chwe
-
Posts
1432 -
Joined
-
Last visited
Content Type
Forums
Store
Crowdfunding
Applications
Events
Raffles
Community Map
Posts posted by chwe
-
-
Todays special: What could probably go wrong? (It's in german, sorry for that..
)
Random company decides that people may want a app which may help you to fill your yearly tax form (that's the latest point where I probably would be skeptical but well, seems that people want to share all their tax data with *random company*). To save money, they decide to outsource the development for such a app to an contractor from india, contractor horribly fails to develop the app which is based on an AWS Bucket with public read/write rights for more or less everyone (as far as I understand the text correctly).. *random security researcher* tells the company that they do wrong, they think it's more appropriate to just ignore him than to see if there's a real issue until media covers up the story. One of the databases he found stored all the user credentials in plain text and all data uploaded to the app could be seen by *random security researcher*.. Just WTF?!??
If you don't have the resources to develop such an app on your own just don't do it! There's absolutely no excuse to bring such a shitshow to your end-users.. At least you need in-house capabilities to review the garbage you bought from your contractor.. How can you think it's somehow okay that user credentials for an app storing tax data of them in plain text? If you don't even get this point right you shouldn't be allowed to offer such a service. And the best part of it, the app is called 'Steuern59.ch' (tax59) because this app cost you 59CHF (~60$)..
reminds me to this one:
Edit:
just for fun, from the steuer59.ch website:
QuoteSicherheit
Ihre Daten werden sorgfältig aufbewahrt und nicht an Dritte weitergereicht. Nachrichten werden verschlüsselt und Ihr Login basiert auf einer 3-Faktoren-Authentifizierung, wie Sie es aus dem E-Banking kennen.
'free translation'
QuoteSecurity
Your data will be stored carefully and will not be passed on to third parties. Messages are encrypted and your login is based on 3-factor authentication as you know it from e-banking.
So I guess
1 factor: the guy who wants to see you taxdata needs Internet access
2 factor: he probably needs java script activated to brows through AWS services
3 factor: he needs a free AWS account cause those open AWS Buckets are only open to people having such an AWS account
Sorry for being sarcastic but that's the only way to deal with such news..
BTW: there is a java based tax app provided by the govt here in switzerland since years. You can load the form from the last year so that all your data is there and you just have to adjust it to the new year.. And it's not only Windows.. They provide an OS X and even a Linux version for it.. It's not that we've to submit those forms on tablets of stone...
-
1 hour ago, tkaiser said:
either need to use Maskrom mode (or how it's called exactly) which requires another host and USB
did you try https://github.com/rockchip-linux/rkdeveloptool?
And then a 'rkdeveloptool ef' might do it.. As far as I see this is plain c++ without any blobs so you could probably compile it on one of your arm boards as well..
chwe@chwe-acer:~$ rkdeveloptool ---------------------Tool Usage --------------------- Help: -h or --help Version: -v or --version ListDevice: ld DownloadBoot: db <Loader> UpgradeLoader: ul <Loader> ReadLBA: rl <BeginSec> <SectorLen> <File> WriteLBA: wl <BeginSec> <File> WriteLBA: wlx <PartitionName> <File> WriteGPT: gpt <gpt partition table> WriteParameter: prm <parameter> PrintPartition: ppt EraseFlash: ef TestDevice: td ResetDevice: rd [subcode] ReadFlashID: rid ReadFlashInfo: rfi ReadChipInfo: rci ReadCapability: rcb PackBootLoader: pack UnpackBootLoader: unpack <boot loader> TagSPL: tagspl <tag> <U-Boot SPL> -------------------------------------------------------
-
1 hour ago, DaveKimble said:
Apart from learning that "it doesn't have to be exactly 5.000 V", still no one has answered the question "What is the acceptable range?". And if the board doesn't tell you continuously what it has actually got, ANYTHING can attributed to a "power failure".
okay some starters:
1 hour ago, DaveKimble said:And if the board doesn't tell you continuously what it has actually got, ANYTHING can attributed to a "power failure".
