Bugzilla – Bug 11133
Duplicate MAC addresses on SB Touch
Last modified: 2009-09-08 09:25:46 UTC
Per bug 11132 and depending on the investigation into duplicate MACs on Duet, we need to put in place appropriate precautions to prevent duplicate MACs on any Fab4. This bug is a placeholder on that action point.
I propose the factory line has Internet access and every Fab4 register the MAC address with SN or some other system as it is being programmed. If the MAC already exists in the database, a red error screen appears and the unit is set aside to be fixed.
We need to have a plan for this. Duplicate MAC = broken player = return for exchange = lose money
Dan told me we haven't had any reports about dupes for Boom. So whatever the factory did for Boom was probably right.
I don't think they did anything differently. They just print out unique labels and scan those during production. Is that right, David?
But that definitely didn't work with SBC's, of which there have been a lot of dupes I believe.
Do we know if any of these units have numbers that don't match the labels? either there are duplicate labels being printed or the wrong mac address is getting put on the unit. There is another bug that lists an off-by-one situation, which is likely operator error (scanning the wrong barcode or something)
We need to foolproof the process and protect against operator error. I don't know how we can do that unless we have each unit talk to SN over the Internet.
Problem is that sometimes on the factory floor they need to reprogram them, so given that, there's no way except through human engineering and user confirmation to prevent these things from getting out into the world. One thing we do have in fab4 is the existing MAC address in the wifi card. in theory we could pair this against that database as a safety check. Not sure how hard that would be to extract, though.
Boy, these bugs have a life of their own... I am including on Fab4 what was added to Boom and Jive. A list of session written MACs is maintained and if any previous MAC during the session is repeated the operator is issued an "are you sure" message before allowing the write. This has proven a successful method for avoiding most MAC mistakes on the line. If this continues to be an issue I will see if a separate station can be used to independently verify the MAC on each device.
Ah, that's great, David! It sounds like we need to see some new duplication scenarios before we can fix them. Re-open if you have one.
I don't have access to #11132. Are you talking about duplicates between different systems, or duplicates on the same box, different interfaces having the same MAC? My fab4 has the same MAC address on both its Ethernet and wireless interfaces: eth0 Link encap:Ethernet HWaddr 00:04:20:22:00:AF inet addr:192.168.9.29 Bcast:192.168.9.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1988 errors:0 dropped:0 overruns:0 frame:0 TX packets:1698 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:319206 (311.7 KiB) TX bytes:455092 (444.4 KiB) Base address:0x8000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:2 errors:0 dropped:0 overruns:0 frame:0 TX packets:2 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:260 (260.0 B) TX bytes:260 (260.0 B) wlan0 Link encap:Ethernet HWaddr 00:04:20:22:00:AF UP BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:23 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:5724 (5.5 KiB) TX bytes:0 (0.0 B)