Werner changed the topic of #armbian to: armbian - Linux for ARM development boards | www.armbian.com | Github: github.com/armbian | Commits: #armbian-commits | Forums Feed: #armbian-rss | Type 'help' for help | Logs: -> irc.armbian.com
<xwigg> oy
IgorPec has quit [Ping timeout: 240 seconds]
ChriChri_ has joined #armbian
ChriChri has quit [Ping timeout: 246 seconds]
ChriChri_ is now known as ChriChri
livingsilver94 has joined #armbian
livingsilver94_ has quit [Ping timeout: 256 seconds]
<Tonymac32> ahoy!
torv has quit [Remote host closed the connection]
torv has joined #armbian
oida has quit [Remote host closed the connection]
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
oida has joined #armbian
<nekomancer[m]> ahoy
<lanefu> chips!
<lanefu> okay I thin i have my fail2ban stuff ironed out
<lanefu> woohoo we be banning
<chewitt> @nekomancer[m] are you building from my branch? .. did you enable CONFIG_RTC_DRV_PCF8563=m in defconfig?
<chewitt> because it works for me :)
<nekomancer[m]> no, i miss this point. retry, thank you.
<nekomancer[m]> (late for today, will build tomorrow)
bigbrovar has quit [Ping timeout: 265 seconds]
bigbrovar has joined #armbian
Net147 has quit [Quit: Quit]
Net147 has joined #armbian
<lanefu> I forgot who was sleuthing with me last night.. but Kobol is off the hook as helios4 is 32bit, and arm64.gz packages are being pulled :)
<Miouyouyou> lanefu: I think the issue might be from OpenMediaVault 4 images
<Miouyouyou> It enables 2 timers and 2 cron activities that do apt updates and upgrades
<lanefu> Miouyouyou: OMV was a suspicion
<lanefu> let me check my helios omv
<Miouyouyou> While the OMV 5 image takes the latest Armbian images and add the necessary, the OMV 4 is premodified one.
<Miouyouyou> I don't think it's the one provided with Helios. They tend to use official Armbian images, so they should be clean, yeah
<lanefu> bastards
<lanefu> okay and so the OMV4 i have installed via the armbian-config script probably doesnt have that either
<lanefu> k i can't math in my head that well.. you think that's causing it to run up to ... every 10 minutes?
<Miouyouyou> every 10 minutes, I don't think so
<lanefu> what abotu only 40 times a day?
<lanefu> i had different frequencies i calculated
<lanefu> man so that'll be a fun bug report to open with OMV
<Miouyouyou> I'll have to check systemd syntax, because I don't know why one is randomized with 12H and one is randomized with 60m
<Miouyouyou> If they're the culprit... The best way would be to test this directly on a board
<Miouyouyou> Who has an Odroid XU4 ?
<lanefu> i'm asking a buddy of mine who is rich in Xu4 to install
<Miouyouyou> Alright. Tell him to check the network activity of that board... Or ask him his IP and check in the logs.
<lanefu> ha yeah i know his dyndns host, so should be easy to check
bigbrovar has quit [Quit: Quit]
bigbrovar has joined #armbian
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
<lanefu> finally image downloades are showing up in the bandwidth reports :) im crashing early tonight.. cya
<lanefu> thanks for sleuthing Miouyouyou
_whitelogger has joined #armbian
martinayotte has quit [Ping timeout: 260 seconds]
fizikz has quit [Quit: Leaving.]
fizikz has joined #armbian
baffo32 has quit [Remote host closed the connection]
baffo32 has joined #armbian
<Werner> Good morning
<Tonymac32> Good morning
livingsilver94 has quit [Ping timeout: 264 seconds]
livingsilver94_ has joined #armbian
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
_whitelogger has joined #armbian
sassinak-work has quit [Ping timeout: 256 seconds]
sassinak-work has joined #armbian
<bigbrovar> Morning guys.
torv has quit [Ping timeout: 240 seconds]
torv has joined #armbian
_whitelogger has joined #armbian
igorp_ has joined #armbian
<Werner> Morning Igors
<igorp_> good morning
<Werner> Always nice to have a spare Igor
<igorp_> sorry :)
<Werner> No worries :D
<PPA> Good morning
<igorp_> morning
IgorPec has quit [Ping timeout: 256 seconds]
<igorp_> it seems lanefu manega to fix repository abuse
<Werner> Did he block the source?
<igorp_> it bans those ips for 7 days
<Werner> Any information through whois.sc for those ips?
<igorp_> i didn't check
<Werner> Might be worth a try
<Tonymac32> my understanding the IP's were very random
<Werner> If they were random it would not make sense to explicitly ban them :P
<Tonymac32> they repeat offended of course, but the individual IP's have no connection
<Werner> I dont know them so I cannot assume that.
<Tonymac32> the theory is it's someone's image for a product with a bad script
<Tonymac32> end user likely unaware
<Werner> Yeah
<igorp_> fail2ban is proper solution IMO
<Werner> Worth a try. Limits the knocks
<Miouyouyou> However, angry end user might complain on the forum... and might provide the name of the product
IgorPec has joined #armbian
<igorp_> yeah, will see
<Werner> Only if they know the relation to armbian
<Tonymac32> omv people might not, as an example
<Miouyouyou> Then they'll complain on the product's forum or the makers email, which might ricochet onto us
<igorp_> its possible that this is some kind of lame mirror making
<Miouyouyou> The other solution is force some kind of "telemetrics" onto the users to get their device brand, which is dodgy
<igorp_> i know for at least one from the era we didn't have ipv6
<Miouyouyou> Given the disparities of the IP and the sheer number, it's less likely that it's a mirroring system going wrong.
<igorp_> anyway, its an abusive behaviour
bigbrovar has quit [Ping timeout: 240 seconds]
<Miouyouyou> Yeah
<Werner> telemetry might bring answers faster but I wont risk any good reputation for adding this to armbian
bigbrovar has joined #armbian
<Werner> Leave that to Redmond companies :P
<Miouyouyou> Windows WokCheap Edition
<igorp_> yes, we are going to change hosting soon and we don't know exactly how services will be defined
<Werner> I already pushed some ideas for locations depending on use (hetzner auction server for cheap storage and traffic and ovh cloud vps for reliable hosting with decent uptime guarantee)
<Werner> Btw check your query igorp_ if you dont have already
igorp_ has quit [Quit: Leaving]
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
<IgorPec> werner: it looks like it works
<Werner> Yes, just verified.
<IgorPec> tnx. i had some troubles reconfiguring
<Werner> You can use the nickserv module to store your credentials but you dont have to. Simply use the /msg nickserv identfy command and it will hold as long as the bnc stays connected to freenode which ideally should be a couple of month
<Werner> The bnc will use the /away command automatically 60 seconds after you disconnect from it to let others know that you are gone.