Most boards don't have a PMIC which is capable of reporting the DC-In voltage. That's a fact and it's likely that this won't change. Those SoCs are mostly made for TV-Boxes, so the corresponding PMICs are developed with the assumption that the (TV-Box) boardmaker understands how proper powering works (e.g. barrelplug with a known good PSU), unfortunately SBC boardmakers decided that 'the RPi way of powering' (e.g. telling your customers that they can use their crappy phone charger with a crappy USB cable first before 'providing' a 'RPi approved' microUSB PSU which delivers 2.5A at 5.1V to overcome all those undervoltage situations). And that's why we're where we are.. having boards and dealing with undervoltage issues and trying again and again to convince people that undervoltage might be a reason why *random board* doesn't run as expected. If there would be a known 100% reliable way to detect undervoltage on each board we support, I'm sure @tkaiser would've implemented it the day he knows that it's working to save himself a bunch of time explaining again and again (and again) why *unexpected error* is likely related to underpowering. If the error would look exactly the same everytime, I'm sure such a script would also be shared, underpowering can have all sorts of funny errors and it's an error which can be avoided buy being not the scrooge trying to save 2-3 bucks by using random crap.
If random OPi board has an barrel-plug, buy opis PSU my OPi PC+ works over a year 24/7 without issues (headless). If you plan to stick random crap via USB to it probably with turned on wifi as well and then mix it with some heavy tasks on the CPU, even the one provided by xunlong my f*ck up and quit. If you've an microUSB powered one (e.g. OPi Zero etc.) go for good microUSB cables (e.g. ~AWG20 for powerwires - you can even find sometimes such cables on aliexpress). I use 1m cables branded 'Fonken' with AWG21 powerlines and AWG28 for datalines together with a Ikea handycharger named 'KOPPLA' and never got fooled by them (doesn't mean that nobody ever will be fooled by this combination but at least headless OPis worked quite well). If you consider to stick random SSDs or HDDs on a microUSB powered SBC over USB powered by the board itself, please reconsider your plan or use Y cables and ask @TonyMac32 when his "mezzanine" board:
will be available (he might need some testing not that he gets an article in the Register similar to team RPi https://www.theregister.co.uk/2018/09/11/raspberry_pi_poe_hat_issue/
). Or build your own GPIO powered solution.
Sometimes a simple Multimeter can help you to measure between GND and a random 5V pin on the pinheader to get a clue if the average voltage is okay.. Some of the shitty PSUs have low voltage spikes which aren't detected by an el Cheapo multimeter which makes debugging even harder. Some SoCs and their boards may be more sensitive to undervoltage than others. But the solution is simple don't save 2-3 bucks by going for the cheapest one you find... Btw. I assume that RPi's official PSU might also be considered as a 'good one' cause it was developed after team RPi realized that average phonechargers are to crappy to power their board reliable...
-
On 7/27/2018 at 9:18 AM, frank-w said:
@chwe if you make changes which fixing problems in mainline please send me the patches si can include them in my repo
Sure, I stepped down a littlebit with my BPi-R2 testings at the moment. There was/is a patchseries which should make it possible to define multiple CPU ports (https://lore.kernel.org/patchwork/patch/793311/). Unfortunately it never reached Linus tree and the patch doesn't apply at the moment. Maybe I'll have a look into it and see if this enhances the performance.
-
well https://github.com/armbian/testings will hopefully be part of tool/work-flow to avoid frustration with updates..
so that the average user doesn't need freezing his system in productive use-cases.
-
19 hours ago, DaveKimble said:
I would suggest that you all spend some time bringing the wiki up to date (for beginners) and maybe that will set more up people for doing testing. Sunxi ought to get their board naming up to date by going to "chip.RAM.ethernet" (e.g. H5.2GB.1000), and putting big heat sinks on everything.
first http://linux-sunxi.org/Main_Page != Armbian
linux -sunxi was formed around kernel development for Allwinner SoCs whereas Armbian tries to build an usable Ubuntu/Debian on top of their work (and on top of other Kernels for other SoCs). There's contribution by armbian members to the Sunxi wiki (e.g. @tkaiser quite often and others probably too) but as armbian, that's community work! so you can also contribute in case you're not satisfied by the current state.
13 hours ago, DaveKimble said:maybe in /sys/class/power/ , that says what voltage it actually has got?
unfortunately not.. or at least not for Allwinner H-series boards. The A series SoC have a PMIC with voltagecheck
13 hours ago, DaveKimble said:I appreciate why you need a good power supply and SD card, but if the mains is browning out there is no way that a transformer will give exactly 5.000 V,
you don't need 5.000 V 4.999 will also do...
