Werner changed the topic of #armbian to: armbian - Linux for ARM development boards | www.armbian.com | Github: github.com/armbian | Commits: #armbian-commits | Developer talk: #armbian-devel | Forum feed: #armbian-rss | Type 'help' for help | Logs: -> irc.armbian.com
CrashTestDummy2 has joined #armbian
Redentor has joined #armbian
CrashTestDummy3 has joined #armbian
CrashTestDummy has quit [Ping timeout: 268 seconds]
CrashTestDummy2 has quit [Ping timeout: 240 seconds]
stipa has joined #armbian
<stipa> got myself powerd usb3 hub with RTS5411 chip and it's practically unusable while connected to op lite 2 USB3 port
<stipa> looks fine when connected into opi lite 2 USB2 port
<stipa> didin't find any info online
<stipa> the driver for that chip is in the kernel since 2.6
<stipa> so, i guess the problem is in H&
<stipa> H6*
<stipa> got another USB3 hub without power Genesys Logic and it's relatively fine
<stipa> what a bummer
<stipa> got one usb3 port on a laptop, got to check what's going on there
<Macer> wifi seems to come back on nightly. i'm not really sure what state it is going in when i tell it to sleep
<stipa> Macer: try rfkill
<Macer> May 17 19:52:11 scipine kernel: [ 6174.584320] rk808-rtc rk808-rtc: Failed to update RTC control: -110
<Macer> May 17 19:52:11 scipine kernel: [ 6175.608315] rk3x-i2c ff3c0000.i2c: timeout, ipd: 0x10, state: 1
<Macer> odd noise in logs. i'm just trying to take a look at the suspend part
jschwart has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
jschwart has joined #armbian
CrashTestDummy2 has joined #armbian
CrashTestDummy3 has quit [Ping timeout: 240 seconds]
stipa has quit [Read error: Connection reset by peer]
archetech has joined #armbian
stipa has joined #armbian
stipa has quit [Quit: WeeChat 3.0.1]
<Macer> May 17 21:27:28 scipine systemd-sleep[6895]: /etc/systemd/sleep.conf:26: Unknown key name 'SuspandState' in section 'Sleep', ignoring.
<Macer> May 17 21:27:28 scipine kernel: [12307.340906] PM: suspend entry (s2idle) <- there it is. so it seems armbian handles it the same way manjaro on pbp does. s2idle
<lanefu> womp not the best typeo
<lanefu> gonna fix that now in repo
<lanefu> thanks
<lanefu> fail by yours truly 8 months ago
stipa has joined #armbian
<archetech> firefly 3566 is out
<lanefu> whats diff between 3566 and 3568
<lanefu> Ha toms is wrong
<lanefu> 3566 and 3568 have same cores and clock
<lanefu> And the said 3568 is faster
<lanefu> So 3568 has lots more IO options,ECC support and displayport
<Tonymac32> needs 3588 or why bother. ;)
<Tonymac32> also the next gen Amlogic needs to land, and soon
<Tonymac32> poor allwinner looks dead to me :D
<Tonymac32> esp. with PowerVR in there, Apple is dropping Imagination so I assume they will stop spending any of the tiny amount of money they might have been developing a decent Linux driver
<Tonymac32> how they still exist is actually a question. Who actually does use it in a product?
<lanefu> What's imagination
Redentor has quit [Quit: Leaving]
CrashTestDummy has joined #armbian
CrashTestDummy2 has quit [Ping timeout: 240 seconds]
Elpaulo has quit [Read error: Connection reset by peer]
Elpaulo has joined #armbian
ikmaak has quit [Ping timeout: 240 seconds]
ikmaak has joined #armbian
ikmaak has quit [Read error: Connection reset by peer]
ikmaak has joined #armbian
<Tonymac32> powervr gpu company
<Tonymac32> basically any SoC with a PowerVR is DOA for linux
archetech has quit [Quit: Leaving]
xtron has joined #armbian
Tony_mac32 has joined #armbian
Tonymac32 has quit [Ping timeout: 268 seconds]
prefixcactus has joined #armbian
xtron has quit [Read error: Connection reset by peer]
xtron has joined #armbian
flyback has quit [Ping timeout: 245 seconds]
MaxT[m] has quit [Quit: Idle for 30+ days]
flyback has joined #armbian
cheakoirccloud has quit [Quit: Connection closed for inactivity]
xtron has quit [Ping timeout: 260 seconds]
stipa has quit [Ping timeout: 265 seconds]
xtron has joined #armbian
sahm_ has joined #armbian
sahm_ has quit [Client Quit]
Findus has quit [Quit: meem]
kayterina has quit [Ping timeout: 245 seconds]
Findus has joined #armbian
psydruid[m] has quit [Ping timeout: 276 seconds]
hencke[m] has quit [Ping timeout: 245 seconds]
Sebastian[m]1 has quit [Ping timeout: 245 seconds]
kayterina has joined #armbian
hencke[m] has joined #armbian
psydruid[m] has joined #armbian
Sebastian[m]1 has joined #armbian
jstefanop has joined #armbian
<ArmbianTwitter> @ArieDeLight (Mac): @onnowpurbo Di install armbian? https://tinyurl.com/yemwr382 (13s ago)
cheakoirccloud has joined #armbian
CrashTestDummy2 has joined #armbian
SebastianLee8 has joined #armbian
CrashTestDummy has quit [Ping timeout: 240 seconds]
stefano79 has quit [Quit: ZNC - https://znc.in]
stefano79 has joined #armbian
SebastianLee has quit [Quit: Ping timeout (120 seconds)]
Elpaulo has quit [Remote host closed the connection]
SebastianLee8 is now known as SebastianLee
Elpaulo has joined #armbian
rpardini has joined #armbian
sunshavi has quit [Ping timeout: 265 seconds]
<cheakoirccloud> I broke cgroups, or they were never working. `● cgroupfs-mount.service\nLoaded: masked (Reason: Unit cgroupfs-mount.service is masked.)`...
