Nick

Thank you very much, I'll try that with the ReadyNAS tonight and let you know how I get on

@tkaiser Yes it is proprietary, it's a Netgear Ready NAS Duo (the Arm V7 version). It was second hand from eBay when I bought it. Though as a simple NAS box in the strictest meaning of the words it's not bad at all. It is also very open, with full root SSH access etc. TBH If someone asked me to recommend a cheap NAS box for home or office I would be happy to recommend it for general day to day use. Now onto more unusual uses such as compiling Armbian etc. That I didn't know about... FreeNAS looks interesting, I'm guessing it needs a little more than a H3 though I'll certainly do some more research into iSCSI, it looks like the Ready NAS supports it as well, so I might have a play with that first to get my head around it. Edit @zador.blood.stained yes only 1 lan port, but then so does my PC at the moment...

Good to know, though for me personally I haven't had a need to go back to debootstrap, the ng version works perfectly for me. Quite possibly, I did just run Armbian natively on my desktop at first, but then there were issues such as /tmp permissions and simply the fact that I was running alot of code as root that made me nervous. So I set up the VM. As for the laptop, I would strip it down to just the motherboard and stick it in an old 19" rack mount case so cooling could be addressed, but then there is still the problem of yet another box running 24/7 or having to come up with some sort of power switch / wake on lan etc. Sadly my motherboard has issues with 2 hard drives, it insists on booting from the wrong one regardless of settings I think I'll just get another SSD, or maybe a whole new PC

That's why I like posting here, within a few hours I've discovered two great new projects (new to me) distcc and OpenCL. It's unlikely that I would go down the Pi cluster root certainly not to the x10 route anyway, the post was in part inspired by another article I was read. I'm using debootstrap-ng so I am making use of ccache and to be honest, the uboot and kernel compilations are pretty quick on my desktop PC anyway despite being in a virtual machine. At the moment I musing over how I might replace my Armbian VM (the other inspiration for the post), the main problem I have right now isn't really compilation as such it's disk I/O. For space reasons the VM image is hosted on the NAS drive via a Gigabit Ethernet connection. My options are simply invest in a larger SSD for my desktop, which isn't a problem however it is only Armbian that is driving this requirement, replace the NAS drive or build a dedicated compilation box out of an old laptop I have kicking around. At some point in the future I'm going to replace the NAS drive anyway, as it's old and underpowered in lots of ways, the question is do I just buy something generic off the shelf that has decent performance, or do I try and build something. My existing NAS drive already performs extra features, for example it's hosting apt-cacher to speed up Armbian building (and target development) so I wondered about having it build Armbian as well. If I'm honest, I'm thinking of just buying a bigger SSD for my desktop and being done with it, but any thoughts would be appreciated.

martinayotte I have Ethernet, USB and SPI working, see attached patches. I've not communicated with an SPI slave yet, but the clock line is pulsing at 50kHz (set by the dts file) according to my scope. Thanks for the most part to your patches. Hope this helps you. I'm off to figure out GPIO in DTS now woop Edit: There was no DTS GPIO problem, rather a user is an idiot problem. I was using old code that was trying to export GPIO pins that don't physically exist eth-usb-spi-opi-pc-patches.zip

Are clusters able to do anything useful yet? I'm vaguely aware of open stack and I know that googles server farms are full of consumer PCs all networked together and even people making RasPi clusters of varying sizes, but are they any good for normal uses? For example if I bought 10 OPi PC's and stuck them in a case with an Ethernet switch and a hard drive could I build Armbian really really quickly (typing make -j40 would be nice ;-) or would it ultimately just get bogged down and do nothing?

Section "2.6 Interacting with the board" on page 11 of the User manual would be a good start Are you running windows or linux on your PC? You will need to run something like putty or minicom on the host PC to use that cable.

Have you tried plugging a serial port adapter into the UART to get an idea of why it isn't booting?

No problem, i'll look at it as well tomorrow. SPI is my next task :-)

martinayotte did you solve your spi timeout problem? If not, you might want to change the max speed in the dtb file. There was / is a bug in the spi driver where it ignores the speed setting in the user application therefore it defaults to the max set in the dtb file.

I have tar.bz2 files containing my user patches and updated armbian patches folders, however the forum wont let me up load them :-( It doesn't like the file type sadly. Can someone adjust my permissions or suggest an alternative format .zip maybe? that the forum likes. The patches are for working USB and EMAC on the Orange Pi PC.