A good powersupply will deliver around 5V under high load.. that's where a shitty powersupply may fail which ends in threads collected here: https://forum.armbian.com/forum/31-sd-card-and-power-supply/
last one:
may help
20 hours ago, DaveKimble said:for people who already know what they are talking about.
to find informations needed. Summarizations have to be written, as long as everyone expects that 'someone else' will do it nobody does it..
-
10 hours ago, tkaiser said:
https://github.com/BPI-SINOVOIP/BPI-W2-bsp --> (not so) surprisingly a 404!
Well, as long as it is marked private I hope that you get a 404. Otherwise GitHub would have serious security issues....
From the README.md we all see that the repo isn't 'polished' yet. It might need some time to write at least a short tutorial how to use those sources, especially due to the usage of a own compiler (see toolchain/asdk...).
my 2 cents towards GPL:
11 hours ago, tkaiser said:It's really that simple if you want to be GPL compliant. Hope it helps!
quote my self:
On 9/13/2018 at 3:49 PM, chwe said:from the GPL:
QuoteFor example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.
there's no need to make sources available on GitHub, indeed to be GPL compliant you've to release the sources for those using your images.
I would also prefer to have the sources publicly available on GitHub or GitLab or whatever git service you use before the board is shipped to the first customers but that's not what GPL forces you to do, that's more what development for such boars would make a way easier cause people dealing with it would have it easier to get some 'warm up' time before support requests for such a board arrives at *random distribution*.
In a perfect world, the documentation for this buildscript would be written before you release the first Debian/Ubuntu for it so that you could release the Image together with the sourcecode on GitHub, probably with a short reminder that this images are in early stage development and that errors problems must be expected at the moment (as armbian does with WIP and/or nightly images). But as shown by multiple boardmakers, including sinovoip, documentation has room for improvements.
-
7 hours ago, tkaiser said:
Lime2 for this.
btw. the table you mentioned would be happy to see some reports for the Lime2
-
Yesterday was about linus don't understand that some people might felt offended by his communications style. Today Coraline Ada's blog post about her year at github.
https://where.coraline.codes/blog/my-year-at-github/
It's somehow interesting... IMO they face both the same problem. Their communication style isn't fully compatible with the community they are in..
Coraline Ada:
QuoteIn speaking up like this, I felt like I was simply doing my job.
QuoteThe fact is, people need to know what my position on things are. And I can't just say "please don't do that", because people won't listen.
Coraline Ada:
QuoteStarting in December, in my weekly one-on-one meetings with my manager, we would review all of my written communication (issues, pull requests, code reviews, and Slack messages) to talk about how I could improve. It felt ridiculous but I went along with it, and did my best to address my manager's feedback and concerns. I even got a book on constructive communication and effective collaboration and reported in regularly on what I was learning from the reading. My manager seemed satisfied with my progress.
QuoteThis is more like the time I got out of kernel development for a while because I needed to write a little tool called "git". I need to take a break to get help on how to behave differently and fix some issues in my tooling and workflow.
For me this is somehow ironic. Both stand hard for their position which might not be a mainstream position which brings them quite often in trouble and they have to defend them-self. IMO, being 'inclusive' means that you accept both behaviors as long as they don't cross some lines of common sense. We've to accept that not everyone has the same opinions on several topics and what people assume as 'normal behavior' . Both, Linus and Coraline aren't fully what I would call average but in an inclusive environment both should find their place and felt somehow welcomed as long as they accept that maybe not everyone shares their opinion on a *random topic*. Cause actually, some of the features she and her team developed for github are really useful e.g. as she describes:
QuoteOne small feature that I shipped got a lot of attention: the first-time contributor badge. Basically, this was a little badge that a project maintainer would see next to the name of a new contributor to that project.
I think it's beneficial for bigger projects to spot new contributors either to have a closer look at the patch or just to congrats for the first contribution..
and the other one:
QuoteA couple of months later, I finished up a feature that I was very excited about: repository invitations. With repository invitations, no one could add someone else to a repository without their consent. Being invited to contribute to a repository resulted in an email notification, from which the recipient could accept or decline to join and even report and block the inviter.
IMO I would call it a bug that the invitation is only visible via email and not reminds you friendly as soon as you login with your github account but well.. Nothing is perfect..
But I also like that I could not be randomly added to groups I don't know or I don't want to be part of..
