X11SBA-LN4F vs A1SRi-2558F



  • Hi

    @mr2dave:

    If it helps; I bought an E200-9B from superbiiz.com last week. I received rev 1.02 hardware and haven't had any issues with it so far.

    I ordered this from a UK supplier which ordered it, I was told, from Supermicro as required.  My revision was 1.01.  I emailed Supermicro tech support and was told the board had been revised with the following:

    1. Solve only 2 LAN show up to BIOS menu for LN4F SKU
    2. Solve LAN EEPROM show up PXE-E05 error message issue
    3. Add rework instruction

    I think the first are likely a BIOS update only as the board came with the latest version, and the rework instruction is the fix applied.

    You can check for this rework yourself on a version 1.01 of the board, it is pretty obvious once you have located the component which is U35, to see it has been changed and hand soldered, see the attached image taken from Newegg's detailed images of the board.  My rework looked a bit neater but it was still obvious looking at the solder and comparing to other components it had been reworked.

    Hope that helps anyone else purchasing one.

    Regards

    Phil




  • Holy crab, don't they have someone who can actually solder SMD devices?
    If one of my guys would call such a job complete he'd have a hard time…



  • Hi

    Just seen on the X11SBA-LN4F that despite having 1600 DDR3L memory, which is reported in the BIOS at 1600MHz, the memory is reported as actually running at 1066MHz, this is seen in the IPMI system hardware list and if doing a #dmidecode in pfSense shell and viewing the memory info.

    Memory is Crucial CT51264BF160B 2 x 4GB.  Everything is running good but it's a bit annoying if memory is clocked slower than it needs to be.  Unfortunately there are no manual memory settings in the BIOS to force anything to run any differently.  It could just be a reporting error from the board perhaps.

    Anyone else seeing this.

    Regards

    Phil



  • @Phil_D,

    Mine is the same.  I'm not going into BIOS right now to snoop (it's running OK and has a 250 day uptime, lol).  Should it ever need a reboot, I'll investigate further.



  • Hi

    @Engineer:

    @Phil_D,

    Mine is the same.  I'm not going into BIOS right now to snoop (it's running OK and has a 250 day uptime, lol).  Should it ever need a reboot, I'll investigate further.

    Many thanks for checking.  The BIOS will probably report 1600MHz as mine does, so that is either the operating speed and the reported 1066MHz is wrong, or the BIOS is just reporting what the memory is saying it will support and is clocked slower at 1066MHz.

    I agree overall it probably doesn't make too much difference to performance, but annoying to not have it running at it's designed speed.  I suspect Supermicro only support 1600MHz on it's tested memory, which for this board is limited to just one option which I couldn't find available to buy.

    I will drop their tech support an email.

    Regards

    Phil



  • Hi

    Supermicro support have said that only memory they have tested will run at 1600MHz on this board, it seems anything else defaults to 1066MHz with no way of overriding that in the BIOS.  The number of tested memory SODIMMs that will run at 1600MHz is just one, and good luck sourcing it.

    Regards

    Phil



  • Thank you for reporting this.
    The manual though does say in general that 1600 Mhz memory clocks at 1600 Mhz and that only 1333 Mhz memory clocks down to 1066 Mhz. Nothing about tested memory. And I even bought Class 9 memory.




  • Hi

    Exactly, there is nothing about tested memory only being supported in the specifications, which I have raised them.

    Is your memory running at 1600MHz or also reported as running at 1066MHz?

    Their tested memory is rated at CL=11, which is the same as the memory I'm using, so it's not as though their tested memory is anything special, and the Intel CPU/SoC should be fine supporting anything with the correct rating.

    Of course it could be their support is being lazy and just saying they don't want to know as the memory isn't in their tested list (a list of one), and for all I know even if I had the their tested memory I may still see it running at 1066MHz as it may be a bug in the BIOS.

    Regards

    Phil



  • @Phil_D:

    Is your memory running at 1600MHz or also reported as running at 1066MHz?

    My memory is Kingston HyperX Impact Black Series - DDR3L 8 GB : 2 x 4 GB - SO DIMM 204-PIN - 1600 MHz / PC3L-12800 - CL9 and running at 1066 Mhz.