<IgorPec> alrigh, i will not spam anymore then :)
buZz has quit [Ping timeout: 256 seconds]
buZz has joined #armbian
buZz is now known as Guest14883
archetech has joined #armbian
bigbrovar has quit [Ping timeout: 240 seconds]
bigbrovar has joined #armbian
bigbrovar has quit [Read error: Connection reset by peer]
sassinak-work has quit [Ping timeout: 246 seconds]
bigbrovar has joined #armbian
sassinak-work has joined #armbian
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
bigbrovar has quit [Ping timeout: 264 seconds]
macc24 has joined #armbian
bigbrovar has joined #armbian
macc24 has quit [Quit: WeeChat 2.8]
Strykar has quit [Quit: /quit]
Strykar has joined #armbian
macc24 has joined #armbian
dddddd has quit [Ping timeout: 240 seconds]
sassinak-work has quit [Ping timeout: 246 seconds]
sassinak-work has joined #armbian
livingsilver94 has joined #armbian
livingsilver94_ has quit [Ping timeout: 240 seconds]
bigbrovar has quit [Ping timeout: 240 seconds]
bigbrovar has joined #armbian
baffo32 has quit [Ping timeout: 260 seconds]
baffo32 has joined #armbian
phantomse[m] has quit [Quit: Idle for 30+ days]
archetech has quit [Quit: Konversation terminated!]
archetech has joined #armbian
ced117 has quit [Ping timeout: 256 seconds]
ced117 has joined #armbian
ced117 has joined #armbian
ced117 has quit [Changing host]
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
<lanefu> Werner: the abuse IPs were all IPs of different residential ISPs. Mostly USA
<Werner> Hm
c0rnelius has quit [Ping timeout: 256 seconds]
<xwigg> do you know which devices? nmap ident
martinayotte has joined #armbian
c0rnelius has joined #armbian
c0rnelius has quit [Ping timeout: 260 seconds]
c0rnelius has joined #armbian
<lanefu> Yeah i dont think nmaping their firewalls is going to buy us much
<lanefu> Miou has a pretty good running theory right now around OMV4 images for Odroid Xu4
<lanefu> IgorPec: ill cut an RC branch later this morning and update forum.
baffo32 has quit [Ping timeout: 260 seconds]
baffo32 has joined #armbian
<lanefu> Miouyouyou: well xu4 omv image theory is out.... xu4 is 32bit and both 32bit and 64bit stretch indexes are being downloaded
<xwigg> /nag I'm still in favor of http vs https, isp proxy can cache
Guest14883 has joined #armbian
Guest14883 has quit [Changing host]
Guest14883 is now known as buZz
<lanefu> xwigg: yeah i see the point now. I'll tinker with disabling redirect
<lanefu> conversely there's advantages to HTTPS.. as that means http 2.0 is supported and can do multiplexing
<lanefu> ..although i dont know if apt supports that :P
<ArmbianHelper> although i dont know if apt supports that :P [en~>eng]
<lanefu> xwigg: http://apt.armbian.com
<xwigg> works
<lanefu> IgorPec: any opposition to disabling redirect to https, for dl,apt,beta .armbian? regular site and docs should stay as ssl
<xwigg> get 404 on favicon.ico
<lanefu> lol
igorp_ has joined #armbian
<xwigg> and set expires max on deb files
<lanefu> xwigg: try favicon now
<lanefu> xwigg: can you give me some guidance on expires max?
<xwigg> okay, nice and gray
<lanefu> lemme try differnt icon
<xwigg> heh
<lanefu> k hopefully better now
<lanefu> wonder if we can find some best practices specifically for apt
<xwigg> arm
<xwigg> though apt.armbian won't have any use for icon ;)
<lanefu> lol you're the one that complained
<igorp_> o, do we have an icon there?
<lanefu> igorp_: i made an include file so we can put one there
<lanefu> igorp_: in case you missed... any opposition to disabling redirect to https, for dl,apt,beta .armbian? regular site and docs
<lanefu> should stay as ssl
<igorp_> you mean to disable automatic redirection?
<igorp_> we have a Jira to do that i think, sure
<igorp_> do it
<lanefu> haha okay
<lanefu> apt. done :) doing others now
<igorp_> don't fix too much :)
<xwigg> debian sets cache to month
<xwigg> (ftp.debian.org)
<lanefu> igorp_: don't worry i'm really good at being 80% done with things
<igorp_> than this is perfect ;)
<xwigg> location ~* \.deb$ {
<xwigg> expires 1M; // 1 month
<xwigg> add_header Cache-Control public; // invisible proxies
<xwigg> access_log off;
<xwigg> }
<lanefu> 1M makes me quesy. i guess proxies still check for file size changes regardless?
<xwigg> all debs are versioned ;)
<xwigg> gzip off;
<xwigg> proxies honour cache settings, at least they should
<lanefu> Ahhh okay.. right.. so .deb is never gonna change :P
<lanefu> makes the sense
<xwigg> yup release etc. do but point to different
<xwigg> the only thing that changes is armbian-...-20.08-trunk, but I do not now if you can pull that
archetech_ has joined #armbian
<xwigg> now=know
<igorp_> tonymac32: so we can go with meson64 with 5.7.y "just" audio will be missing?
archetech has quit [Ping timeout: 256 seconds]
igorp__ has joined #armbian
<lanefu> xwigg: http://ix.io/2rYt
<igorp__> werner: alter nicks are staying online now
<Werner> ?
igorp_ has quit [Ping timeout: 240 seconds]
<igorp__> it seems i can't reuse igorpec
<Werner> You aren't connected to the bnc which holds the nick atm
<igorp__> ahaa, let me check
igorp__ has quit [Client Quit]
igorp__ has joined #armbian
igorp__ has quit [Client Quit]
<xwigg> lanefu; looks good, please set server_tokens off; in http
<IgorPec> werner: works now i guess :)
<lanefu> whats wrong with server tokens
<lanefu> all the cool kid sare doing it
<Werner> awesome
<lanefu> s/kid sare/kids are
<ArmbianHelper> lanefu meant to say: all the cool kids are doing it
<xwigg> lanefu: automated scripts from eastern europe/china in case 0-day vulnerability
<lanefu> tell me more
<lanefu> ohhh
<lanefu> nevermind got it
<xwigg> ideally something vulnerable as PHP/wordpress on different server
<lanefu> yeah we're working on a new solution for the archive/mirror stuff.. and things will be more segmented then
<IgorPec> yes, we can have web and forum separated from others
<IgorPec> should
<IgorPec> i think this is not a problem
<IgorPec> i hope next week we will be able to start with transfer
<xwigg> okay, gotta go shoppinh
xwigg has quit [Quit: bye]
<lanefu> IgorPec: do we want to cut RC branch soon? or do you want to do some more work first...
<IgorPec> i plan to scan irc log to see what we said we will fix
bigbrovar has quit [Ping timeout: 240 seconds]
bigbrovar has joined #armbian
<IgorPec> and those larger changes, like shell we move sunxi 5.4.y to legacy and 5.7.y to corrent now?
<IgorPec> i mean today
<IgorPec> also rockchip, current to 5.7.y ?
<lanefu> dolphs was making progress on with dev sunxi 5.8, so makes sense to me
<lanefu> my sunxi 5.5 boxes have better uptime than my 5.4
<IgorPec> 5.4.y is really old
<IgorPec> the question is are we ready for that?
<IgorPec> we move ... and than two weeks only bugs
<lanefu> martinayotte: is probably the best for casting that vote?
<Werner> Since there is no reason to keep 4.19 I think it is fine to move 5.4 to legacy. We wont loose anything
<IgorPec> we can also plan to go out with a release when i am back in office, so i will be able to assist better
<IgorPec> 4.19.y might be good for some special cases. i have seen people still using it
<IgorPec> but they can switch to that kernel and remain there
<IgorPec> in case of urgency
<IgorPec> check here https://freenode.irclog.whitequark.org/armbian/2020-07-04 we talked about ...
<IgorPec> rockchip we have confirmation to move current to 5.7 too
<IgorPec> but we have no Jiras for this move
<Tonymac32> igorpec that is all I see as a problem. Let me review the other known troublesome topics like C2 and it's shitty USB configuration
<IgorPec> C2 USB got worse in 5.7.y ?
<Tonymac32> Not that I know of, but it is never consistent, from kernel to kernel it works or deosn't. I have not verified it yet
<lanefu> Tonymac32: I had HerculeP test a 5.7 kernel and USB was fine.... still no sound..
<Tonymac32> thanks lanefu
<IgorPec> usb is random issue
wi has quit [Ping timeout: 260 seconds]
<Tonymac32> The C2 is a victim of being first, it has a lot more non-standard stuff going on, so that GPIO enable for the USB power is a moving target
<IgorPec> usb problem is here since ever ... let's forget about that. sound?
