2 Replies Latest reply on Jan 26, 2010 10:16 AM by JReese

    SE7525GP2 Memory Issues


      I mange a small office of 20 with a server running this board (SE7525GP2) with two 2.8GHz Xeons and 4GB (1GBx4) of Corsair Registered ECC PC2700 DDRSDRAM. The server is running Windows Small Business Server 2003 and performing duties of AD/DNS/DHCP and File & Printer Sharing and it has been in use since 2005. When it was originally built it started with 2GB of RAM but about a year ago an additional 2GB of RAM was added and I have not had any issues prior to this. However, while performing maintenance on the server over the weekend I started getting post errors on reboot that halted the system reading the following.


      Error (8502): Bad Memory in DIMM 1A

      Error (8501): Bad Memory in DIMM 2A

      Error (8505): Bad Memory in DIMM 2B

      Error (8506): Bad Memory in DIMM 1B

      Error (8510): All memory marked as failed. Force all memory back online.


      Press F1 to Resume, F2 to Run Setup


      I find it quite strange that all four memory modules would suddenly just go bad; especially since two of the modules are from a different manufacturing time frame so I am wondering if there is something else I should be looking at first. I haven't tested the modules with Memtest86+ yet but I plan to do that tonight after hours. However, now that the server has more of a load on it this morning (Monday morning) than it did over the weekend I have seen a few application crashing issues on the server that seem like it may be memory related. Also, some overall sluggishness where it normally would be seen.


      Besisides testing the memory modules is there anything else I can look into to try and track this down?

        • 1. Re: SE7525GP2 Memory Issues

          Even if the memory is physically okay, if it's not on the list at


          you're going to get unpredictable results

          • 2. Re: SE7525GP2 Memory Issues

            I would have to disagree that is not nessicarily true; that is why Intel even states it right on the memory specifications page.

            Note: This list is not intended be all-inclusive. It is
            provided as a convenience to Intel’s general customer base, but Intel
            does not make any representations or warranties whatsoever regarding the
            quality, reliability, functionality, or compatibility of these memory


            Also, since the system has been running without any issue at all since June of 2005 with the original two modules and since November 2008 with all four modules; I really don't think "compatibility" issues would start becoming an issue all the sudden. I have worked with Intel boards for quite a while and although I know they are one of the more "picky" (if you want to call it that) brand of motherboads when it comes to memory; as long as you buy a good brand of memory that adheres to the specifications that Intel sets for the motherboard then there are usually no issues at all.


            Anyway, after some hardware testing last night after business hours I believe I found the issue to be an actual bad module of the first pair bought in June 2005. I haven't validated that yet with Memtest because it would have taken more time than I had last night but I hope to this weekend. I believe that the reason that all four of the modules were being labeled as failed during POST was because the module that seems to be bad was in the first slot (DIMM1B); which is the primary slot that needs to be populated first. I believe since this primary module had somewhat failed it was making the system throw the POST errors on the rest of the modules since the first one wasn't being fully detected. As of now the system appears to be working normally on the two newer modules purchased in November 2008. Testing this weeked will confirm I hope.


            In any case thanks for the reply.