  • Hi

    @gogol:

    My memory is Kingston HyperX Impact Black Series - DDR3L 8 GB : 2 x 4 GB - SO DIMM 204-PIN - 1600 MHz / PC3L-12800 - CL9 and running at 1066 Mhz.

    I recommend yourself and anyone else with this problem getting in touch with Supermicro support on their website, the more people that complain the more likely they will fix it and unlock the BIOS to more memory.

    Regards

    Phil



  • I have been researching this MBD-X11SBA-LN4F-O board, I saw all this talk here before ordering and decided to give it a shot.

    FYI, for anyone interested in buying this board. I ordered from Newegg and received the 1.02 hardware version.

    I lost access to the web console twice, so I was a little worried I might have the same problem that Engineer described. However both times I was messing with the firewall so I am sort of thinking that had something to do with it. Anyway I digress…. I am testing it my work and it's been up three days and I have had a constant web video stream going 24/7 (part of the time through two of the LAN interfaces at the same time.) No crashes or anything weird so far since I was messing with the firewall.

    Engineer do you have any suggestions I could try to really test it out before I put it into "production?" This first build will just go into my apartment with a 300 mbps fiber connection. I just don't want to piss my roommates off and have it crash so I want to test it for a week or so. Just wondering what people would suggest specifically for this board relating to the earlier problems with the timeouts.

    P.S. Thanks for all the info, this was all very interesting and helped me decide to give this board a shot. I have been running a Gigabyte embedded 1007u board on PFSENSE for quite awhile with zero issues. It's been rock solid, but I wanted to lower the power draw, and get AES, and Intel NICS etc. This SM board seems to hit all the points for my situation. Anyway, just curious what people (esp Engineer) would suggest to put it through its paces.



  • I got an X11SBA-LN4f with the 1.02 pcb revision and still encountered the watchdog issue on igb1(LAN). Even after upgrading the bios to 1.0b. It took about 43 days of uptime before it triggered. I RMA'd it and got an even more defective unit.

    The X11SBA-LN4F is a good system on paper but these problems make it too unreliable for a router. I would advise new buyers to avoid it.



  • Very good thread, was just looking at these 2 SM boards and found this using google. I do have a few questions;

    1. Is this problem only on the X11SBA-LN4f board or does it also affect the A1SRi-2558F?

    2. are there any reported problems with the  A1SRi-2558F board pfsense? Planning on using OpenVPN.



  • 1. Is this problem only on the X11SBA-LN4f board or does it also affect the A1SRi-2558F?

    X11SBA-LN4f only.

    2. are there any reported problems with the  A1SRi-2558F board pfsense? Planning on using OpenVPN.

    No.



  • Hi

    Just an update from me. I've been running this box for days and days and not experience any timeout issues relating to the LAN1, I have version 1.01 of the board which Supermicro confirmed by serial number it had been modified and fixed for the watchdog issue.

    In addition there is a new firmware for the X11SBA-LN4F but no release notes, but I will give it a go.

    Regards

    Phil



  • In addition there is a new firmware for the X11SBA-LN4F but no release notes, but I will give it a go.

    Phil, if you have updated it, it would be nice to hear from you about, please drop us a couple
    of lines about it. I am also thinking to get hands on one or two of this boxes.



  • @Phil_D:

    In addition there is a new firmware for the X11SBA-LN4F but no release notes, but I will give it a go.

    Yeah, supermicro is funny about their firmware updates.



  • Just an update that my build has been up for 341 days since the repair with no watchdog faults.  Still running 2.2.5.



  • Since there appear to be alot of A1SRi-2558F owners in this thread. Does anyone else experience hangs during boot? It is random, but FreeBSD will hang at the following line:
    ACPI APIC Table: <intel  tiano ="">I was running BIOS 1.01a and just upgraded to 1.01c that I saw recently came out. No idea if that addresses it.

    This issue make reliable recovery from power failure impossible. Anyone use any sort of watchdog to handle reset when this happens - I haven't tried/looked yet.</intel >



  • @krbvroc1:

    Since there appear to be alot of A1SRi-2558F owners in this thread. Does anyone else experience hangs during boot? It is random, but FreeBSD will hang at the following line:
    ACPI APIC Table: <intel  tiano ="">I was running BIOS 1.01a and just upgraded to 1.01c that I saw recently came out. No idea if that addresses it.

