Fix 0X000009D4 Remoteboot Connection Error

Cause #1: Windows Firewall blocking the RPL port

What's actually happening here is your Windows Firewall (or a third-party one like Norton or McAfee) is dropping the UDP packets the Remoteboot server sends back. The RPL protocol uses UDP port 4011. When the server tries to reply, the firewall says nope.

I see this most often on Windows 10 workstations where someone enabled the firewall after the boot setup worked fine for months. The error appears suddenly, right after a Windows Update that reset firewall rules.

The fix

1. Open Windows Defender Firewall with Advanced Security as admin.
2. Click Inbound Rules then New Rule…
3. Choose Port → UDP → Specific local ports: 4011
4. Allow the connection. Check all three profile boxes (Domain, Private, Public) if you're not sure.
5. Name it something like "RPL Remoteboot"
6. Do the same for Outbound Rules.

Then reboot the workstation. The reason this step 3 works is because the Remoteboot server listens on UDP 4011 for BOOTP/DHCP discovery messages. If the firewall blocks those, the server never gets the request, so it can't tell the client where the boot image is.

Cause #2: Wrong server name or IP in the client settings

The error code 0X000009D4 (which is decimal 2516 by the way) maps to NERR_StartingRplBoot. That means the client did make contact with a server, but the server said "I don't know who you're talking about."

This happens when you moved the Remoteboot server to a different machine or changed its IP, but didn't update the RPL database or the client's DHCP reservation.

The fix

1. On the Remoteboot server, open the Remoteboot Service management console.
2. Check that the server name in the service config matches the actual machine name. No typos.
3. Open a command prompt as admin and run: nbtstat -a [client-IP] to verify NetBIOS name resolution is working.
4. If the client uses DHCP, make sure the DHCP scope option 060 (Vendor Class Identifier) is set to MSRPL, and option 066 (Boot Server Host Name) points to the correct server name or IP.

What often trips people up: the server name in the RPL database can't have spaces or special characters. Keep it simple like RPLSRV01. Also, Windows DNS might resolve the name to an old IP if you didn't flush the cache. Run ipconfig /flushdns on both sides.

Cause #3: Corrupted RPL database or boot block records

This one's rarer but nasty. The RPL database (stored in %SystemRoot%\System32\RPL\RPLSVC.DAT on older Windows versions) gets corrupted after a crash or improper shutdown of the server. The error won't show up immediately — it'll surface after a few days when the server tries to read a bad record.

The fix

1. Stop the Remoteboot service: net stop rplsvc
2. Back up the database file: copy C:\Windows\System32\RPL\RPLSVC.DAT C:\RPLBACKUP\
3. Run the database repair tool: rplboot -d -r (this rebuilds the index from the backup of boot block records)
4. Start the service again: net start rplsvc
5. Check the event log for RPL source errors. If you see Event ID 100 with a description about a missing boot block, you need to re-add the client's MAC address in the RPL manager.

The reason this step 3 is the nuclear option is because rplboot -d -r remaps all the boot block pointers. It won't fix a physically damaged database file, but it does fix index corruption. If the file itself is toast (rare on NTFS), you'll need to restore from backup or rebuild the whole RPL database from scratch — which is a pain.

Quick reference table

If none of these fix it, check if the client's network card supports RPL at all. Some cheap Realtek chipsets don't. You'll see that as a different error though — 0X000009D4 specifically means the server connection failed, not the card.