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
lids has quit [Ping timeout: 240 seconds]
lids has joined #armbian
<lanefu>
sup
<nekomancer[m]>
.whatis sup
<[TheBug]>
.sup
<lanefu>
sup == What's up?
<lanefu>
..sup
<ArmbianHelper>
sup [en~>eng]
<lanefu>
..'sup
<ArmbianHelper>
'sup [en~>eng]
ChriChri_ has joined #armbian
ChriChri has quit [Ping timeout: 264 seconds]
ChriChri_ is now known as ChriChri
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
DaRock has joined #armbian
TiD91 has quit [Ping timeout: 246 seconds]
swedneck has quit [Ping timeout: 246 seconds]
BenZucker[m] has quit [Ping timeout: 246 seconds]
agates[m] has quit [Ping timeout: 246 seconds]
psydruid has quit [Ping timeout: 246 seconds]
solderfumes[m] has quit [Write error: Connection reset by peer]
nekomancer[m] has quit [Remote host closed the connection]
phantomse[m] has quit [Write error: Connection reset by peer]
marcodiego has quit [Quit: Leaving]
TiD91 has joined #armbian
psydruid has joined #armbian
swedneck has joined #armbian
nekomancer[m] has joined #armbian
phantomse[m] has joined #armbian
agates[m] has joined #armbian
solderfumes[m] has joined #armbian
BenZucker[m] has joined #armbian
archetech has quit [Quit: Leaving]
torv has quit [Remote host closed the connection]
oida has quit [Remote host closed the connection]
oida has joined #armbian
RzR has quit [Ping timeout: 260 seconds]
torv has joined #armbian
DaRock has quit [Ping timeout: 264 seconds]
RzR has joined #armbian
RzR has joined #armbian
RzR has quit [Changing host]
DaRock has joined #armbian
martinl has joined #armbian
oida has quit [Remote host closed the connection]
torv has quit [Write error: Connection reset by peer]
<martinl>
I added another set of patches to R2 test build, hoping to get hdmi working
<martinl>
I think after that it should be useful enough to do first PR
torv has joined #armbian
oida has joined #armbian
martinl has quit [Read error: Connection reset by peer]
martinl has joined #armbian
BenG83 has joined #armbian
martinl has quit [Read error: Connection reset by peer]
BenG83 has quit [Ping timeout: 246 seconds]
martinl has joined #armbian
psydruid has quit [Quit: killed]
swedneck has quit [Quit: killed]
agates[m] has quit [Quit: killed]
BenZucker[m] has quit [Quit: killed]
phantomse[m] has quit [Quit: killed]
nekomancer[m] has quit [Quit: killed]
solderfumes[m] has quit [Quit: killed]
TiD91 has quit [Quit: killed]
martinl has quit [Quit: leaving]
martinl has joined #armbian
martinl has quit [Client Quit]
martinl has joined #armbian
martinl has quit [Client Quit]
martinl has joined #armbian
martinl has quit [Client Quit]
martinl has joined #armbian
solderfumes[m] has joined #armbian
agates[m] has joined #armbian
phantomse[m] has joined #armbian
swedneck has joined #armbian
nekomancer[m] has joined #armbian
TiD91 has joined #armbian
psydruid has joined #armbian
BenZucker[m] has joined #armbian
RzR has quit [Quit: ZNC 1.6.5+deb1+deb9u2 - http://znc.in]
RzR has joined #armbian
RzR has quit [Changing host]
RzR has joined #armbian
martinl has quit [Read error: Connection reset by peer]
martinl has joined #armbian
redhoticaro has joined #armbian
redhoticaro has quit [Remote host closed the connection]
archetech has joined #armbian
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Ping timeout: 245 seconds]
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Excess Flood]
hipr_c has joined #armbian
hipr_c has quit [Ping timeout: 245 seconds]
hipr_c has joined #armbian
martinl has quit [Read error: Connection reset by peer]
martinl has joined #armbian
BenG83 has joined #armbian
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
ubuntu has joined #armbian
ubuntu is now known as Guest94484
lids has quit [Ping timeout: 240 seconds]
lids has joined #armbian
oida has quit [Ping timeout: 240 seconds]
<nekomancer[m]>
anyone can to understand uboot output?