<IgorPec> i read about someone change a lot , right?
* lanefu taking a load of brush to the dumb.. bbiab
<lanefu> s/dumb/dump
<ArmbianHelper> lanefu meant to say: * lanefu taking a load of brush to the dump.. bbiab
<IgorPec> and our window is closing, so
<Tonymac32> I will work on it today to verify the other boards, and clean up the sound some more to retry reverting that sound patch that broke everything
<Tonymac32> I spent the last weeks building test boxes for my company's products that I designed, cost me all my other activities
<IgorPec> sure, we have to work with what we have / can
<IgorPec> my wife will also kill me for siting here :)
<Tonymac32> On a similar note, how is the SD multiplexing thing going?
<Tonymac32> I had reworked the old ones, but it seems not worth forwarding to lanefu I think
<IgorPec> i only briefly checked, a prototype will be made
<IgorPec> SD switcher on old is broken, console/power switching can be used
* Tonymac32 is currently filling out documents for home schooling the children
xwigg has joined #armbian
<Tonymac32> ok, I can work on that. I'll make sure I have all the fixes in place, I saw the display needed work too
<IgorPec> yeah, i think its better to wait for v2
<IgorPec> v1 was made for educational purposes :)
* Tonymac32 knows how hard it is to roll something like that out
<Tonymac32> my test machines will be handled by people all over the world, and I'm kind of nervous about it
<IgorPec> yeah, hw design is different world from my perspective
<IgorPec> or let's say from sw support side
<Tonymac32> especially if you're going to tie in software
wi has joined #armbian
<Tonymac32> I did the whole stack in this case, chose the microcontroller, developed the firmware/algo, and then the packaging
<Tonymac32> There is no Linux on board. :P
<IgorPec> and after you order first batch you discover fatal problem :)
<Tonymac32> There are always problems. I have 2 modifications to the board, yes haha
<Tonymac32> Maybe I'll build a NanoPi Duo into something next, only if there needs to be some special requirements for data logging/uploading I think
<Tonymac32> Our local schools are saying no recess, no gymn class, no moving out of the classroom without masks, etc etc etc etc etc
<Tonymac32> maybe no transportation to/from the school
<xwigg> what area?
<Tonymac32> Michigan
<Tonymac32> North end of the Detroit region
<xwigg> I thought Detroit was abandoned, covid regulations here (NL) are somewhat loosened
kolla has quit [Quit: %fog relay%]
redliondj has joined #armbian
<Tonymac32> haha the city itself has a very low (comparatively) population from the "Old Days"
kolla has joined #armbian
<Tonymac32> the neighborhoods near it that worked in automotive are very bad, abandoned, etc
<xwigg> yeah, sad thing, but GM really stuck too long on cubic inches
<Tonymac32> lol that wasn't the problem, it was quality and cost
<Tonymac32> a GM pushrod engine is still competitive with a Ford dual overhead cam model, in power and fuel efficiency
<Tonymac32> despite the displacement differences
<Tonymac32> and management. Oh my goodness management
<Tonymac32> their management/marketting is terrible
<Tonymac32> In any case, you get 30 minutes away from the city by car and it is a normal area
livingsilver94_ has joined #armbian
livingsilver94 has quit [Ping timeout: 240 seconds]
<Tonymac32> (I have, in the past, worked in vehicle emissions testing and fuel management, so my familiarity with engine systems and their relative pros/cons is due to that))
macc24 has quit [Quit: WeeChat 2.8]
<IgorPec> lanefu: large code drop in the PR :)
<xwigg> ah, I just did some tinkering in the past, now a lease bucket
<Tonymac32> I drive things until they fail so completely they cannot be slavaged
<Tonymac32> s/slavaged/salvaged
<ArmbianHelper> Tonymac32 meant to say: I drive things until they fail so completely they cannot be salvaged
<xwigg> hehe
<Tonymac32> and looking at all these new cars with their leaky European turbo designs... well... :P
* Tonymac32 had a bad experience (ok, many bad experiences)
<Tonymac32> My next vehicle probably won't be domestic or Euro looking at the market and the known quality problems. I'll probably wind up with a Toyota
<xwigg> yes, emission requirements results into tiny 1L blocks which are too small
<Tonymac32> GM has a 1.2 and a 1.3L 3-cylinder, it gets worse fuel economy than my 2.5L Mondeo
<xwigg> fiat blocks?
<Tonymac32> hmmm, it is GM so probably Opel
<Tonymac32> from before the shuffle
<Tonymac32> the 1.4L 4-cylinder they have is hte smallest useful engine in their lineup, and it is a quality nightmare
<Tonymac32> also an Opel
<xwigg> opel is psa nowadays... very bad quality
<Werner> Um.... IgorPec why am I assigned for AR-366?
<ArmbianHelper> AR-366 [Task] "Move rockchip/64 current to 5.7.y" reported by Igor Pecovnik at 2020-07-18. Status: To Do
<Tonymac32> My least efficient engine, and the one to give me the most serious trouble: https://en.wikipedia.org/wiki/Ford_Cologne_V6_engine
<Tonymac32> the 4.0 overhead cam version
<IgorPec> werner: if you got time, otherwise i will do it later
<Werner> Okay
<IgorPec> now i have to report to my supreme commander ... be back soon
<Werner> GL HF
<Tonymac32> hahahha
<Werner> Remember. No respawn if you fuck up :P
<Tonymac32> xwigg the engineers were drinking too much https://en.wikipedia.org/wiki/K%C3%B6lsch_(beer) when they designed the timing chain set
<xwigg> old bmw are nice
<Tonymac32> like 80's old?
<Tonymac32> yes
<xwigg> early 90s without i-drive also
<Tonymac32> of course I don't think we get all the models here
<Tonymac32> but yeah, the newer ones, even the people who like them here say: "You lease it, then, instead of taking it back to the dealer, save time and deliver it directly to the scrap yard"
<Tonymac32> I like the cars, they fit into segments the American manufacturers won't fill. Muscular sedans for example
<Tonymac32> (The Charger does not count, it drives like a barge)
<xwigg> yup, spoke to BMW engineer ~2000s, they actually fix the door hinges to not exceed hinges
<Tonymac32> oh my
<Tonymac32> :-P
<Tonymac32> That seems like a large number of actuations though
<xwigg> yup, but to spend time downgrading something is appaling
<Tonymac32> that is the basic premise of engineering
<Tonymac32> do as much with as little to meet the specified requirement
<Tonymac32> they probably saved 0,01
<xwigg> they earned a lot more, new cars....
<xwigg> but an old mercedes G is fine too
<lanefu> Tonymac32: IgorPec k i'm back.. my brain is a little slow, if you guys just wanna tell me what to do, i'll do it :)
<Werner> First things first. Get some brain food -> coffee :P
<Tonymac32> yes
<lanefu> bagles and coffe, check
<xwigg> when is aug release?
* Tonymac32 remembers he has lox and bagles...
<lanefu> xwigg: we cut RC branch today-ish.. then test and harden and release when its good. or good enough
<lanefu> aka.. sometime in august hwen it feels ready
<lanefu> obvosuly sooner is better than later
<lanefu> hopefully we can aim for better than good-enough
<xwigg> okay, H3 seems stable, though only headless
<lanefu> coordiantg with community for testing is the harder, more critical part
<Tonymac32> right
<xwigg> yeah well, most people buy the stuff for production use I guess
<Tonymac32> lanefu I owe a test of the newer xu4 kernel on BOINC
<Tonymac32> were the changes to the DMC merged?
<lanefu> Tonymac32: yeah long time ago
<Tonymac32> I figured, I'll pop that in while I'm waiting on a build
<Tonymac32> I assume a nightly is available?