And my last two cents on this one:
QuoteGitHub touts its values, but has consistently failed to live up to them. Values that are expressed but that don't change behavior are not really values, they are lies that you tell yourself.
Well IMO those 'values' are written by HR and Lawyers (to ensure they can't be sued for them..
) and that's why a bunch of people inside a company don't give a fuck whats written there.. They don't grow organic it's more some sort of GMO, the majority accepts them cause they're cheap... e.g.
Quote"Positive Impact. We believe in making the world a better place through our work. Make GitHub a role model for the industry. Be a great neighbor and member of the community. Build inclusive culture."
-- From "Values" by Chris Wanstrath
smells like the marketing intern together with the law intern played a round of buzzword bingo to make an HR senior happy who had to make management happy cause they pushed him to write down some sort of 'what we are'.. If this stuff doesn't come from the people who do the work.. They don't feel responsible to follow these rules/values.. That's why IMO the companies HR department should have different 'values' than the development unit and they may have different values than the production department..
-
4 minutes ago, tkaiser said:
If clockspeeds increase by just running armbianmonitor in monitor mode then this is a clear indication that there's something wrong with cpufreq scaling
that's about the C-tool written to replace the bashscript for armbianmonitor -m. It's default update delay is 0.1 seconds.. that's why cpu freq. probably goes up.
25 minutes ago, tkaiser said:we need to accept that this forum starts to get flooded with recommendations like switching to conservative or powersave governor since Armbian doesn't care about such things
I don't think so. But I'm not sure what would be needed to fix this. E.g. zram got rolled out due to your testings (btw it's a great enhancement for my outdated 4GB RAM ubuntu notebook too, browser with 20-30 tabs aren't a problem anymore..
). What's needed to fix thermals in general (this seems even in mainline not fully clear - most thermals I saw are rather conservative)? Sum up the current state and we might find a working way to solve those issues.
I hoped that we can 'extract' such projects out of this one:
Didn't work that well. Maybe, I as trying to moderate this discussion was the wrong one, maybe it was just the wrong time to bring up this discussion back in january to april..
-
1 hour ago, tkaiser said:
Is not a great idea since this monitoring stuff generates some light load itself which might already trigger cpufreq rising. Better always use
armbianmonitor -m(this defaults to 5 seconds which gives a more realistic picture).
It may work with https://github.com/armbian/build/pull/824 by changing the defaults from 0.1 to something more realistic.. Otherwise you clearly see the impact (I guess SSH may create similar amount of work than the code itself..
) I tried it on a Tinker and an OPi PC+ and it worked without issues, never tested it on a big/LITTLE board etc.
-
https://lkml.org/lkml/2018/9/16/167
Linus takes a break (as it sounds at least for the rest of the 4.19 release) to fix his own behavior..
-
BTW. you might replace thermalpaste.. Mine was a bit dry (not terrible dry, but I guess they also go for the cheapest thermal past the can get..
)
It was accessible via USBA to UBAA cable (I had to kill two of my el cheapo microUSB cables,
)
My soldering tip died and resoldering of the UART (one cable directly soldered broke) was nasty.. It works somehow.. but not really well. I may need an HDMI screen as well to get a clue.. according to console out.. the emmc should be empty now.. so my next attempts will start soon.
-
1 hour ago, martinayotte said:
sounds like the alsaces.. We would be neighbors then..
didn't knot that ham rock is so famous.. I guess next time I've to prepare something else.. Maybe carrot cake? -
Does my 'kitchen vocabulary' sound fluent? Well not really..
For the recipes.. that's mostly a 2 or 3 liner in my 'private' lab-notebook e.g.
4 Haxen 2 Zwiebeln 2 Karotten + 300mL Wein + Bouillon --1.5h/150--0.5h/200-> feucht halten, am ende mit Honig glasieren 500g Mehl + 300mL Milch + 3 Eier + Salz/Muskat --30'-> portionenweise Kochen (Lochsieb nicht heiss werden lassen!) --> in Butter schwenkendocumentation during daily work needs 10-20% of my time (e.g. like this one: http://www.orgsyn.org/Content/pdfs/procedures/v95p0218.pdf).. So at home I'm not that motivated to do the same.. So you've no excuse..