<scope>
depends on what it is
<nekomancer[m]>
at last I connect serial console to my N2 and see how it can't boot when I build btrfs image https://pastebin.com/Hzng1HeG
<nekomancer[m]>
seems it not see /boot partition of card at all?
<[TheBug]>
seems whatever sdcard you are using to boot from isn't formatted correctly or something?
<[TheBug]>
ohh
<[TheBug]>
you didn't make /boot ext3/4 did you?
<nekomancer[m]>
<[TheBug] "seems things are mising from you"> I see it can't find nothing at all. but, fr example, file `** File not found boot/dtb/amlogic/meson-g12b-odroid-n2.dtb **` present on /boot
<[TheBug]>
you can make the secondary partition brtfs but pretty sure uboot expects an ext based filesystem to read boot information from
<Guest94484>
lagecy u-boot, please also use vendor provided OS image.
<[TheBug]>
based on that output it can't access the filesystem on the sdcard to read information
<nekomancer[m]>
uinitrd present too
<[TheBug]>
and the uboot is set to try all devices so then it next tries to find information on usb and fails there also
<nekomancer[m]>
[TheBug]: I use armbian image build script. it does /boot as ext4
<[TheBug]>
hm
<[TheBug]>
then not sure why the mmc isn't readable
<[TheBug]>
you should be able from uboot to test this
<[TheBug]>
by manually doing ext4load
<[TheBug]>
amd then using ext4ls to list content on the sdcard
<[TheBug]>
if it can't list anything on the card.. well it won't be able to read anything there either
<Guest94484>
please destroy your u-boot on Emmc
<nekomancer[m]>
isn't readable when flashed with btrfs image, but all the same (including sdcard itself) readable if ext4 image?
<Guest94484>
than you can boot u-boot in SDcard
<[TheBug]>
or you could just write the newer uboot to the emmc/spi on the board..
<nekomancer[m]>
there no emmc — if you speak about emmc card
<scope>
try ext4 as root
<[TheBug]>
N2 isn't allwinner, are you sure it actually works that way and auto check sdcard for uboot next?
<nekomancer[m]>
ext4 as root works.
<[TheBug]>
as it should
<[TheBug]>
sounds like uboot you are using doesn't support brtfs
oida has joined #armbian
<scope>
I doubt u-boot 2015 recognizes any btrfs
<[TheBug]>
whats scope said
<nekomancer[m]>
I not "use" any uboot. I "use" armbian image generation script.
<nekomancer[m]>
Isn't uboot I see running from same sdcard?
<nekomancer[m]>
where it located at all?
<nekomancer[m]>
I don't know
<[TheBug]>
nekomancer[m]: I will bet dollars to donuts your loading the default uboot that came on the SPI on the board -- this may not be the newest uboot for the board which may also in newer versions support brtfs -- this is what people are saying to you, you can either try and build newest uboot your self and write to sdcard and assuming board boots from sdcard first you should boot newest uboot
<[TheBug]>
with compatability
<[TheBug]>
but most likely uboot y ou are currently booting from does not support brtfs thus fails to read filesystem on sdcard and thus fails to boot
<nekomancer[m]>
any "howto" somewhere?
<[TheBug]>
I would go to hardkernels site
<[TheBug]>
and check their howtos
<[TheBug]>
they usually have pretty compitent documentation
<nekomancer[m]>
but there are ext4 /boot on sdcard! should'nt uboot finish work on loading kernel(with btrfs support) from ext4 /boot?
