Good morning. I have two Banana Pi Pro8x units. The main one had been running without a reboot for 12 days.
Yesterday I decided to reboot it, and when it starts up I get this error:
[ 48.965524] sfp sfp1: SM: enter probe:down:down event timeout
[ 48.982610] sfp sfp1: module OEM PAC-10G-T rev A6.1 sn Mav2509006 dc 250909
[ 48.992037] sfp sfp1: sfp: support mode 00,00000000,00000800,00006040
[ 48.998484] mxl862xx mdio-bus:10 mxl_lan5: optical SFP: interfaces=[mac=4,22-23,27,29, sfp=27]
[ 49.007101] mxl862xx mdio-bus:10 mxl_lan5: requesting link mode inband/10gbase-r with support 00,00000000,00000800,00006440
[ 49.018226] sfp sfp1: SM: exit present:down:down
[ 49.022838] sfp sfp2: SM: enter probe:down:down event timeout
[ 49.038470] hwmon hwmon6: temp1_input not attached to any thermal zone
[ 49.045792] sfp sfp2: module FINISAR CORP. FCBG110SD1C03-1R rev A sn WYD02SE dc 170925
[ 49.055186] sfp sfp2: sfp: support mode 00,00000000,00000400,00006040
[ 49.061628] mtk_soc_eth 15100000.ethernet eth1: optical SFP: interfaces=[mac=1-4,22-24,27,29, sfp=27]
[ 49.070863] mtk_soc_eth 15100000.ethernet eth1: requesting link mode inband/10gbase-r with support 00,00000000,00000400,00006040
[ 49.082416] sfp sfp2: SM: exit present:down:down
[ 49.100695] mt7996e_hif 0001:01:00.0: assign IRQ: got 125
[ 49.106150] mt7996e_hif 0001:01:00.0: enabling device (0000 -> 0002)
[ 49.112518] mt7996e_hif 0001:01:00.0: enabling bus mastering
[ 49.118306] mt7996e 0000:01:00.0: assign IRQ: got 122
[ 49.123357] mt7996e 0000:01:00.0: enabling device (0000 -> 0002)
[ 49.129380] mt7996e 0000:01:00.0: enabling bus mastering
[ 49.205547] mt7996e 0000:01:00.0: attaching wed device 0 version 3.0
[ 49.253511] platform 15010000.wed: MTK WED WO Firmware Version: ____000000, Build Time: 20230218204509
[ 49.262852] platform 15010000.wed: MTK WED WO Chip ID 00 Region 2
[ 49.322900] mt7996e_hif 0001:01:00.0: attaching wed device 1 version 3.0
[ 49.586374] mt7996e 0000:01:00.0: HW/SW Version: 0x8a108a10, Build Time: 20250605125803a
[ 49.586374]
[ 69.605522] mt7996e 0000:01:00.0: Message 00000007 (seq 12) timeout
[ 69.611797] mt7996e 0000:01:00.0: Retry message 00000007 (seq 12)
[ 90.085500] mt7996e 0000:01:00.0: Message 00000007 (seq 13) timeout
[ 90.091757] mt7996e 0000:01:00.0: Retry message 00000007 (seq 13)
[ 110.565504] mt7996e 0000:01:00.0: Message 00000007 (seq 14) timeout
[ 110.571763] mt7996e 0000:01:00.0: Failed to start patch
[ 131.045505] mt7996e 0000:01:00.0: Message 00000010 (seq 15) timeout
[ 131.051775] mt7996e 0000:01:00.0: Retry message 00000010 (seq 15)
[ 151.525499] mt7996e 0000:01:00.0: Message 00000010 (seq 1) timeout
[ 151.531668] mt7996e 0000:01:00.0: Retry message 00000010 (seq 1)
[ 172.005500] mt7996e 0000:01:00.0: Message 00000010 (seq 2) timeout
[ 172.011669] mt7996e 0000:01:00.0: Failed to release patch semaphore
And of course Wi‑Fi doesn’t load because it errors out. I had it on a flashed SD card; I took the SD out and formatted it three times — same error, same error. I tried a brand‑new SD card, never used — same error. I tried booting from the factory NAND — same error. Obviously, since I have a second Banana Pi Pro8x, in order to have Wi‑Fi I changed the configuration and connected the XGS-PON, because if you look at the log, it’s not the real one from yesterday since right now I’m using a DAC cable.
Right now I have the second router (which I used for compilation tests) as the main one, working perfectly with the XGS-PON and with Wi‑Fi throughout the house.
I rule out the image because it’s the one I always use, and also because I tested the factory NAND and it gives the same error of not loading the Wi‑Fi card drivers.
Two things come to mind:
-
Some cable may have come loose — maybe when I removed dust from the router while cleaning with a feather duster, a cable got loose and is causing the failure.
-
The BE14000 may have failed.
There can’t have been an overvoltage in any way because I use this power strip for each router and computer I have at home to prevent damage. It already happened to me with a 500-euro router because there was a power outage in the neighborhood due to construction work and when power came back the router was dead.
Power strip I use: Belkin Regleta de enchufes con 8 tomas y 2 puertos USB-A (2,4 A) con protección contra sobretensiones, 900 julios, ladrón, cable de 2 metros, base múltiple con protección eléctrica, hogar y oficina : Amazon.es: Electrónica, so I rule out it being related to that.
About this product 8 protected AC outlets: Protects your electronics at the office or at home from harmful voltage fluctuations, power surges, and current spikes up to 900 joules with our three‑wire AC surge protection. Continuous protection: Protects your devices against unexpected voltage spikes such as short circuits and lightning. Also acts as a safeguard against sudden voltage fluctuations or low-level surges and brownouts. 2 built-in USB-A ports: Charge your mobile devices quickly while shielding them against damage and data loss thanks to the two integrated USB-A ports (2.4 A shared). Safe, smart design: This USB power strip/extension lead features high-quality construction and a durable design, including a damage-resistant housing that protects the circuits against fire, impact, and corrosion. 2 m cable: The durable 2-meter power cable lets you place this smart power strip discreetly out of sight, while ensuring optimal power flow to your devices. Trusted surge protection: Its smart design includes safety features that provide effective defense against power surges and voltage spikes. Buy with confidence knowing it’s backed by a 2-year limited product warranty and a connected-equipment warranty of up to €65,000.
Right now, since I’m compiling and using it without Wi‑Fi, I don’t need it. Later I’ll put on the gloves I have so that no part of my body can deliver any static electricity, and I’ll fully disassemble it to see what could have happened — whether it’s a cable issue, the BE14000 is damaged, or some cable came loose.
I hope the router isn’t broken, because getting the Return Merchandise Authorization (RMA) was incredibly difficult. It was a real struggle to get the official store to send it. Thanks to a very compromising email—due to a mistake by the sender requesting something I’d rather not mention to avoid damaging Banana’s reputation—I finally got the RMA; they even told me it was an RMA, pretending not to know.
I mention this because it’s not a cheap router or motherboard. Besides, getting the RMA again would be extremely complicated. Since this is the second router, I hope they don’t ask me the same silly questions as last time, and that they’re a reputable company and send me the RMA quickly, because I don’t want another broken router.
I’m posting it in the forum to give everyone a heads-up. When the compilation finishes, I’ll work on it. The thing is, if it fails, I’ll have to remove one of the two BE14000 units I’m using and try it in this one, to be 100% sure whether it’s the BE14000’s fault or the router’s fault.
I’ll tell you later how it goes, and if it turns out to be the router (or development board, however they prefer to call it), hopefully this time there won’t be problems with the RMA.
Best regards.