-
1 minute ago, tkaiser said:
Delicious taste, bad 'look&feel'. Never did them again
well.. your pictures look better than mine.. Cause no woman ever tried haxe here I don't care that much.. And for the 'feel' part, it can depend on the flour you use.. You can mostly fix it.. e.g. let them cook a bit longer in water and/or roast them with butter for 2-3 min after its.. But the mess you've to clean after its.. different story (keep the parts you used wet helps..)
and from:
6 minutes ago, tkaiser said:Just had a look
I expect at least on recipe from you too...
BTW: Haxe is also delicious could as meat in a sandwich (together with mayonnaise mixed a bit of the sauce you get from cooking and some freshly cutted onions, and some tomatoes).
-
2 hours ago, botfap said:
We should probably start a H96 Max TV Box thread
Splittened this one..
2 hours ago, botfap said:I have discovered I have 3 different revisions of the board (dark blue, light blue and green pcb's) mfg in 2017-11, 2017-08, 2017-06. They all have 4G+32G+AP3255 stickers and are visually identical though I havent had chance to actually compare them yet
Compared to the pictures from the link:
2 hours ago, chwe said:it looks similar despite:
- SK hynx ram not 'ELPiDA'
- eMMC is not from samsung
- 'QL1720' not 'QL1716'
but both are labeled ZH-RK3399-TVBOX v2.0 even the date matches (2017-03-30).
2 hours ago, botfap said:You might have to zero the first 4M of the eMMC to force SD boot or I think there may be a reset button behind the audio port.
I may need an USB-C cable now..
All my computers are from the time before USB-C was standard..
2 hours ago, botfap said:On the serial console side, I think it may switch uart's when the bootloader starts but Im not 100% sure, it was a while ago and I see a lot of different boards
I guess this behavior can be fixed with some board config stuff (e.g. https://github.com/armbian/build/blob/9c15130252828c7b2ecf3c18b9c3a5dec8494bf9/patch/u-boot/u-boot-rk3399/nanopc-t4.patch#L62-L63 and https://github.com/armbian/build/blob/9c15130252828c7b2ecf3c18b9c3a5dec8494bf9/patch/u-boot/u-boot-rk3399/nanopc-t4.patch#L785 for the time before DT kicks in.. schematics would be nice..
but well.. I assume they don't change that much from default for cheap tv boxes)..
-
thread was splittened. See:
-
Quote
I'm depressing myself, back to sipping whisky and eating Spätzle <--- if that isn't comfort food, I don't know what is. @chwe we should start a "local foods" thread. Mostly because I'm hungry and want to learn some recipes.
challenge accepted
Todays special:
Schweinshaxe mit Knöpfli
Knöpfli is Switzerlands pendant to 'Spätzle'. Schweinshaxe (Ham hock), and for those not knowing them:
here we go..
user: ~$ time meal.sh real ~2h30min user ~40min sys ~2hFirst we start with the Haxe. We rost them of both sides for around 2-3mins in a metal bowl on the hotplate.
First on the side where the bones are smaller, then on the other side. We add 2 roughly cutted onions and carrots and let them rost for around 5mins before adding a glass of vine.
We let the vine now reduce a little bit before adding bouillon (as much as needed so that they are not completely covered). They're no cooked for around 1h30mins at 150°C and the last 30mins at 200°C in the oven (use the liquid to 'moisten' them every 15min).
Knöpfli:
We mix 500g flour with 300mL milk (or 150 milk 150 water), 3 eggs a bit of salt (one teaspoon) and if you like a bit of nutmeg. As soon as it is well mixed we let it rest for 30mins.
A big pan is filled with water and heated up until it boils. The knöpfli dough is then pressed through a 'perforated sieve' (there are special ones for 'Knöpfli' but mine comes from a steam cooking set and works also well
).
as soon as they swimm (normally immediately), they can be skimmed wit a normal sieve and collected. If you want you can roast them now in butter and add some roasted onions on top but it also works well without, or as I do add cheese on top. Similar you can add some honey on top of the 'haxen' for the last 5-10minutes for a nice crust (I guess the germans adding Soda in the end for the crust, maybe the Bavarians here knowing it?
).
In the end, it should look somehow like that:
Things you need (for ~4 persons):
- 2 onions
- 2 carrots
- 4 haxen
- 500g flour
- 3 eggs
- 300mL milk
- a glass vine, a bit of bouillon, salt, oil (butter)
-
5 minutes ago, botfap said:
Also is there a 2nd uart header on the board? I seem to remember there being 2 serial headers but I may be wrong
There are two.. Unfortunately they have a 2mm pitch not 2.54 so soldering is nasty..
