Supermicro servers report memory trouble in three places: the BMC event log, the POST code on the board or in the BMC, and the BIOS memory screens. Each points to a specific DIMM or stage, which turns troubleshooting into a short, ordered process. This guide covers how to read those signals, the steps Supermicro's manuals and FAQs recommend, and when to replace a module.
Read the BMC first
- Health Event Log (System, Health Event Log on X12/H12 BMCs; the same section exists on X13/H13). Memory entries name the slot, for example "Correctable Memory ECC @ DIMMA2(CPU1) - Asserted". The log can be exported, which helps when you need to compare dates.
- Component Info, Memory: every DIMM with health, type, ECC, speed, size, serial and part number. A missing slot here usually means a seating or population problem.
- POST snooping: on BMCs that offer it, shows the current BIOS POST code, useful on headless servers.
POST codes that point to memory
Supermicro boards use AMI Aptio V BIOS. If the board stops during boot, the last code on the onboard display, or in BMC POST snooping, tells you where:
| POST code | Meaning (AMI Aptio V, per Supermicro) |
|---|---|
| 2B to 2F | Memory initialization in progress (SPD read, presence detect, timing, configuration) |
| 31 | Memory installed |
| 50 | Invalid memory type or incompatible memory speed |
| 51 | SPD reading failed |
| 52 | Invalid memory size or modules do not match |
| 53 | No usable memory detected |
| 54 | Unspecified memory initialization error |
| 55 | Memory not installed |
One beep at power-on means memory is not installed. Older Supermicro FAQs mention other codes, but the list above is the one Supermicro publishes for Aptio V boards.
Step-by-step troubleshooting
- Note the slot and the code. Read the Health Event Log entry or the POST code.
- Check the population. Supermicro boards support only the DIMM counts and layouts listed in the manual, with identical modules and both sockets populated the same way. See how to install RAM in a Supermicro server.
- Reseat. Power off, unplug, remove the flagged DIMM, clean and reseat it until it snaps into place.
- Swap test. Move the flagged DIMM to a known-good slot. Supermicro's FAQ: if the error follows the DIMM, replace it.
- Minimal boot. If the board does not POST, boot with one DIMM in DIMMA1 on CPU1, then add modules back channel by channel.
- Clear CMOS and update BIOS. Clear CMOS with the jumper or pads in the manual, then update to the current BIOS.
- Confirm the module type. A board that takes registered DIMMs will not run unbuffered modules, and the reverse. Mixed speeds or DRAM densities are not allowed on X13.
Built-in memory protection
Supermicro boards include memory RAS features that keep a server running through some faults. On Intel platforms, the Memory RAS Configuration menu (Advanced, Chipset Configuration, North Bridge, Memory Configuration) offers SDDC, which corrects faults within a single x4 DRAM device, ADDDC for a second device, and runtime soft Post Package Repair (sPPR), which repairs failing rows while the system runs. H13 EPYC boards offer Enhanced PPR, which runs a full memory test at boot and adds to boot time. For how correctable and uncorrectable errors differ, see how to troubleshoot ECC memory errors.
When to replace the DIMM
- Correctable ECC entries that follow the module to a new slot.
- Any uncorrectable ECC event on the same module after reseating.
- A module that fails the single-DIMM boot test in a known-good slot.
See also how to identify a failing server DIMM and what to do when a server will not POST after installing RAM.
Replacement memory
- Supermicro memory hub and all Supermicro listings.
- Examples: X13DEI, X12 dual-socket boards and H13SSL-NT.
- Which memory fits which generation: the Supermicro RAM compatibility guide.
Frequently asked questions
Where does a Supermicro server log memory errors?
In the BMC Health Event Log (System, Health Event Log on X12 and H12 BMCs). Entries name the DIMM, for example "Correctable Memory ECC @ DIMMA2(CPU1) - Asserted". The BMC Memory tab under System, Component Info shows each DIMM's health.
What does "Correctable Memory ECC @ DIMMA2(CPU1)" mean?
The memory controller corrected an error on that DIMM. Supermicro's FAQ says to remove, clean and reseat the DIMM, then swap it into a known-good slot to see whether the error follows it. If it does, replace the DIMM.
What does POST code 53 mean on a Supermicro board?
On AMI Aptio V BIOS, Supermicro lists 53 as no usable memory detected. Codes 50 to 55 cover memory type, SPD, size and presence problems.
How do I test a Supermicro board with one DIMM?
Power off, unplug, and leave a single module in DIMMA1 of CPU1 (P1-DIMMA1 on dual-socket boards). If it boots, add modules back one channel at a time.
Does Supermicro support third-party memory?
Supermicro recommends memory from its tested memory list and asks users to check its website for recommended modules. The list is published as a convenience without warranties. Use modules that match the board's supported type, speed and population.
How do I clear CMOS on a Supermicro board?
Unplug the system and short the clear CMOS jumper or pads named in the manual (JBT1 on the H13SSL), then power on and reload BIOS defaults.
Buying for more than one system
IT teams and integrators that maintain Supermicro servers can standardize on one kit per model. Request a quote with your models and quantities, or send us your system model and we will confirm the right module before you order.
Steps and rules are summarized from the vendors' published documentation; follow the guide for your exact model.