    This issue make reliable recovery from power failure impossible. Anyone use any sort of watchdog to handle reset when this happens - I haven't tried/looked yet.</intel >

    Hello,

    Yes, I have the same issue exactly on my A1SRi-2758F board.  Using the reset function on IPMI to reboot usually fixes it, but I've had times when it took 4-5 reboots to fix the issue.  I've only seen this since pfSense 2.3.2 so I'm wondering if it's something in the patch of FreeBSD that's in use.  Could be fixed in pfSense 2.4 running on FreeBSD 11?

    Brian





  • Hi @Hugovsky,

    Thanks for pointing this out.  I've seen this before but I thought that once you saw this that the device would never boot again.  The fact that I can reset the unit and it works made me think that it isn't the problem.  I'm going to contact SuperMicro and see if this errata might be the cause.

    Brian



  • Hi,

    Alright, got a reply from Supermicro and they said that this isn't (for me at least) being caused by the Intel erratum.  In my case, they said that the memory I'm using may be the culprit.  Seems strange that the memory could be the issue though as it's always at exactly the same place, and once we get past that point in the boot things are fine.

    Brian



  • Got my Supermicro X11SBA-LN4F today - Rev. 1.02 :)
    Was a little disappointed that i have to order the correct I/O Shield for my 505 Rackmount case. For that price it could have been included…

    What i wanted to ask the Supermicro owners. The CPU Cooler is wobbling. So it does not sit tight. you can move it with your finger. Is that normal? There are springs on the pins. I do not understand how this construction work or should work...



  • Was a little disappointed that i have to order the correct I/O Shield for my 505 Rackmount case. For that price it could have been included…

    Normally it let you safe money! It is included in the desktop and in the 1U bare bone and for custom cases
    you may don´t need it because you will be need there also a custom I/O shield.

    What i wanted to ask the Supermicro owners. The CPU Cooler is wobbling. So it does not sit tight. you can move it with your finger. Is that normal? There are springs on the pins. I do not understand how this construction work or should work…

    Any cooler should be sitting tough on the chip, and if there will be something wbbling it cou ld be that during
    the transport one of the small pins (with the springs) is broken or only jumped out of the hole in the board
    to get a better overview over that you should turn around that board and watch from the underlying or
    bottom side that this both pins are inserted and not has broken snappers. This are in normal $ cent
    articles and can be mostly obtained from the next electronic parts shop for cheap. For low budget
    you be able to order them at amazon.com for $2.14 (10 pieces) as an example.

    Here are some examples of this pins (with springs):

    ![different push pin models.jpg](/public/imported_attachments/1/different push pin models.jpg)
    ![different push pin models.jpg_thumb](/public/imported_attachments/1/different push pin models.jpg_thumb)
    ![metal and plastic pins with springs.jpg](/public/imported_attachments/1/metal and plastic pins with springs.jpg)
    ![metal and plastic pins with springs.jpg_thumb](/public/imported_attachments/1/metal and plastic pins with springs.jpg_thumb)
    ![metal pins with springs.jpg](/public/imported_attachments/1/metal pins with springs.jpg)
    ![metal pins with springs.jpg_thumb](/public/imported_attachments/1/metal pins with springs.jpg_thumb)
    ![metal push pins with spring and washer (disc).jpg](/public/imported_attachments/1/metal push pins with spring and washer (disc).jpg)
    ![metal push pins with spring and washer (disc).jpg_thumb](/public/imported_attachments/1/metal push pins with spring and washer (disc).jpg_thumb)



  • Frank,
    thank you very much for your very detailed explanation!
    I checked it and pins and springs are OK.
    There are only 2 Pins for the cooler and you can move it so that the springs are become more tention.
    So it seems that it works as it should work and that there is some lash for the cooler.
    Temp is about 50° in my supermicro 505 rackmount case without active cooling in a full network rack.

    I took a better look at the standard shield from the case. and yes, it works if you open every inlay and leave it closed for displayport/hdmi.