<lanefu> apparently not. i hsould have an image handy stand by
<Tonymac32> (also bought my daughter an i7-3770 Dell from eBay for $150, it will maybe have other jobs in the background) :P
<lanefu> j/k. i dont. lemme build one
<Werner> martinayotte, do you want a project affiliation cloak? To say @armbian/developer/martinayotte where the middle part is exchangeable to what fits best (like maintainer, contributer, donator and so on...).
<Werner> I cannot randomly assign them, I have to ask for permission :)
<xwigg> https://forum.armbian.com/topic/14148-2-opipcs-same-boot-different-results/ OpiPC+ running stable, though capped at 840 (powered by old nokia adapter)
<Werner> xwigg, since you are contributor you can have voice as well as a project affiliation cloak if you like to.
* lanefu wants his armbian tshirt
<Werner> Wait, we have tshirts? #shutupandtakemymoney
<xwigg> werner: i do not exactly know what that means
<Werner> xwigg, voice or cloak? :P
<lanefu> haha i wish we did.. would be a good fundraiser maybe
<xwigg> both
drobo_00 has quit [Read error: Connection reset by peer]
<lanefu> xwigg is better had nginx hardening than irc
<Werner> Aight, The voice thing is mostly cosmetic since you are "ranked higher". It also gives permission to talk in the channel if the moderation flag is set (in case of emergency spam bots or something).
<lanefu> /s/had/at
<Werner> The project affiliation cloak replaces your hostname or ip address (in your case an IPv6 address) with something like @armbian/contributor/xwigg
drobo_00 has joined #armbian
<Werner> It kind a hides your source but the cloak is not designed to hide it because there are still ways to determ your source address. Though they show your affiliation to a project
<lanefu> it's like a tattoo, but for IRC
<Werner> Yeah, kind of
<Werner> comes with awesomeness +2 boost :P
<martinayotte> Werner: "contributor" is the best fit for me ...
<Werner> martinayotte, alright then. I'll take care for it
martinayotte has joined #armbian
martinayotte has quit [Changing host]
<Werner> Done
* lanefu having trouble with customize image finding /boot/armbianEnv.txt
<Tonymac32> latest 5.7 is spitting out an emmc patch, checking
<Tonymac32> meson64
<Werner> xwigg, both though requires you to have your nickname registered with the freenode service "nickserv"
<ArmbianTwitter> @armbian (armbian): @leonanavi @github Congratulation! BTW. Ours made it deep down :) https://t.co/PbRC2lpqCm (17s ago)
* Tonymac32 wants his armbian t-shirt as well
<xwigg> ow, I think I'm fine as it is
* xwigg votes for underpants!
<Werner> Its no big deal. Let me know if you change your mind :)
<xwigg> thx
<IgorPec> alright
<Werner> Armbian tangas with the subtext (RK3399) because they usually running hot :P
LanDi has joined #armbian
* Werner looking over to teespring...
<Tonymac32> XD
<IgorPec> tshirt, ok. let's push things for the RC together :)
<Tonymac32> goodness how much changed between 5.7.7 and 5.7.9
* Tonymac32 knows where to look, just surprised at build time
<Werner> First of all the number
<xwigg> sunxi-current can be bumped to 5.4.52
<Werner> 5kft usually pushing upstream patches frequently but you can do that too via PR
<IgorPec> sunxi-current will soon be 5.7
<IgorPec> legacy will remain for 5.4
<Werner> Still should receive upstrema patches ;)
<Werner> s/rema/ream
<ArmbianHelper> Werner meant to say: Still should receive upstream patches ;)
<IgorPec> correct
<xwigg> ok-- perhaps for another time, why 5.7 over 5.4 (LTS)
<lanefu> xwigg: because we're usually backporting patches from newer kernels just to make support work, keeping LTS kernels doesn't buy the SBC community much
<IgorPec> because this is arm
<xwigg> hehe
<lanefu> i do think there should be an optional uber conservative LTS offering from us at standpoint that just focuses on server
<Tonymac32> hmmm, 3-color beer can sleeves are too expensive, would look nice with our polygon penguin though
<lanefu> i want a tshirt of nicod wearing an armbian tshirt on his bike
* xwigg thinks there's already too much considering userspace might change as well
<Tonymac32> LOL
<lanefu> speaking of which. i'm a YEAR late on merging his armbian backgrounds into the desktop
<IgorPec> year late, that's nothing :)
<lanefu> haha
<Werner> lanefu, let me know if you manage to create such a shirt :D
<IgorPec> is there any cheap way to outsource complete operation?
<IgorPec> i have no intention to ship tshirts around :)
<Werner> teespring or spreadshirt are known manufacturers
<Tonymac32> yeah, teespring does the whole deal
<IgorPec> "trending now" face masks :)
<Tonymac32> sigh
LanDi has quit [Read error: Connection reset by peer]
<Tonymac32> lol
<Werner> Armbian facemask? For LTS protection
<Tonymac32> LOL
<Tonymac32> Armbian onesies, for the babies doomed to drink too much coffee and never sleep
<lanefu> hahaha
<lanefu> actualyl an armbian facemask would be badass
macc24 has joined #armbian
<lanefu> hey dumb question
<lanefu> have xu4, and ebin u-boots recently started using boot.ini?
<lanefu> and no armbianEnv.txt?
<ArmbianTwitter> @leonanavi (Leon Anavi): @armbian @github Congrats! This is super cool. (1s ago)
<IgorPec> there is no armbianEnv support when boot.ini is used
<lanefu> Werner: nice
<IgorPec> there was an idea to move to mainline with xu4 once, but i guess troubles, compatilitiy issues
archetech has joined #armbian
<Tonymac32> lanefu once this is set up it will take 5 hours or so to really tell
archetech has quit [Client Quit]
archetech_ has quit [Ping timeout: 264 seconds]
<lanefu> haha okay we must have improved error handlign when invoking customize_image :P
<lanefu> my customize_image scirpt has been changing verbosity in armbianEnv.txt forever, but nevre died becuase file didnt exist before
<lanefu> anyway i added a if statemetn
<lanefu> so sorry for my confusion
baffo32 has quit [Ping timeout: 256 seconds]
baffo32 has joined #armbian
<Tonymac32> OK: I hate the new password system
<Tonymac32> HATE
<Werner> You shall not hate
<Tonymac32> I shall
<Tonymac32> rejected based on dictionary word is going to cause rage episodes from a lot of people
<Werner> Hm well...good point. I use to change the first password always to testtest...
<Tonymac32> exactly
<Tonymac32> for development anyway
<xwigg> add sequence number to previous
<Tonymac32> no, I do that
<Tonymac32> it rejected because a dictionary word was in it
<xwigg> google translate to hindi, it's just a game
<Tonymac32> oh, and I mistyped, then hit the backspace, it kept adding '*' for every backspace
<IgorPec> hahaha
<lanefu> yeah i dont want teh dictionary bullshit
<lanefu> that will slow me down so much
<IgorPec> but it should not be very strickt
<xwigg> keepassx does wonders though
<Tonymac32> OK, added "special characters" and it worked
<IgorPec> we can set it even lower probably, but defaults seems prety reasonoble to me
<Tonymac32> but the backspace incrementing the '*' is just hard to look at
<lanefu> xwigg: you want ot use keepassx on TTL serial?
<xwigg> lanefu: copy paste works fine
<Tonymac32> OMG it doesn't honor backspace either
<Tonymac32> rofl
bigbrovar has quit [Ping timeout: 240 seconds]
<Tonymac32> so if you mistype you have to start over after getting an error
<IgorPec> tonymac32: well, if its a bug ... we have two weeks to fix
<xwigg> tonymac32: DEL
bigbrovar has joined #armbian
<Tonymac32> the DEL key shifts the other way though
<xwigg> depends on tty
<IgorPec> it has to remain terminal friendly, the rest idk
<lanefu> [TheBug]: what's the kernel module for ebin that we're looking ofr
<Tonymac32> DEL should *not* work, so if it does no one will understand
<lanefu> Tonymac32: ctrl-H
<xwigg> men with beards understand DEL
<Werner> Is the armbian logo available in higher resolution somewhere? So 2000px and up?
