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
ChriChri_ has joined #armbian
ChriChri has quit [Ping timeout: 240 seconds]
ChriChri_ is now known as ChriChri
AndroUser2 has quit [Remote host closed the connection]
chewitt has joined #armbian
_whitelogger has joined #armbian
livingsilver94 has joined #armbian
livingsilver94_ has quit [Ping timeout: 272 seconds]
<archetech>
ho
<lanefu>
how ya livin arch
<archetech>
one day at a time
<awf_sbrt>
So, I think I might be a bit closer to figuring out my +2E's hangs. I'll need a few more days to be sure.
<lanefu>
awf_sbrt: do tell
<awf_sbrt>
Setting min and max CPU frequency to the same value seems to keep it stable so far. Maybe it's a frequency switching issue?
<awf_sbrt>
I suspected power issues, but it doesn't seem like that, since I got the hang twice even with max at 648MHz.
<lanefu>
did you ever catch any serial console messages when it hangs?
<lanefu>
well.. it does look like kernel 5.5 seems to have a bit better uptime for me
<lanefu>
lane@hambone:~/GIT/lanesible$ ansible hashi_cluster,armdocker-4.angrybear.com -m shell -a 'uptime;uname -a'
<lanefu>
Linux armdocker-4 5.5.14-sunxi #trunk.094 SMP Fri Apr 3 01:11:05 CEST 2020 armv7l armv7l armv7l GNU/Linux
<awf_sbrt>
I didn't notice any, except systemd complaining about writing journal messages. But I'll try 5.5 later, thanks! I'll give this a few days first to see if setting fixed frequency really makes it stable.
<awf_sbrt>
Oh. Another weird thing was that when I copied ~10GB data from my USB SSD to HDD, three of the files were corrupted. That was before I changed those frequency settings. Also I don't really trust that SSD.
<lanefu>
are you powering the USB from the opi?
<awf_sbrt>
The SSD is USB-powered, the HDD has its own power. It's a random chinese M.2 2242 SSD, so I wouldn't be surprised if it's doing something silly.
<awf_sbrt>
I'm a bit worried about the voltage drop on the cable/connectors. It seems a bit high, but I don't know what's normal. I'm getting about 4.7V on the 5V header, but the power supply I'm using stays around 5.2-5.3V up to 2A.
<lanefu>
Tonymac32: say something smart
on2 has joined #armbian
<on2>
Hallo ppl:-)
archetech has quit [Quit: Konversation terminated!]
on2 has left #armbian ["Leaving"]
archetech has joined #armbian
sunshavi has quit [Remote host closed the connection]
sunshavi has joined #armbian
<Tonymac32>
hello on2
<Tonymac32>
lanefu: taxation is theft
<Tonymac32>
LOL
<Tonymac32>
more seriously: awf_sbrt I had one of those external SSD's hooked to my N2, it pulls a ton of power
* Tonymac32
clarifies: Pulls a ton of power when talking about an SBC that doesn't even pretend to conform to any semblance of the power specifications of USB ports
<Tonymac32>
I'm generalizing as I don't have a +2E, but if it's like literally every other SBC on the planet, true
archetech has quit [Quit: Konversation terminated!]
<awf_sbrt>
That's possible, I should find a better way to power it at some point. But right now it doesn't look like a power issue - it was fine in 4.19, crashes in hours on 5.4, crashes even when I limit max frequency to 648MHz, but so far it seems to be stable at any fixed frequency. I'm giving it a day on each to see what happens, I'm at 1.2GHz now.
<awf_sbrt>
I was having power issues before after I added the SSD, but then I found out that I was powering it from a 1A power supply, switching to a 2A one fixed it.
<Tonymac32>
ok
<Tonymac32>
it could have something to do with the corruption though, even independantly of the crashing
<awf_sbrt>
That's possible, I'll have to try to reproduce that.
archetech has joined #armbian
<Werner>
Good morning
<Tonymac32>
Good morning
R0b0t1`` has quit [Ping timeout: 240 seconds]
R0b0t1`` has joined #armbian
archetech has quit [Ping timeout: 240 seconds]
sassinak-work has quit [Ping timeout: 260 seconds]
sassinak-work has joined #armbian
archetech has joined #armbian
<ArmbianTwitter>
@notkoutsie (): thanks armbian for being awesome https://t.co/6O2J6RUpTH (17s ago)
<HerculeP>
voltage drop from 5.2 to 4.7 is def. too high (also current depending) -> use a better==thicker cable
<HerculeP>
GM btw
livingsilver94 has quit [Ping timeout: 240 seconds]
<HerculeP>
thicker and/or shorter even
livingsilver94 has joined #armbian
<HerculeP>
Vdrop=Rcable(ohms)/I(A)
<HerculeP>
err, V=R*I
IgorPec has joined #armbian
igorp_ has joined #armbian
IgorPec has quit [Ping timeout: 265 seconds]
macc24 has joined #armbian
<ArmbianTwitter>
@armbian (armbian): @MarioMasta64 @notkoutsie Nightly builds from the trunk could be rough and are not officially tested on hardware, stable are. When CI will also run tests on hardware after each code change then this label could be changed. Bullseye is userspace trunk and small problems can be found here and there ... (21s ago)
igorp_ has quit [Quit: Leaving]
<ArmbianTwitter>
@notkoutsie (): @armbian @MarioMasta64 I have a reason why i run it: Testing. Thus i don't mind a reinstall Also never had any issues anyway so yeah (surprising?) And im running stable on my other OPi and in my experience stable branch is actually stable :D (6s ago)
<awf_sbrt>
Mmhm. I'll have to order some connectors and make a cable, since this board uses a barrel connector instead of USB. The cable I ordered with it always seemed a bit thin for its length. Or I could power it through the pins, I guess.
