You’re very close to isolating the real problem. What you’ve discovered is key:
- LSI card installed + NO backplane/SAS cables → server boots fine
- LSI card installed + backplane/SAS cables connected → system won’t complete boot / cycles off
- When you do get into the LSI utility, all drives show “Unconfigured”
That points to
a drive/backplane/cable issue (most common) or
power/PSU spin‑up issue, not Windows.
1) First: your RAID screen is keyboard-only (no mouse)
The screen in your photo is the
LSI BIOS RAID utility that says:
“Use Cursor Keys … Press Enter …”
So you won’t be able to “click” your way through it. If your wireless keyboard isn’t working reliably at boot, it will feel frozen.
Do this
- Use a wired USB keyboard (or PS/2 if your Supermicro has it)
- Plug into a rear USB 2.0 port
- In BIOS, enable USB Legacy Support (sometimes called “USB Keyboard Support”)
This usually fixes the “can’t get into MegaRAID / can’t control it” problem.
2) Stop the failed-drive alarm (beeping)
In that utility it’s typically here:
Objects → Adapter → Alarm Control → Disable/Silence
If you can’t find “Adapter”, tell me exactly what items appear under
Objects and I’ll map it precisely.
Temporary “make it stop now” option
If the alarm is unbearable and you can’t navigate the menu yet:
- Power off and disconnect the chassis speaker/buzzer lead from the motherboard header (often labeled SPKR / Speaker on Supermicro boards), or
- Some LSI cards have an on-card piezo buzzer; if yours does, you can unplug it if it’s on a small connector.
This won’t fix the RAID, but it will stop the noise while troubleshooting.
3) Why Windows cycles off only when the drive cables are connected
This is almost always one of these:
A) A single bad drive (or bad slot) is hanging the controller during scan
When the LSI firmware scans disks at boot, a dying drive can cause timeouts/hangs/resets.
B) Backplane/SAS cable problem
A marginal SAS cable or backplane can do the same thing.
C) PSU / power draw at spin-up
When all HDDs spin up together, the PSU may dip and the system resets/cycles.
4) The most effective next step: find the “poison” drive/slot
Since your data is
not critical, we can troubleshoot aggressively.
Do this test (best bang for time)
- Power off.
- Leave the LSI card + SAS/backplane cables connected.
- Remove ALL hard drives from the backplane (pull the trays).
- Boot.
If it boots reliably now, then:
- Insert ONE drive only, boot again.
- Repeat, adding one drive at a time until it starts cycling/failing.
The drive you inserted right before it starts cycling is very likely bad (or the slot/backplane position it’s in is bad).
If it fails even with no drives inserted, then the issue is more likely the SAS cable/backplane short or the LSI card itself.
5) “All drives show Unconfigured” — what that means
If
all drives show
Unconfigured (and you don’t see “Foreign configuration”), the controller is not seeing valid RAID metadata for an existing array.
That can happen if:
- The array config was cleared at some point, or
- You’re in a different controller/firmware mode than the array was created with, or
- The controller can’t properly read the disks (cable/backplane/drive issue)
Because your data isn’t critical:
The simplest end-state is usually:
- Create a new configuration
- Initialize (this wipes)
- Build your RAID again
Warning: “Initialize” is destructive. Only do it if you’re OK losing what’s on those RAID disks.
6) One quick clarification question (important)
The BIOS utility in your photo shows
M1068e (SAS1068E family). That is
not what I’d expect to see on a MegaRAID 9260-8i “WebBIOS” screen.
Can you confirm what the physical card actually is by:
- Posting a photo of the sticker/label on the card, or
- Writing the exact model shown during POST (it often says SAS1068E / SAS3081E / MegaRAID 9260, etc.)
This matters because the key combos and management tools differ by controller family.
Answering your main goal (“If I can get to the right window I can repair drive”)
Yes—but first we must make the controller stable enough to see the drives without rebooting.
Do these two things next:
1) Use a
wired keyboard and try
Objects → Adapter → Alarm Control to silence the alarm
2) Do the
one-drive-at-a-time test to find the drive/slot that triggers cycling
Reply with:
- What you see under Objects
- Whether the system cycles when no drives are inserted, and if not, which drive number causes the cycling when added back one-by-one.