<Miouyouyou> The best password is hunter2
<lanefu> Werner: if you render nico'ds blender stuff
<Werner> I'd be fine with a simple png ^^
<Tonymac32> Anyway, getting a password into first login just took me 6 tries
<Tonymac32> so expect anger on our forum
<IgorPec> Werner: yes, i have it somewhere
<IgorPec> i might put it to the forum as well
<Miouyouyou> We'll just answer with "lrn2password kthxbye"
<IgorPec> font + psd file
<Tonymac32> it's a shame these armv7 boards are so slow (as XU4 fetches/updates faster than my RK3399 seems to be able to)
archetech has joined #armbian
<Tonymac32> I also replaced my rockpro64 with an XU4 for retro gaming, so much better (that is software I know)
<Miouyouyou> Sdcard ? nvme disk ?
<Tonymac32> SD
<Tonymac32> not even a fast one really
<Tonymac32> the NanoPC T4 with NVMe wins, but it has an unfair advantage. :P
<Werner> Hm svg file would be helpful too since this opens other plotting techniques...
<Miouyouyou> Yeah, once you drop the Sdcard, it suddenly reveals useable performances
<IgorPec> werner: i will check later. open a jira and i will ad it there
<Tonymac32> the XU4 I do retrogaming on has eMMC
<Werner> AR-368 is already there
<ArmbianHelper> AR-368 [Story] "Set up Armbian swag store" reported by Werner at 2020-07-18. Status: To Do
<Tonymac32> hahahahaha
<Tonymac32> OK I've launched Einstein@home on my MC1-solo, let's see what happens
<Miouyouyou> *MC1-solo catches fire*
<Tonymac32> I had 5 running at once, but they would almost never manage to save the task, something about the finalization of a work package murdered them every time
<Tonymac32> power is good, no errors about file system, CPU hang, and hard
<Miouyouyou> Ah yeah, I remember you talking about it... a year ago ?
<Tonymac32> not quite that long lol
<Tonymac32> I want to calculate pulsars and run CI on these
<Tonymac32> :D
<Tonymac32> OK, it's running
<Miouyouyou> I guess you'll just have to put fprintf(stderr, "Still didn't crash #45674") everywhere, like all good programmers do
<Tonymac32> well, unfortunately none of these number crunchers are open source
<Tonymac32> and I tried a monitoring session, got nothing
<Tonymac32> everything was fine, then it just died. no oops or anything
<Tonymac32> so I'm staring at the DMC wondering
<lanefu> "why stainless steel?"
<Miouyouyou> "Will it finishes one day ? Are there other signs of life in the Galaxy ? What am I going to eat tonight ?"
<Tonymac32> ha
<Tonymac32> "In the entire history of humanity, has there ever been a benevolent government?"
<Tonymac32> :P
<Miouyouyou> And did they give donuts to their population ?
<Tonymac32> That would be a sign of benevolence I think
<Tonymac32> As long as they didn't pay for them with taxes
<Miouyouyou> Right now, yeah. Cause I'm hungry.
<Tonymac32> hehe
<lanefu> didn't france try that with cake and it backfired?
<Tonymac32> well, you need to look at translations
<Tonymac32> "cake" was not "cake"
<Miouyouyou> Raiding the fridge, brb
<Miouyouyou> The cake is a lie
<Tonymac32> always
<Miouyouyou> They'll throw you into a fire pit
<xwigg> watchdog not working
<xwigg> ?
<Tonymac32> so einstein at home is claiming a 2H completion time, but is also using massively more RAM per task. I wonder if that will void this experiment
<Tonymac32> I was able to run 8 threads before
<Tonymac32> now it is doing 2
<Tonymac32> prior situation was ~5H completion per task, 8 in parallel. now is 2H completion, 2 in parallel.
<Tonymac32> perhaps the benchmarks it gave larger tasks to use the A15's
<Tonymac32> (The kernel has properly allocated, by the way
<xwigg> improved biglittle schedulinig?
<Tonymac32> yes for sure, but it now has gone to 8 threads, so other stuff must have been happening in the background
<Tonymac32> HTOP wasn't telling me anything that made sense. *shrugs*
<Tonymac32> sorry 7 threads, I just glanced at it (had to get that bagel/lox)\
birkoff has left #armbian [#armbian]
<xwigg> mm. zram swap 50% seems a bit over the top
<lanefu> we have fairly aggressive swapiness because there's performance by running through zram
<lanefu> so 50% is probably sane on a high workload
<xwigg> ah. 'm used to prevent swapping out ;-)
<Tonymac32> this poor cell phone processor was not designed to move heat like this
<Tonymac32> rofl
<lanefu> haha yeah even with the nice heatsink, my mcsolo's throttle when i peg them
<Tonymac32> my XU4 is liquid cooled. It does not throttle
<Tonymac32> however, ridiculous and over the top come to mind
<Tonymac32> (I wanted to see if it was possible to actually get 100% out of one)
<lanefu> wow dude
<lanefu> lol
<Tonymac32> exact size of exynos package
<Tonymac32> used a thin layer of thermal epoxy
<Tonymac32> (Also used said epoxy to put the big passive heat sinks on the Tinker Boards
<IgorPec> werner: config/sources/families/include/meson64_common.inc
<IgorPec> werner: sorry, rockchip64
<Tonymac32> thye entire liquid cooling setup and XU4 combined use less power than one of those. :P
<xwigg> yeah, that's the downside, but it's silent ;)
<Werner> IgorPec, yeah i noticed that and wondered because there is no definition for current and dev...
<IgorPec> yes, its there
<Werner> Ah the include
<Werner> yes
<IgorPec> actual DEV you can leave as is ... not sure if 5.8.y works
<Tonymac32> oh man, dutch is like German with the harshness spellings turned to 11
<IgorPec> this is the case with sunxi ... 5.8RC have some issues
<Tonymac32> IK
<Tonymac32> xwigg in the appropriate size even: https://www.aliexpress.com/item/32855811660.html
oida has quit [Remote host closed the connection]
<Tonymac32> probably too low power though, this tech is not pleasant to package
oida has joined #armbian
<xwigg> just fun to try, can't imagine cooling without moving heat
<lanefu> btw does focal desktop work?
<Tonymac32> that's why the efficiency on these is soo bad, the heat is moved via the electric current, but it is also conducted back through the assembly at the sdame time
<xwigg> no EM problems?
<Tonymac32> in DC no, but if you PWM it I'm sure it sings
<IgorPec> lanefu: yes, focal desktop should work
<IgorPec> where it doesn't?
<lanefu> IgorPec: i'll have to check... a while back i tested and it didn't..... so was checking expectations
<IgorPec> it must work, i mainly building focal desktop
<lanefu> IgorPec: haha okay cool
<IgorPec> even tested with latest changes to 1st login system
<IgorPec> but haven't actually build any for past 10 days :)
<xwigg> noob question; why does cpu freq work on all cores same time on H3?
<Werner> That remembers me of something. Wasn't cpufreq broken on sunxi with 5.7.x?
<IgorPec> IIRC it was absent on some boards
<IgorPec> but xwigg asks why speed goes the same for all cores
<IgorPec> right=
<IgorPec> ?
<Werner> Yeah I know, was just thinking...
<xwigg> yeah
<IgorPec> its designed this way
<IgorPec> in theory each core can have its own scaling, but why?
<lanefu> IgorPec: i'm not sure kotz's kernel config got merged.. mvebu64-dev.config is a symlink to mve64-current.config
<lanefu> i have my own cpy of kotz's kernel so testing it
<lanefu> *kernel cofnig
<IgorPec> ok, could be that i didn't manage ... ok, try ... if you get somewhere, we can just push current to 5.7.y, right?