<[TheBug]>
if you get really stuck and have a question you may consider #odroid and asking in there
<scope>
it says the odroid has a switch for loading u-boot from sd
<nekomancer[m]>
if I click that switch — I can see "petitboot" menu
<[TheBug]>
so you could compile your own uboot and write it to sdcard, then change to boot from sd as documented above
<nekomancer[m]>
I set to boot from sdcard already
<nekomancer[m]>
with switch
<[TheBug]>
then the uboot you are using is old in your compilation
<[TheBug]>
if you are indeed fully building the image and then writing it includeing the uboot to sdcard
<nekomancer[m]>
it means armbian build script build old uboot into image
<[TheBug]>
nekomancer[m]: my suggestion to get a better answer to your question may be asking in #odroid as their engineers usually linger and if y ou ask a good direct question they usually respond, maybe they could tell you the uboot version needed (or if supported at all) for loading from brtfs and maybe ask if they can point you towards instructions for building it
<nekomancer[m]>
Can I "just to look" uboot versiot on sdcard or img file without booting from them?
<[TheBug]>
nekomancer[m]: very possible the build script still references old uboot for that board, sure.. I don't know that my self, lanefu or IgorPec may be able to comment if they are not AFK
<[TheBug]>
no you have too boot to see uboot version as its usually part of the first info printed to screen
<[TheBug]>
I do not know any easy way to just read version from it written on an sdcard
<nekomancer[m]>
so — I see what I see. `U-Boot 2015.01 (Aug 17 2020 - 22:17:13)`
<[TheBug]>
so yeah you are using old uboot as mentioned by scope above
<[TheBug]>
2015 doesn't support brtfs probably
<[TheBug]>
"[10:16] <scope> I doubt u-boot 2015 recognizes any btrfs"
<nekomancer[m]>
ok,ok,ok, I understand — uboot not suported btrfs. then — there are separate ext4 /boot with files
<[TheBug]>
then the next step would be to make sure you are using the correct boot variables
<[TheBug]>
and that its load from /boot and not /
<[TheBug]>
printenv
<[TheBug]>
etc
<nekomancer[m]>
uboot read ext4 /boot, read there all that he needs to run initrd, and it finished.
<[TheBug]>
I know on ebin for examle you have to sepcify a bootprefix variable for t heir script
<nekomancer[m]>
aha! can I do that (check env variables) from console?
<[TheBug]>
yeah, bro no offense, but you been here long enough I shouldn't have to tell you to actually RTFM right? I mean a few minutes of googling you can find all the documentation on how uboot works, viewing envriotment , etc
<[TheBug]>
if you haven't spent time yet to learn
<[TheBug]>
its worth the 20 minutes of poking around
<nekomancer[m]>
maybe they mount /boot as root, but continue to look for files at /boot...
<[TheBug]>
at least to have a base understanding
<scope>
try printenv in u-boot, /boot must not be read from / fs as this is btrfs
<scope>
and current image for odroid has u-boot 2015, so that seems correct
<IgorPec>
there is no modern u-boot support for n2 afaik
<scope>
chain-loading??
<IgorPec>
well, didn't try tghat
<IgorPec>
for us this is not needed
martinl has quit [Ping timeout: 256 seconds]
<nekomancer[m]>
it means btrfs root on N2 unsupported
<scope>
you need to figure out how to ext4load boot in u-boot
<scope>
if you load uinitrd, image and dtb from u-boot, requires some tinkering
<[TheBug]>
nekomancer[m]: do a printenv in uboot, copy all your current settings into a text file somewhere, then take some time to mess with uboot, try using the ext4 tools to view the sdcard and get familiar with how they work, check through the variables and script you are using to make sure it fits your use case -- seriously spending the 20 or so minutes just to play like in sandbox and learn is
<[TheBug]>
worth it -- then when you see issues you will better know what tools you have available.