<cheakoirccloud> unmask does nothing, but doing a https://www.irccloud.com/pastebin/94UW6xgd/working.log
<cheakoirccloud> Espressobin: Armbian 21.05.1 Buster with Linux 5.10.34-mvebu64
<[TheBug]> ?
<cheakoirccloud> I've `dpkg --purge cgroupfs-mount` and apt installed it again, not luck.
caiortp has joined #armbian
caiortp has quit [Read error: Connection reset by peer]
<cheakoirccloud> MPTCP?
sunshavi has joined #armbian
<cheakoirccloud> sysctl: cannot stat /proc/sys/net/mptcp: No such file or directory
CrashTestDummy3 has joined #armbian
CrashTestDummy2 has quit [Ping timeout: 240 seconds]
CrashTestDummy has joined #armbian
CrashTestDummy3 has quit [Ping timeout: 260 seconds]
<[TheBug]> what are you actually trying to do on the Ebin?
<[TheBug]> there isn't really a lot of CPU there so if you are trying to do docker would be curious what you intend to run there
<[TheBug]> but it may be that those things are not built into the kernel by default and you either need to load modules
<[TheBug]> or build the kernel your self to include them
<cheakoirccloud> Networking, it's to simulate the local DMZ... without actually having to have a host on the network for the testing(iperf3/ping/ect) I'd like to perform.
<[TheBug]> um not sure driver will work that way for Topaz
<[TheBug]> I know there is already a lot of fun there just to have networking work correct in Mainline
<[TheBug]> they use bridging in Armbian to handle the interfaces instead of physical speration of interfaces
<[TheBug]> since the driver is buggy
<[TheBug]> if you really want to split out each interface you probably want to test a legacy image
<[TheBug]> also I would bet on the modules you are wanting to use not be compiled in, so I would look to see if modules are there, if not you will likely need to build a kernel your self and either build the modules or compile the kernel your self
<cheakoirccloud> Essentially if I'm testing on the host I'd be using all kinds of fancy routing table/multi-home stuff. If I'm in a container then networking should be through something like a tun adapter, not sure of the specifics only that by default all containers are on the same bridge(decidedly not the same as the hosts networking).
<cheakoirccloud> I'm unsure of what u mean by splitting interfaces, I'm working under the assumption there is only one interface and a network switch.
<cheakoirccloud> As far as loaded modules, I'm able to run docker and networking works... it's just something is wrong with systemd.
<[TheBug]> again I will circle back to asking you why you would run containers on an EBin since its only 2x1ghz cores and really doesn't have a lot of memory either, but i digress. There is no switch in Mainline, that is what I just said above, they bridge all the devices since the drivers poop
<[TheBug]> if you want it to act like that you likely need to use a image with a legacy kernel
<[TheBug]> or the original dist that Globalscale provided
<[TheBug]> I think it was a 4.8 kernel
<[TheBug]> or similar
<[TheBug]> effectively in mainline it just treats it as a 'dumb hub' kinda situation
<[TheBug]> you can manually assign interfaces to the devices after removing them from teh bridge
<[TheBug]> but I am still not sure it will work fully as you are expecting
<[TheBug]> and I have seen weirdness with link control
<[TheBug]> where trying to remove the interfaces from the bridge device can actually cause a kernel panic in some cases as well
<[TheBug]> "one interface and a network switch" with the bridge in place I believe this is actually roughly how it acts
<[TheBug]> you can remove the bridge and then you will have 2 interfaces, wan, lan0, lan1
<[TheBug]> but it can be buggy when used as individual interfaces in some case
<[TheBug]> s/2 interfaces/3 interfaces
<ArmbianHelper> [TheBug] meant to say: you can remove the bridge and then you will have 3 interfaces, wan, lan0, lan1
<[TheBug]> Maybe you can give more information about what you want to accomplish exactly if you are still unclear and I can try my best to help
<cheakoirccloud> I'm bonding the ports, so still not really sure what your getting at. I'm nating a home network to two different ISPs. I could have a pi or something on the home network, for running network tests, but I figure a docker container is close enough. What I'm after is to not be under the `ip route table` and `ip rule` that the hosts have and a container wouldn't(shouldn't?).