<Werner> Actually you could leave sunxi-dev at 5.8. Its dev. If it breaks it breaks. Dolph found some patches in forums iirc that caused trouble
<IgorPec> the problem is that automatic building is not that good
<IgorPec> it will break
<Werner> Hm okay
<IgorPec> which is why the whole pack must build ... at this point DEV targets are not important
<IgorPec> my sunxi is ready for review
<Werner> I'll do some building on that branch overnight and will test some on my boards tomorrow
<IgorPec> werner: you done with rockchip*? i'll check it
<Werner> I guess so. Not really an idea about what I am doing but I hope it suffices ^^
<IgorPec> hehe , no problem ... now it has to be alrigh on the higher level
<xwigg> IgorPec: because it's more efficient to just crank up the freq to 100% of the core in use than to wait for all cores to increase after load on one core is determined to be high?
<IgorPec> bugs on the boards are for the RC part
<IgorPec> xwigg: ok, that's true, but current tables are not done the way this is possible I think
<xwigg> IgorPec: no, it's a hardware limitation I guess, but I would like to know the root cause
<Werner> Aight I'll leave my machine build everything together and check tomorrow. Gonna crash now. Bye for now
<lanefu> cya Werner
<IgorPec> bye
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
macc24 has quit [Quit: WeeChat 2.8]
macc24 has joined #armbian
<xwigg> bad news; EU regulations state from 2021 the tax-free import upto 22E will be gone.
<Tonymac32> Yeah I saw that
* Tonymac32 does not follow up with rant about protectionism
<Tonymac32> the only time I have ever paid an import duty of any kind was when I brought 100 assembled microcontroller boards into the US from China
<Tonymac32> VAT doesn't exist here either
<xwigg> well, aliexpress is building a giant warehouse in belgium, but still, smaller companies will be difficult to reach
<buZz> aliexpress is 29873498273 small sellers though
<buZz> not all will want to pay for warehousing in belgium
<xwigg> perhaps chinese gvt will take the hit
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
<lanefu> [TheBug]: you around?
<lanefu> was just missing CESA module being enabled :P
<[TheBug]> lanefu: sup
<lanefu> [TheBug]: i have an image for you to test in a sec
<[TheBug]> hmm I won't be able to test this fully best I can do atm is pull out my v4, boot it from uart boot and then load the image on it. The v5's and v7 are currently in production so not really able to directly test this moment..
<[TheBug]> I think Kotc has a v5 and he may be able to test it there as well
<[TheBug]> if all you need is confirmation of booting etc then I can do that now, if you want me to fully test long term will need to schedule that for next restart of the v7 machine
<[TheBug]> and change the kernl over
<lanefu> primarly i want you to confirm CESA is there adn working
<lanefu> thats what was missing on the last test image i gave you right?
<ArmbianHelper> AR-227 [Story] "Move Espressobin current to K5.4" reported by Igor Pecovnik at 2020-04-27. Status: To Do
<lanefu> :P I thought igor updated the ticket because i saw a jira update, but it was because my image name has the Ticket # in it :P
livingsilver94 has joined #armbian
livingsilver94_ has quit [Ping timeout: 256 seconds]
bigbrovar has quit [Ping timeout: 256 seconds]
bigbrovar has joined #armbian
<[TheBug]> hmm then thats probably better tested on thg v7 when I can as I have a encrypted volume and can more easily tell if it comes up and works
<[TheBug]> i will put it on my list to test, I am tagged on that Jira so will update there as I get it tested
<lanefu> [TheBug]: you want debs?
<[TheBug]> just a tar.gz of kernel, initrd, config + modules would be easier for me if you can provide that
<[TheBug]> bim running the Buster image already
<[TheBug]> so I can push it there and see
<[TheBug]> lanefu: btw who came up with the networking config on those images
<[TheBug]> the fact its now handled by a bridge and then also is now handled by the networkd by systemd is a huge pain in the ass to configure
<[TheBug]> I had to rip all that out of your image to make it usable
<[TheBug]> removed the bridge and setup the interfaces normally to pull from /etc/network/interfaces
<[TheBug]> seems dumb to use systemd for handling networking on a headless server
<lanefu> careful who you call dumb
<[TheBug]> s/your/the armbian
<ArmbianHelper> [TheBug] meant to say: I had to rip all that out of the armbian image to make it usable
<[TheBug]> 'seems dumb' is an observation ;p
<[TheBug]> not the person but the decision
<lanefu> well purpose of the bridge is for the fastest out-of-box experience
<lanefu> doesnt matter what port you plug in. it works
<IgorPec> hey if you got this ebin working on 5.7.y we can branch, right?
<IgorPec> i merged sunxi / rockchip
<lanefu> IgorPec: yeah give me just a bit to PR.. i have current on 5.4 and dev on 5.7 right now
<IgorPec> booting on hw?
<lanefu> yeah 5.4 boots. been running your testing suite on it
<IgorPec> great, well, so our config is problematic? perhaps too big?
<lanefu> actually
<lanefu> lol i rolled back to our config
<[TheBug]> lanefu: yeah I get it, gotta make it usable for the n00bs, but still drives me nutty like a squirrel :Z
<lanefu> and then enabled the missing kernel module config
<IgorPec> the case is i also got it working once on 5.4.y ... then suddenly i could not boot it anymore
<IgorPec> i suspected the same issue like with had with focal, but not
<IgorPec> anyway, if it works for you, we are on
<[TheBug]> lanefu: also if you have packaged a full image with on of the kernels let me know a link and I will pull it and at least confirm it boots on v4 with recent uboot
<lanefu> [TheBug]: huh?
<[TheBug]> have you built a full image for the new kernels or only the kernels?
<[TheBug]> if you have a whole armbian image I can grab quickly and write to sd
<[TheBug]> I will boot it on v4 ebin
<[TheBug]> if not I will have to wait till later till I can inject the kernel in an image and test
<lanefu> [TheBug]: yeah i sent you links already in this channel :)
* [TheBug] must be blind person
<[TheBug]> ahhh
<[TheBug]> grabbing both now
* [TheBug] rubs eyes
<lanefu> its a good thing i have the ebin in bridge moode so i can plug in my other ebin and test it :P
<[TheBug]> LOL
<[TheBug]> :)
<[TheBug]> okay time for a smoke break and then I will pull out the v4 and test those
<lanefu> party time
<IgorPec> sleep time :)
<lanefu> have a good evening igor
<lanefu> whats left of it
<lanefu> IgorPec: we'll cut RC branch tomorr
<IgorPec> ok, that's fine. i will also switch nightly there when this is done
<lanefu> awesome
<lanefu> [TheBug]: 5.4 works on my v4 adn v5... 5.7 didnt work on my v4.. didnt test v5
<lanefu> bwaah i forogt abotu the mac address stuff
bigbrovar has quit [Read error: Connection reset by peer]
bigbrovar has joined #armbian
<[TheBug]> ohh yeah
<[TheBug]> thats a big mess
<[TheBug]> thanks for reminding me
<[TheBug]> anytime I boot that with ethernet connected I end up with another server going offline cause I haven't updated the mac from another armbian image I am using
<[TheBug]> LOL
<[TheBug]> always fun to boot it and half my storage network magically disappears
macc24 has quit [Ping timeout: 260 seconds]
baffo32 has quit [Ping timeout: 260 seconds]
kolla has quit [Quit: %fog relay%]
kolla has joined #armbian
<lanefu> bwaah where's the bridge get its mac
<[TheBug]> loading uboot
<[TheBug]> weeeee serial speeeds
<[TheBug]> hmm
<[TheBug]> what am I missing
<[TheBug]> you have a different file layout there?