* nekomancer[m]
see in boot.ini `if ext4load mmc ${devno}:1`, devno=3, `ext4ls mmc 1` returns list of /boot files, `ext4ls mmc 3:1` returms error MMC Device 3 not found
* nekomancer[m]
newer trade with uboot before and not have idea how to talk with it. internet have tons of references completely unrelated, with commands that just not known with my uboot
drobo_00 has joined #armbian
<[TheBug]>
nekomancer[m]: let me say this, all the people here who work with it and know (some) of it are in that position because they sat there at the prompt, typed 'help' and then went through help for each of the tools to understand what it does and then manually step through and boot your board your self instead of relying on a script. If you can get through those steps, then debugging why it
<[TheBug]>
won't boot becomes easier because you have been exposed to what each tool does -- YES it will likely take 20-30 minutes of your time to test and figure out but again, it is worth it because the return on your investment is you now have a better idea of how to use it
<nekomancer[m]>
👍️
<nekomancer[m]>
addresses — hardcore. /me not deals with addresses after 1996
<archetech>
but I just want to come here and say wont boot so you'll make it work for me :P
* [TheBug]
kicks archetech under the table with his steel toed boot
<[TheBug]>
:Z
<[TheBug]>
lol
<nekomancer[m]>
<archetech "but I just want to come here and"> I just hope it's something like "just use this magic value"
<archetech>
tried arch wiki ?
archetech has quit [Quit: Leaving]
<nekomancer[m]>
not for this case. but often.
archetech has joined #armbian
<archetech>
based on the Phoronix site tests btrfs vs ext is a wash
Guest94484 has quit [Quit: Lost terminal]
<archetech>
no advantage
* nekomancer[m]
uses snapshots and deduplication with docker
<scope>
my 5T usb disk says otherwise
<archetech>
ext4 has that its called tar xcvf
<scope>
online fsck /defrag is nice
<nekomancer[m]>
where comes ${devno} in uboot? why it seems wrong?
<archetech>
on arm its a waste on a big vm server ok
<scope>
archetech: do you have a link?
<archetech>
deb wiki
<archetech>
just scanned it its full of risks
<archetech>
it competes with zfs and raid 5
<IgorPec>
in recently moved my nas to zfs
<archetech>
I use raid 0 for 2 disks and back it up occas
<IgorPec>
raid0 :) OMG
<archetech>
simple fast easy
<IgorPec>
ssd or mechanical?
<archetech>
I dont have alot of data
<archetech>
2 x 1TB spinners
<archetech>
I had my proxmox server pointed toit for nfs was very nice
<archetech>
prox needed almost no local storage :)
<archetech>
and it did nas duty too
<IgorPec>
i need a bit of reliable storege for armbian production
<archetech>
zfs on linux yeah thats production NAS/SAN way to go
<IgorPec>
yeah, its really nice. just got two hdd failures recently
<IgorPec>
and ... no problem ;)
<archetech>
you use a zfs distro or hand made?
<IgorPec>
hand made
<IgorPec>
actually i used debian + OMV for easiery management
<archetech>
yup nice
<archetech>
I use omv now no prox
<scope>
what hardware?
<archetech>
just a humble E5300 2 gb ram and couple disks
<archetech>
in a showbox itx case
<archetech>
shoe
<IgorPec>
i use some i7 but moving to some proper xeon soon
<IgorPec>
no ecc atm
<IgorPec>
8 x sata + 10Gbe
<archetech>
yeah I had a dual xeon ecc mirrored proxmox setup
<IgorPec>
i have dual xeon only for compiling
<IgorPec>
and i am thinking to put it away
<scope>
threadripper?
<IgorPec>
not sure. got ryzen 3950x wkst
<Werner>
Decent enough
<IgorPec>
its almost 2x faster then this old dual xeon
<archetech>
so use that for your new zfs
<IgorPec>
that's biiig
<IgorPec>
28C server for NAS? :)
<archetech>
heh
<archetech>
I had my pmox in HA drbd mirrored servers
<archetech>
just for fun I could poweroff one and the cluster kept going
<scope>
make -j 56 would be nice to see
<archetech>
yeah why upgrade that
<archetech>
sound s nice even if its older xeons
<IgorPec>
like i said, this is old hardware
<IgorPec>
it looks impressive, but its slow from todays perspective
<scope>
yeah electricity is not cheap here, moved away from q6600 as server
Bitschnuppe has joined #armbian
<IgorPec>
also that, eats more power for sure
<archetech>
I used L5410's low power cpu's
<archetech>
2 quad cpu's
DaRock has quit [Ping timeout: 256 seconds]
<archetech>
what xeons model is it?