<[TheBug]> I am sorry but that is still clear as mud, I think though I have answered your question a few different times / ways. Simply put the drivers you are wanting are not built-in with the default Armbian kernel build for Espressobin, so you will need to check if there are modules provided -- if not you will either need to build the needed modules for the same kernel version or build the kernel
<[TheBug]> your self. Past that, the network stack on there is buggy in general, so I would tune your expectations as such when using mainline.
<cheakoirccloud> AFAICT the bond works, the switch auto-detects a LASG group and when I do xfers I get two/three switch ports blinking as I'd expect for one link being used for full duplex.
\` has quit [Ping timeout: 245 seconds]
<cheakoirccloud> *LAG, sorry I dropped there.
<cheakoirccloud> AFAICT the bond works, the switch auto-detects a LAG(group is redundant) and when I do xfers I get two/three switch ports blinking as I'd expect for one link being used for each direction of a full duplex.
<cheakoirccloud> [TheBug]: I'm running some commands by hand... I think a reasonable system default would be for systemd to run these commands. All this networking talk is way above the point.
<cheakoirccloud> I'm using an Ebin instead of a WiFi router specifically because I can spin up docker containers, the big feature missing from OpenWRT and the like.
stipa has joined #armbian
<[TheBug]> cheakoirccloud: well I would say since in https://forum.armbian.com/topic/15442-orange-pi-3-how-ot-enable-cgroups/ lanefu suggested that cgroups was enabled there in some images though he doesn't say spcifically for all images there, only confirms for OPi3. As such, it is very possibly since EspressoBin is now CSC (community support) it may not have been getting needed updates to support
<[TheBug]> cgroups and you may have to work to do that your self. The best ways to confirm would be to ask IgorPec or lanefu or post on the forum.
<IgorPec> what?
<[TheBug]> Well tbh I still don't fully understand what he is trying to do but his original complaint was about cgroups being missing on Buster image for EBin above
<IgorPec> no idea, they are enabled
<IgorPec> we also dropped dealing with ebin. there are no maintainers
<[TheBug]> I did state that above
<cheakoirccloud> I't trying to upgrade docker, apparently I may have been trying to use cgroups v1 on a system with v2.
<IgorPec> hard to say. docker should work on all our builds, unless broken
<IgorPec> since automated testing facility is not very well done, we can't check this feature in automated way yet
<lanefu> whats docker error? i thought i added missing kernel configs on everything
<lanefu> cheakoirccloud: fyi bandwidth from the switch chip to cpu is 1gigbit
<lanefu> so you have a filesystem error when launching container?
<lanefu> and how are you launching container?
<cheakoirccloud> That fixed it: so when docker says `Error starting daemon: Devices cgroup isn't mounted`, likely talking about cgroups v1 on a system with v2. The solution is to upgrade docker.
<lanefu> sweet
<[TheBug]> :)
<ArmbianTwitter> @armbian (armbian): Check Armbian 21.05 Jerboa release docs https://t.co/k9utGTV9Rz https://t.co/0YK0Z3LmuL https://tinyurl.com/yjjcbof8 (5s ago)
<IgorPec> i knew it couln't be Armbian fault ;)
<lanefu> lol
<[TheBug]> haha
<[TheBug]> PEBKAC?
<[TheBug]> :D
indy has quit [Ping timeout: 240 seconds]
sunshavi has quit [Read error: Connection reset by peer]
sunshavi has joined #armbian
rpardini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
prefixcactus has quit [Ping timeout: 240 seconds]
<ArmbianTwitter> @joysfera (Petr Stehlík): @armbian Looks great but it would be cool if Armbian implemented read-only root option like Raspbian does. I mean log2ram is good but truly read-only disk (~ memory card) is far more durable. For certain installation types it would be blessing. https://tinyurl.com/yz6zu3fz (27s ago)
<lanefu> lol doesn't armbian literally support that via armbian-config?