<[TheBug]> ## Executing script at 06d00000
<[TheBug]> ** File not found /boot/armbianEnv.txt **
<[TheBug]> ** File not found /boot/dtb/marvell/armada-3720-community.dtb **
<lanefu> yeah i get that too, boots eventually
<lanefu> does yoru stall?
<[TheBug]> mine dies and doesn't go on complaining of ramdisk
<[TheBug]> Wrong Ramdisk Image Format
<[TheBug]> Ramdisk image is corrupt or invalid
<[TheBug]> thats v4
<[TheBug]> Wrong image format for "source" command
<[TheBug]> hmm
<[TheBug]> that is the 5.4.52
<[TheBug]> image
<[TheBug]> Armbian_20.08.0-trunk-lane_Espressobin_buster_current_5.4.52.img
<lanefu> are you booting frmo SPI?
<[TheBug]> no SPI is fried I am loading uboot via uart and then loading from mmc
<lanefu> oh
<lanefu> this is booting from SPI
<lanefu> your tftping u-boot via uart?
<lanefu> whatif you dumb the u-boot commands from here https://www.armbian.com/espressobin/
<[TheBug]> look at image
<[TheBug]> taht is what I am doing
<[TheBug]> I just abrev them by not including the other boot devices
<[TheBug]> is armbian uboot w/ that image
<[TheBug]> will keep checking
<lanefu> my v4 is booting from it..
baffo32 has joined #armbian
<[TheBug]> so it fails to load the armbianEnv.txt
<[TheBug]> cause it isn't there
<[TheBug]> it loads the image
<[TheBug]> loads the initrd
<[TheBug]> checks for inital dtb, then finds the one your using as second choice
<[TheBug]> proceeds to unpack initrd and then produces ramdisk image format error
<[TheBug]> I will try the other image
<[TheBug]> have another thought too
<[TheBug]> I will retest with a different uboot I think
<[TheBug]> have a thought maybe something weird with memory or whatever so will test a differnt one
<lanefu> yeah id say itsa u-boot issue
<lanefu> some support differnet image formats
<lanefu> teh no armbianenv is expected
<lanefu> although i think i want to add one
<lanefu> i think there usd to be one
<lanefu> ha yes found it
<[TheBug]> will try 5.7 then go back
<[TheBug]> its loading up a different uboot now
<[TheBug]> see if makes a difference
<nekomancer[m]> chewitt ping
sunshavi has quit [Quit: nil]
<[TheBug]> no dice
<[TheBug]> im comming up snake eyes over here
sunshavi has joined #armbian
<[TheBug]> same exact issue
<[TheBug]> Wrong Ramdisk Image Format
<[TheBug]> Ramdisk image is corrupt or invalid
<[TheBug]> 1 sec
<[TheBug]> gonna try a differnt set of uboot I have
<[TheBug]> this has diferent timings lowered ones
<[TheBug]> lets see if that makes it go
sunshavi has quit [Quit: nil]
<[TheBug]> okay
<[TheBug]> making progress
<[TheBug]> had to clock it down to 800/800
<[TheBug]> 5.7 is booooting..
<[TheBug]> at least starting..
<[TheBug]> and then..
baffo32 has quit [Ping timeout: 240 seconds]
<Tonymac32> nekomancer[m] what's up?
<[TheBug]> weird
<[TheBug]> has some kernel error during intial start
<[TheBug]> not sure if its crashed or taking forever
<Miouyouyou> Maybe a hung task
<[TheBug]> since uboot seems to wokr
<[TheBug]> will switch back to 5.4.52
<[TheBug]> and try again there
<nekomancer[m]> <Tonymac32 "nekomancer what's up?"> no-no, just stupid question on chewitt's github repo
<[TheBug]> since you said 5.7 failed on your v4 also
<Tonymac32> ah ok
baffo32 has joined #armbian
<Miouyouyou> Check for DTS and DTSI changes, maybe
<[TheBug]> It's lanfus dts so
<[TheBug]> :)
<[TheBug]> am testing
<[TheBug]> he said same issue in his test with 5.7
<[TheBug]> so yeah probably something in dts change
sunshavi has joined #armbian
<Miouyouyou> Something like half of the regulators not being defined, or something like that :3
<Tonymac32> I saw that on another board
<Tonymac32> ...
<ArmbianHelper> . [en~>eng]
<Tonymac32> be quite ArmbianHelper
<Tonymac32> lol
<Miouyouyou> ..。。。
<ArmbianHelper> 。。。 [en~>eng]
xwigg has quit [Quit: Leaving]
<[TheBug]> Okay! lanefu 5.4.52 is booting! 5.7 is not.
<Miouyouyou> Black magic
<Miouyouyou> Look at the dmesg errors
<Miouyouyou> In order to know which one you should forget about
<Tonymac32> what pmic is in there
<lanefu> [TheBug]: okay cool.. same here
<[TheBug]> I need to get and look at schematics for v5 sometime here
<lanefu> will build another 57 imag shortly with more verbosity
<[TheBug]> I had a fan attached to one of them on the molex adapter
<[TheBug]> and the fan fried its self and dhorted
<lanefu> building clean v5.4 now
<[TheBug]> shorted*
<[TheBug]> and blew the fuse for the molex
<lanefu> man that molex port is useless
<[TheBug]> I need to find and replace
<Tonymac32> because I think it was an allwinner board, the pmic was not being talked to by the kernel
<Miouyouyou> So now TheBug house is on fire
<Tonymac32> martinayotte?
<Tonymac32> my memory is bad
<[TheBug]> ohh that happened the other day lol, its running find just molex port is not functioning i am sure because fuse
<Miouyouyou> Remember kids, never do molex
<[TheBug]> s/find/fine
<ArmbianHelper> [TheBug] meant to say: ohh that happened the other day lol, its running fine just molex port is not functioning i am sure because fuse
<Tonymac32> ya hafta know how ta yield it, aye?
<Tonymac32> isn't the espressobin the board we couldn't get any samples of?
<[TheBug]> Lol I actually consider those who got v4 to have some kinda luck
<Miouyouyou> Shouldn't be that hard to find an espresso machine, though
<[TheBug]> I did get one but it seemed that yield was ONLY sent to kickstarter
<[TheBug]> as I ordered more as soon as was public available and got v5
<lanefu> i'm the knucklehead to got a v4 from kickstarer and added it ot armbian
<[TheBug]> and the v5 is one I have most of
<[TheBug]> but yeah no demo devices cause I think they burned v4 after they ran out of them
<[TheBug]> this one I have has some fun issues after several years of use
<[TheBug]> like SPI is fried, even tried replacing spi chip
<[TheBug]> and it has a coil whine from the led
<Tonymac32> SPI can be replaced with care
<Tonymac32> oh
<Tonymac32> you tried
<Tonymac32> hmmm
<[TheBug]> I think some power issue on it
<[TheBug]> its just my test box now
<Tonymac32> so I see V3, V4, V7
<[TheBug]> I like the v7 alot actually, at least so far.. and you can't get v5 anymore and v6 was some secret production no one got
<[TheBug]> lol
<[TheBug]> v3 I don't think was public nor v6
<[TheBug]> there is v4, v5 and v7
<[TheBug]> I have 2v5 and one of v4 and v7
<Miouyouyou> So you got a collectioner board
<Tonymac32> this schematic is upsetting
<Tonymac32> just like a bag of cats tried to document
<lanefu> marvell losing interesting in the SoCs doesnt help either
<Miouyouyou> [ Food ] <-> [ Sofa ]
<[TheBug]> I have a feeling when they came up with this they had vendors ordering a version of it for routers and stuff and they just kinda hacked a version together of the chips and stuff and ran it at same time for lower BOM originally and v4 was kinda a hack job
<lanefu> get 'em Miouyouyou
<[TheBug]> v5 was always much more stable
* Miouyouyou throw a Pokeball on the Espressobin
* Tonymac32 thinks that was cat documentation of a process
<[TheBug]> v7 I can't really judge but seems to be okay so far, only a month or so in
<lanefu> okay man at first it seemed like v7s were worse
<lanefu> but there's like a PFSSense appliance on the v7 isnt there
<[TheBug]> im not sure about that but what I can say is the faster memory and cpu timings has a real impact on its IO and networking performance
<[TheBug]> a measurable one
<[TheBug]> now what that means for stability I don't know
<lanefu> yeah im sure the ddr4 is speedy
<[TheBug]> running original kernel from Globalscape on v7 and doing high IO results in kswapd bug that kills all IO and esstentailly crashes
<Miouyouyou> "Well, it can run fast, yeah ! ... Not very long, but you know, an espresso shot is like 20 seconds, right ?"