Soldered my uart to the one labeled debug and the first messages also show that this seems to be the right one. But I might solder to the other uart as well. At least firstrun doesn't came up.. otherwise the FS would be expanded.
BTW did you load something to the eMMC? cause there's a short bootlog when I boot up without SD-Card (the binary creation date changes, so I assume there's a difference)...
-
On 8/25/2018 at 3:11 PM, botfap said:
Edit: H96 Max Android image boots without problem and everything seems to work fully except the WiFi chip which isnt detected. From @tkaiser this may be because its looking for a LTM8830
depends on version you look at.. E.g. this one http://freaktab.com/forum/tv-player-support/rockchip-based-tv-players/rk3399-devices/645643-h96-max-rk3399-tv-box-4-32gb/page2
has an the AP6255.
Unfortunately, UART get's silent quickly after boot and I still don't understand exactly why.. using mainline u-boot. Well, let's see what will work with @ayufans u-boot
-
https://bugs.python.org/issue34605 @Igor shall we rename armbians master branch too?
I don't really understand what this 'clean ups' try to achieve? Does someone really felt offended by master/slave terminology in programming languages? Maybe I miss something here..
we could name it Gru:
well, maybe @zador.blood.stained might not think that GRU is appropriate..
Spoilerwhen I started as a lab technician we called them 'condoms for cats' - or 'condoms for x' (replace x with someone you want to piss off)...
-
Ce sont pour les Québécois sympathiques.
I had to learn most of the console work too when I did more stuff on SBCs, as long as they're able to read and you see progress, I hope this forum is also a place for 'linux beginners'.. Debian/ubuntu can be a hard change if you come from windows-world. But especially for python etc. a linux shell is a way more userfriendly than windows once your familiar with the shell.. Actually I hope that a beginner writes tutorials after he got his first successful little python project running, I think they're best in writing a tutorial for other beginners cause they probably face the same issues when they start. The more experienced once often fail to explain stuff for beginners (remembering my quantum mechanics classes where youtube was often the better source of information than the weekly lecture )..
Spoilerbest explanation I ever had for group theory...
3 profs tried to explain it, nobody did it better... at least for the calculations..
-
On 9/12/2018 at 2:42 PM, tkaiser said:
But instead we should emphasize that on our officially supported boards w/o Ethernet a remote console sits on the MicroUSB port waiting for the user to login.
never tested but:
https://github.com/armbian/build/blob/8ea0b98845e9759266108e699d11aefd55cac882/config/boards/bananapim2zero.csc#L7 even on the CSC it should be there..
Maybe we should point it out in the documentation e.g.
QuoteHow to login?
Login as root on console (HDMI / serial) or via SSH and use password 1234. You will be prompted to change this password at first login. You will then be asked to create a normal user account that is sudo enabled (beware of default QWERTY keyboard settings at this stage). Please use this tool, to find your board IP address.
e.g.:
For the following boards:
- nanopineoplus2
- nanopineocore2
- nanopineo2
- nanopim3
- nanopifire3
- bananapim2plus
- orangepizeroplus
- bananapim2zeroSerial console can be accessed via USB with gserial (link to a little how-to to work with gserial).
On 2/9/2018 at 1:13 AM, iwl said:Looks like somebody has modified Armbian for bpi zero allready for that image, so it's Armbian with Ubuntu on top?
the board is csc supported, even with a patch which shouldn't harm other boards and make sure that your boards survives a kernel and/or u-boot update. https://github.com/armbian/build/blob/8ea0b98845e9759266108e699d11aefd55cac882/patch/u-boot/u-boot-sunxi/add-bananapi-bpi-zero.patch#L184
you've to build it on your own but besides that, the board should work and survive updates (mine is collecting dust since months due to no need for it at the moment, never did long run tests to keep an eye on durability).
OPi Zero does not get an IP address
in Allwinner sunxi
Posted
nope it can be either accessed via USB-UART adapter (that's what those 3 pins near to RJ45 are for.. Or with armbian via USB gadget mode.
yep
hmm that sounds strange.. But with your previous armbian it worked properly on the same router via DHCP?