<IgorPec>
CPU E5-2683 v3 @ 2.00GHz
<IgorPec>
two servers, two switches and 30 arm boards = 300 W in idle
* lanefu
CPU E5-2650 v3 @ 2.30GHz
<scope>
if you cluster the arm boards, are they faster than the xeon?
<IgorPec>
they are single mode, for auto testings
<lanefu>
scope: it would take a pretty optimized workload and probably more than 100 SBCs to get into xeon territory
<IgorPec>
solo for that purpose
Strykar has quit [Quit: /quit]
<scope>
distcc kernels?
<archetech>
300w idle gah yeah thats not good
<IgorPec>
that's everything together ... while around 100W goes for this dual xeon IIRC
<IgorPec>
and at least on that one, things works as expected
<leventel>
cool
thefinn93 has quit [Remote host closed the connection]
thefinn93 has joined #armbian
Bitschnuppe has quit [Ping timeout: 260 seconds]
BenG83 has joined #armbian
BenG83 has quit [Quit: Leaving]
jwb__ has quit [Remote host closed the connection]
<leventel>
IgorPec: yep, that image is just working fine :)
<leventel>
thanks again
R0b0t1 has joined #armbian
<R0b0t1>
hi, I need a board with USB3 device mode
<R0b0t1>
it seems RK3399 parts will fit, but based on a parametric search someone linked here only board I know that will support it is the Boardcon EM3399 which I can't find on website
<R0b0t1>
otherwise there's two boards that might qualify, but I am unsure if the USB3 port with the device mode has been brought out to be usable
<R0b0t1>
OrangePi 3399 and Firefly 3399
jwb_ has joined #armbian
<scope>
friendlyarm has 3399 afaik
<scope>
m4v2
<leventel>
could it be an SD card issue, if during a reboot armbian stuck here forever http://ix.io/2uzv
<[TheBug]>
well
<[TheBug]>
trick question
<[TheBug]>
does it contain a brtfs file system?
<leventel>
this is the base image, I didn't modify anything yet
<[TheBug]>
if it does and it's freezing there.. seems quite odd
<[TheBug]>
if it doesn't and it's freezing there, even more odd maybe
<leventel>
haha
<[TheBug]>
hm
<[TheBug]>
is this first run?
<scope>
would expect to bail out to initrd
<[TheBug]>
and how long are you waiting before you decide it is froze?
<leventel>
it's after a reboot, I was waiting 5+ mins
<[TheBug]>
hmm
<[TheBug]>
if you power off fully
<[TheBug]>
then boot again
<[TheBug]>
does it boot?
<leventel>
yep, and it's boot in 15 secs
<[TheBug]>
okay so there is something wrong with the state of the board on a reboot where it likely isn't resetting fully
<[TheBug]>
sounds like it tries to start boot but then encounters problem and just fails outright
<[TheBug]>
leventel: what board?
<[TheBug]>
OPi3?
<leventel>
orangepi 3 (2GB with eMMC)
<[TheBug]>
IgorPec: something worth trying to replicate or known issue ?
<leventel>
I'm going to install the OS to eMMC
<R0b0t1>
scope, oh, I've bought things from them, they look pretty good
<R0b0t1>
only issue is confirming the USBC ports do 3.0/3.1/3.2
<R0b0t1>
a few I have found just wire up the 2.0 pins
<R0b0t1>
and the dwc3 USB driver is behind a hub
<[TheBug]>
leventel: would you be able to try a slightly older image version and see if the same issue occurs?
<[TheBug]>
leventel: this would help to know if its some type of issue needs checked in current before release as well, so if you can see it as something that changed in a specific version it would help us to know
<leventel>
hm, so I experienced the same with the current stable 5.4 image
<[TheBug]>
right I saw above that IgorPec mentioned some issue in current builds being worked on, why I am curious how far back this issue goes
<[TheBug]>
may help us to know if it is 'all versions' or started in a specific version