version did you use?

[learningcoding]

Thanks. word-wrap: break-word; helps so the text stays inside the column but there is now a huge gap between the text on the first line and text on the second line) and also when I remove width and height, the column won't be a circle anymore-I just want it to stay a circle (all 3 columns circles should have the same size).

[Spencer Rohan]

@learningcoding That's because you have a line-height of 250px

[Spencer Rohan]

I've updated your css above -- Again, your problem isn't super specific to what you are trying to accomplish. You can also share these class names(or IDs) to share properties. You are declaring the same exact values for different properties. Instead of ID 1,2,3 -- maybe try class ="column"

[learningcoding]

Thanks but when I remove width and height, the circle is just as small as the text and if I remove line-height, somehow all words after the first one are invisible..

[Spencer Rohan]

@learningcoding - I think you need to be more specific in your explanation on what you are looking for. Add a height: 200px; to the css