03:27 <Whiskey`> ok got me a usb cable.... and failed to connect to rdnis network =\

03:28 <zmatt> Whiskey`: on the programmed one you mean? well if it's a highly custom system then there's no reason to expect an rndis interface there

03:34 <Whiskey`> uh thats an unprogrammed one i think

03:34 <Whiskey`> waiting on a new uart adapter, cant find mine and my son's seems to have no tx

03:35 <zmatt> then what do you mean exactly by "connect to" rndis? it will show up as rndis interface but there's nothing to "connect" to, the beaglebone will act as bootp/tftp client

03:35 <zmatt> also, I think newer Windows systems don't recognize it because it is notoriously difficult to get windows to recognize usb devices and they keep changing things

03:36 <Whiskey`> linux is just being derpy then. its giving a error 'activation of network failed'

03:36 <zmatt> what is?

03:36 <Whiskey`> network manager

03:37 <zmatt> network manager is probably trying to get an IP address, which it won't get

03:48 <zmatt> you'll need to setup static configuration

03:48 <zmatt> (obviously, since the bbb will be getting its ip *from you* via bootp)

06:43 <Guest8> Does anyone know what could cause the onboard EMMc to fail to load with a fresh image on the SD card?

06:44 <zmatt> what do you mean? what exactly are you observing?

06:44 <Guest8> (Beagleboard Black AM3358)

06:44 <zmatt> beaglebone black, not beagleboard black

06:44 <Guest8> Right you are

06:44 <Guest8> I can't seem to flash the EMMc storage

06:45 <Guest8> Flashed a brand new light image on the SD card

06:45 <Guest8> dmesg readout gives error -110

06:45 <Guest8> As far as I understand it means that it got a timeout error

06:46 <zmatt> I've seen that happen when the eMMC has been worn out by too many writes

06:47 <zmatt> (which doesn't seem to be handled very well by this eMMC's internal firmware, causing it to go unresponsive upon attempted write)

06:47 <Guest8> That /would/ make a lot of sense

06:48 <Guest8> Is there a way to check that?

06:48 <zmatt> well, if the eMMC is initially detected but subsequently goes unresponsive, it seems almost certain that that's what's going on

06:49 <Guest8> I actually don't see it being detected at all in the dmesg

06:50 <zmatt> can you share (using pastebin.com or similar paste service) whatever is getting logged related to mmc1/mmcblk1 ?

06:51 <Guest8> Sure, give me a minute

06:53 <Guest8> https://pastebin.com/pkc4eW8S

06:53 <zmatt> well that ain't good

06:54 <zmatt> but could still be due to wearout, I've seen it manifest in different ways... in particular, I've seen it go completely dead by attempting to erase once it was already worn out

06:54 <zmatt> like what would happen if you were to attempt to reflash it

06:55 <Guest8> It did happen while trying to reflash the EMMc with a newer version

06:55 <zmatt> exactly

06:55 <Guest8> It immediately fails because it dosen't recognize the EMMc

06:56 <zmatt> like, the only other explanation would be actual hardware damage, but that would be unlikely unless something actually happened to the beaglebone that could explain that

06:57 <Guest8> The only possible sort of physical damage may be from power-outtages

06:57 <zmatt> I've never heard of that causing physical damage

06:58 <zmatt> we literally never perform clean shutdowns (we embed the BBB in a product that has an on/off switch that just cuts power)

06:59 <zmatt> (though we do reconfigure the eMMC into a pseudo-SLC mode, intended to increase endurance and reliability, including robustness against power failure)

07:00 <zmatt> (at the cost of losing 50% of storage space)

07:01 <Guest8> How do you configure it in pseudo-SLC mode?

07:03 <zmatt> this can be done using mmc-utils, though the procedure is somewhat delicate and tricky, and since it involves writing one-time-programmable registers (i.e. irreversible actions) I actually added a custom command to mmc-utils that performs a bunch of sanity checks and then performs the relevant steps

07:04 <zmatt> https://github.com/dutchanddutch/mmc-utils/commit/d550d3e2edaa4518af99ffdc101a32e93c5d37a5

07:05 <zmatt> of course we also put some effort into eliminating unnecessary writes to eMMC (such as logging)