<IgorPec> working on image to show that
<ArmbianTwitter> @armbian (armbian): @joysfera We are sure we had read-only rootfs options several years before Raspbian -> armbian-config Limited to Ubuntu based builds. https://t.co/4gP7UqXAAX https://tinyurl.com/yzwk6o35 (15s ago)
<lanefu> lol
<HerculeP> raspi-config: system..> Overlayroot Enable virtual read-only root filesystem
<HerculeP> its there
<IgorPec> so raspian is now our new copy cat :) lol
<HerculeP> err armbian-config ofc
<IgorPec> i think its there for at least 5-6 years out of my head
oida has quit [Ping timeout: 240 seconds]
<HerculeP> raspi-config has a similar option
<IgorPec> yeah, apparently
oida has joined #armbian
<HerculeP> I just wanted to confirm that armbian has it (silly typo, lol)
<IgorPec> yeah
<IgorPec> thx
rpardini has joined #armbian
<ArmbianTwitter> @joysfera (Petr Stehlík): @armbian I remember there was such option on very old kernels (like 3.4.x) but then it disappeared. If it's back then I am seriously outdated and need to give it a try ASAP. Thanks for correcting me, I'll be back with more thanks after testing it out! https://tinyurl.com/yeejfm2m (10s ago)
lids has joined #armbian
rpardini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
xtron has quit [Ping timeout: 260 seconds]
bzyx has quit [Ping timeout: 240 seconds]
bzyx has joined #armbian
rpardini has joined #armbian
indy has joined #armbian
rpardini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<ArmbianTwitter> @joysfera (Petr Stehlík): @armbian Ah, so the crucial info is that it is in the Ubuntu builds only. I always go for Debian ones beacuse I feel like Debian is for servers and Ubuntu is for desktops :-) Downloading Ubuntu for Orange Pi Zero right now... Thanks! https://tinyurl.com/ydkwznao (4s ago)
<Macer> so the default sleep mode is s2idle afaik for armbian. is there a way to test out actual suspend or is that still not working in armbian as well?
<lanefu> actual suspend doesnt work on pinebook on any distro far as i know... closed firmware issue
<lanefu> s2idle is the most reliable
<lanefu> i mean when i spell it right
<lanefu> s/pinebook/pbp
<ArmbianHelper> lanefu meant to say: actual suspend doesnt work on pbp on any distro far as i know... closed firmware issue
<Macer> lanefu, ah ok. i vaguely remember there being a patch for it from the pbp guys that hasn't been pushed yet but this fix/workaround did not extend to pcie
rpardini has joined #armbian
<lanefu> yeah i havent been tracking the forums over there in quite a while
<Macer> there's that heh
sunshavi has quit [Ping timeout: 252 seconds]
<Macer> but yah i saw a TF-A patch elsewhere that seemed to be a workaround for the deep sleep issue
sunshavi has joined #armbian
rpardini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
rpardini has joined #armbian
stipa has quit [Quit: WeeChat 3.0.1]
jstefanop has quit [Remote host closed the connection]
jstefanop has joined #armbian
jstefanop has quit [Ping timeout: 265 seconds]
danboid has joined #armbian
<danboid> Is the armbian build script known not to run in a (hirsuite) LXD container? It didn't work for me
<danboid> mknod: /root/build/.tmp/rootfs-62604dc6-01a2-482f-ab4e-ce0d70293e4f/test-dev-null: Operation not permitted
<danboid> I'm running Ubuntu 20.04 but I wanted to build a bullseye image so I tried LXD first
<danboid> I could use a VM and it should work
<danboid> but I wanted to know if anyone has got it to work with LXD?
<danboid> [ error ] ERROR in function create_rootfs_cache [ debootstrap.sh:189 ]
<danboid> So it built the kernel etc but failed when packaging it
<danboid> There could be some LXD storage pool setting to get this to work
<danboid> How much disk space is required to do an armbian build?
rpardini has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<lanefu> danboid: don't think anyone has tried. If building kernel should be fine. If building image it will need priviledges to mount loopback devices
<lanefu> So at the createrootfs cache stage its probably trying to mount
<danboid> Sounds about right
rpardini has joined #armbian
kayterina has quit [Ping timeout: 250 seconds]
psydruid[m] has quit [Ping timeout: 245 seconds]
nekomancer[m] has quit [Ping timeout: 258 seconds]
agates[m] has quit [Ping timeout: 258 seconds]
kayterina has joined #armbian
psydruid[m] has joined #armbian
agates[m] has joined #armbian
nekomancer[m] has joined #armbian