xwigg has joined #armbian
sassinak-work has quit [Ping timeout: 272 seconds]
sassinak-work has joined #armbian
RzR has joined #armbian
RzR has quit [Changing host]
_whitelogger has joined #armbian
_whitelogger has joined #armbian
livingsilver94_ has joined #armbian
livingsilver94 has quit [Ping timeout: 240 seconds]
<ArmbianTwitter>
@DieZuckerbude (Ben Zucker 🍰): @orangepixunlong @OPi_Community any idea why this #OrangePi One keeps crashing after a while? https://t.co/8PVkGY0aQl (19s ago)
awf_sbrt has quit [Read error: Connection reset by peer]
<lanefu>
HerculeP: do you have any issues iwth USB currently?
baffo32 has joined #armbian
<baffo32>
hey i'm trying to build the latest sunxi64 kernel because it's not on the apt mirror and my compile.sh is failing with a kernel build error: is there a way to make it show errors?
Hokedli has quit [Quit: Konversation terminated!]
<lanefu>
baffo32: did you check output/debug/*.log ?
<lanefu>
baffo32: are you trying to build sunxi-5.8?
<baffo32>
i'm trying to build teres-a64. ah-ha debug folder! thank you
<baffo32>
kernel ver is 5. ... something .45 . hum i tried to boot my system to check and it no longer boots ... that changes my priorities
<baffo32>
needed to be plugged in. 5.4.45-sunxi64 is my goal; not sure how to specify that to compile.sh but it's the latest teres-a64 armbian release
<Werner>
In case of sunxi it would be easy to go to 5.4.45 by removing all upstream patches from the sunxi-current folder.
<baffo32>
errr
<baffo32>
it sounds like you aren't expecting 5.4.45 to work yet?
<HerculeP>
lanefu: installed on SD, tested: reboot hangs, CR browser: no sound ("default source: none")
drobo_00 has quit [Read error: Connection reset by peer]
drobo_01 has joined #armbian
drobo_01 has quit [Ping timeout: 256 seconds]
<lanefu>
anyway lesson learned, calling someone on github a dick doesn't get you anywhere
<c0rnelius>
what did you think was gonna happen?
<xwigg>
haha, did you kiss and makeup?
<c0rnelius>
I was actually interested in that lil bit to see if it made a diff
<lanefu>
I was hoping it would cause a moment of reflection
<c0rnelius>
I've never had the urge to interact with TK, but I have seen that he has been a bit nicer to peps on the random webs.
<c0rnelius>
i thought he was trying to turn over a new leaf
<c0rnelius>
sorta say
<lanefu>
yeah i've seen him be quite nice on other forums
<lanefu>
but yeah he's not happy with armbian's path for sure
<c0rnelius>
seems that way
<lanefu>
my feathers get really ruffled when he says no one cares about XYZ
<c0rnelius>
the boy is moody. you prob shouldn't take it to heart. but I understand.
<c0rnelius>
he also seems to be missing a filter or two :)
<c0rnelius>
which i can relate too. the webs are the webs.
<nekomancer[m]>
<c0rnelius "the boy is moody. you prob shoul"> how not to take to heart code removal from repo?
<xwigg>
would be nice if sales would provide him with a board, solves a lot of heat
<c0rnelius>
nekomancer[m]: by testing the code and if its actually beneficial, putting it back. Give TK credit and tell him not to cry next time.
<c0rnelius>
puts him on the spot and suggests ur looking out for the greater good
<c0rnelius>
if the code is garbage... well fuck it.
<lanefu>
like ive learned some cool shit from him in the past, but another method of coaching is needed.
<c0rnelius>
same could be said for the lot of us. I understand ur point of view though. bs gets tiresome.
<baffo32>
everyone gets exhausted working in tough conditions. we can all learn ways to communicate with taxed people that our needs are unmet without triggering further degrees of taxedness.
<xwigg>
point a date at which someone will have tested the code
<lanefu>
xwigg: watcha mean?
<xwigg>
I mean; collaborate with others when patch can be tested and set a date in the answer.
<lanefu>
yeah that would be pretty good
<lanefu>
i wish we had more board maintainers
<xwigg>
it's what it is, just make sure he knows the patch won't get in /dev/null
<c0rnelius>
I'm tryin to convince some one to test for us. I don't own the N2.
<c0rnelius>
he is.. reluctant.
<lanefu>
ha
<lanefu>
cuz he hates armbian or because he's using his N2?
<lanefu>
I guess I should go through and make a list of optimizations that have been done for SBCs in the past and add them to like a checklist for adding a new board
<xwigg>
yeah, something like a status matrix mentioned before
<lanefu>
testing matrix? or onboarding matrix?
<xwigg>
what's onboarding matrix?
<lanefu>
tracking like adding a new board to armbian
livingsilver94 has quit [Ping timeout: 240 seconds]
livingsilver94 has joined #armbian
<xwigg>
gotta go get some sleep, bye
xwigg has quit [Quit: Leaving]
oida has quit [Remote host closed the connection]
oida has joined #armbian
<ArmbianTwitter>
@TonyMac_32 (Thomas McKahan): @groguard My Linux build server is an old cooler master case, I had to deflect it to fit my 780 TI at the time. These days the overclocked i7-3770 peacefully crunches Rosetta at home and builds Armbian (32s ago)