* Tonymac32 glances nervously at announcement of N2 +
<[TheBug]> running new mainlin on v7 seems to be stable
<lanefu> Tonymac32: lol no kididing
<[TheBug]> been running 5.4.45 stable for about a month
<[TheBug]> so yeah don't run old kernels on v7
<[TheBug]> 4 series seems to be instable on it
* Tonymac32 is honestly over this meson64 patch directory
<Tonymac32> you just can't do this
<[TheBug]> s/instable/unstable
<ArmbianHelper> [TheBug] meant to say: 4 series seems to be unstable on it
xwigg has joined #armbian
<lanefu> Tonymac32: do we need to break out glx patches frmo others ?
<[TheBug]> alright I will bbl time for some dinner and house chores
<Tonymac32> no that's not the problem
<[TheBug]> lanefu: I had to use the uboot clocked at ddr3 800/800 with 2chips - 1gb ram to bring up on v4
<Tonymac32> the problem is, completely indiscriminate copy/pasting from Khadas/chewitt/baylibre/mainline with absolutely no checking to see how they worked together
<[TheBug]> for reference
<Tonymac32> So it's a spaghetti mess
<Miouyouyou> And Tony doesn't like uncooked spaghettis
<Tonymac32> like this, in meson64: 01-10-soc-sunxi-convert-to-devm_platform_ioremap_resource.patch
<Tonymac32> ???
<Tonymac32> So I open it,
* [TheBug] Someone call for Al? Al Dente? (awful joke.. couldn't resist)
<Tonymac32> and it's meson64 clock and canvas patches
<Miouyouyou> Maybe someone working for meson64 boards, has devised a new way to randomize patch contents
<Tonymac32> and we have patches for the libretech PC, which I think only BayLibre even has any of
<Tonymac32> And the N2 got added with it's stoneage garbage u-boot we have to tiptoe around....
<Tonymac32> </endrant>
<Miouyouyou> You might be able to resolve this mess, using an old secret technique
<Miouyouyou> rm *.patch
<Tonymac32> hahahahaha
<[TheBug]> LOL
<lanefu> Tonymac32: know what happens when you put a le potato image into a tritium?
<Tonymac32> uhhhhhh
<Miouyouyou> You got a tritium potato
<Miouyouyou> They're toxic
<Tonymac32> can be used as lamps though
<lanefu> lol
<Miouyouyou> If you don't fry them look at TheBug
<[TheBug]> lol
<[TheBug]> I have the h2+, h3 and h5 versions
<[TheBug]> I don't have the amlogic on
<[TheBug]> one*
<Tonymac32> I have the tritiums, the potato/frite
<Tonymac32> and the renegade and renegade elite
<Miouyouyou> That's a strange evolution
<Miouyouyou> The potatoes went renegade ?
<Tonymac32> The elite I need to spend more time with IF I EVER GET THE MESON64 SORTED OUT
<Tonymac32> lol
<Tonymac32> renegade potatoes
<Miouyouyou> The new game from EA
<[TheBug]> lvrp16 still never got back to me on emmc for my h5 :Z
<Tonymac32> sounds about right
<[TheBug]> but thats okay, it still has it's place on the wall of boards :D
<Tonymac32> I was talking about EA, not lvrp16 lol
<[TheBug]> i know, lol
<[TheBug]> I was just picking on him to see if he was lingering ;p
<Tonymac32> making sure lvrp16 knows hahaha
<Tonymac32> I'm starting to try to find tasks for all my misfit boards
<Tonymac32> like my NanoPi M1 Plus
<[TheBug]> hmm
<[TheBug]> brave words
<[TheBug]> if you like that you will probablyu *gasp* like the BananaPi M2 boards better
<Tonymac32> I like that layout, a shame they didn't stick with it
<[TheBug]> believe it or not
<[TheBug]> I have M1 and M1 plus
<[TheBug]> and I have BPi M2 H2+ and H3
<[TheBug]> and believe it or not
<[TheBug]> M2 boards are active use
<Miouyouyou> I don't believe it
<[TheBug]> Nano boards sit in board
<[TheBug]> I know its hard to believe
<[TheBug]> the emmc is a nice touch
<[TheBug]> but H2+ does run hot from single voltage (no voltage regulator)
<Tonymac32> yeah, I learned that on the Tritium
<Tonymac32> AHA! found my powerline ethernet adapter! Now my shed will have internet again
<[TheBug]> If I had to say for H3/H2+ which order roughly I woud go something like OPi Plus 2e, Opic PC Plus, OPi PC, BPi M2 h2+ and h3, Beelink X2, then Nano Pi M1 plus, Nano Pi M1
<[TheBug]> something like that
<[TheBug]> ohh I left out the libre boards
<[TheBug]> they would actually probably come before nano but just a slight bit
<[TheBug]> problem with Friendly elec boards is
<[TheBug]> very small pcb with limited copper in pcb so doesn't disipate heat well
<[TheBug]> then to use video reliably you have to clock down dram
<lanefu> Tonymac32: yay tater booted to focal desktop..... no sound :)
<[TheBug]> or you get glitching in video
<lanefu> built from your recent commits an hour or two ago
<Tonymac32> my nanopi's of interest are the NEO, CORE, DUO
<Tonymac32> headless beasties
<Tonymac32> I have a NanoPC T4
<Tonymac32> that thing is actually very nice
<Tonymac32> and I like the K2
<Tonymac32> Like(d) since the SoC is gone
<Tonymac32> and they discontinued
<[TheBug]> Tonymac32: most of my experience / testing has admittedly been for H3Droid (Android) more so than Linux, but we also tested Armbian which side loads with H3Droid so I was aware of weirdness more so in Android but sometimes in Armbian as well
<[TheBug]> Why I have some many H3 boards :D
<Tonymac32> I started my Armbian journey with the Tinker Board, which I have... something like 5 of
<Tonymac32> including a 1 GB RAM model (yeah, never sold)
<[TheBug]> bbiab
<Tonymac32> Hey, did we get the NanoPi M4V2 stable? I put it away after a couple tries and haven't gotten it back out. :/
<lanefu> i cant even spell nanopi
<Tonymac32> oh holy shit we have a patch changing the text offset to make uboot 2015.1 happy
<Tonymac32> I don't give a shit about 2015.1
<Miouyouyou> Well, you don't eat nano pies very often
<Tonymac32> now I'm just getting upset
<Tonymac32> lol
<Miouyouyou> You don't give a shit about uboot 2015 and kernel 3.10 ? But it's the FUTURE !
<Tonymac32> oof
<Tonymac32> N2 is supported in mainline, I had a kernel hang when I tried, and THIS IS PROBABLY WHY
<Tonymac32> ok that's a lie, it probably isn't
<Tonymac32> but it might be
<Tonymac32> setup MAC from command line?
<Tonymac32> what kind of patch is that?
<Miouyouyou> Maybe coupled with a Wifi driver with injection support
<Miouyouyou> To "assert the security of my neighbors wireless equipment, of course"