It seems very unstable. Maybe a buffer overflow or something? I had the ethernet port lock up just now, the only way to fix it was a reboot. The board itself was fine (still had a serial console working) but no ethernet. On a different note: Did you also have to patch linux/reset-controller.h ? without it core.c wouldn't compile. Here is the patch that I came up with: --- /include/linux/reset-controller.h 2016-03-18 20:51:25.529580823 +0000 +++ /include/linux/reset-controller.h 2016-03-20 22:16:00.000000000 +0000 @@ -31,6 +31,7 @@ * @ops: a pointer to device specific struct reset_control_ops * @owner: kernel module of the reset controller driver * @list: internal list of reset controller devices + * @reset_control_head: head of internal list of requested reset controls * @of_node: corresponding device tree node as phandle target * @of_reset_n_cells: number of cells in reset line specifiers * @of_xlate: translation function to translate from specifier as found in the @@ -41,6 +42,7 @@ struct reset_control_ops *ops; struct module *owner; struct list_head list; + struct list_head reset_control_head; struct device_node *of_node; int of_reset_n_cells; int (*of_xlate)(struct reset_controller_dev *rcdev,

A little more info root@orangepipc:~# ethtool eth0 Settings for eth0: Supported ports: [ TP MII ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Supported pause frame use: No Supports auto-negotiation: No Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Half 1000baseT/Full Advertised pause frame use: No Advertised auto-negotiation: No Link partner advertised link modes: 1000baseT/Full Link partner advertised pause frame use: No Link partner advertised auto-negotiation: No Speed: 10Mb/s Duplex: Half Port: MII PHYAD: 0 Transceiver: external Auto-negotiation: on Link detected: yes

Interesting... I have managed to compile Armbian with the on SoC emac running yay I can ssh to it from the outside world, I can ping it from the outside world apt-get update works fine, but I can't ping other devices from the OPi PC. I'm guessing there are still some bugs around ICMP? Edit: Just checked dmesg and found this: [ 700.437256] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437270] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437283] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437297] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437311] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437324] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437338] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437351] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437372] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437386] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437400] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437414] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437428] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437442] sun8i-emac 1c30000.ethernet: ERROR: TX is full [ 700.437474] sun8i-emac 1c30000.ethernet: ERROR: TX is full Lots of them :-P Also this: sun8i-emac 1c30000.ethernet eth0: Link is Up - 10Mbps/Half - flow control off May explain your poor iperf results

Ahh that would be why my attempts didn't work either. I assumed it was unavailable because it wasn't needed. Great to hear that it's working :-)

4.4.6 not booting is a missing opi pc dtb file (at least for me). When I looked into it I think the h3 dtsi file was missing from the linux source. I'm guessing Armbian was including it as a patch? The dtsi file is in V4.5 of the kernel source (I found it in the git repo) so I figured I would use that to create a patch that created the file. Long story short it got me a little further but still no boot There is a commit in the Armbian repo from a couple of days ago that details the move of the patch files, I figured all I would have to do is reverse that commit locally, but I'm guessing something else has changed as well.

Thanks, Ill look at those patches later on. I'm trying to get back to where I was the other week, before all of the H3 stuff was moved to dev. Everything was working nicely then :-P That said, shouldn't complain as once the EMAC stuff is working properly it will be awesome :-)

Is USB broken for the H3 in dev for everyone or just for me? I saw something in the sunxi chat logs about Montjoe's ethernet patches breaking USB, but I thought that was fixed? root@orangepipc:~# dmesg | grep usb [ 3.183397] usbcore: registered new interface driver usbfs [ 3.183453] usbcore: registered new interface driver hub [ 3.183526] usbcore: registered new device driver usb [ 3.319078] ehci-platform 1c1b000.usb: EHCI Host Controller [ 3.319108] ehci-platform 1c1b000.usb: new USB bus registered, assigned bus number 1 [ 3.321579] ehci-platform 1c1b000.usb: can't setup: -110 [ 3.321596] ehci-platform 1c1b000.usb: USB bus 1 deregistered [ 3.321641] ehci-platform: probe of 1c1b000.usb failed with error -110 [ 3.321755] ehci-platform 1c1c000.usb: EHCI Host Controller [ 3.321778] ehci-platform 1c1c000.usb: new USB bus registered, assigned bus number 1 [ 3.324127] ehci-platform 1c1c000.usb: can't setup: -110 [ 3.324138] ehci-platform 1c1c000.usb: USB bus 1 deregistered [ 3.324167] ehci-platform: probe of 1c1c000.usb failed with error -110 [ 3.324264] ehci-platform 1c1d000.usb: EHCI Host Controller [ 3.324288] ehci-platform 1c1d000.usb: new USB bus registered, assigned bus number 1 [ 3.326671] ehci-platform 1c1d000.usb: can't setup: -110 [ 3.326685] ehci-platform 1c1d000.usb: USB bus 1 deregistered [ 3.326715] ehci-platform: probe of 1c1d000.usb failed with error -110 [ 3.326992] ohci-platform 1c1b400.usb: Generic Platform OHCI controller [ 3.327017] ohci-platform 1c1b400.usb: new USB bus registered, assigned bus number 1 [ 3.327145] ohci-platform 1c1b400.usb: irq 24, io mem 0x01c1b400 [ 3.385212] ohci-platform 1c1b400.usb: init err (00000000 0000) [ 3.385222] ohci-platform 1c1b400.usb: can't start [ 3.385293] ohci-platform 1c1b400.usb: startup error -75 [ 3.385305] ohci-platform 1c1b400.usb: USB bus 1 deregistered [ 3.385343] ohci-platform: probe of 1c1b400.usb failed with error -75 [ 3.385461] ohci-platform 1c1c400.usb: Generic Platform OHCI controller [ 3.385485] ohci-platform 1c1c400.usb: new USB bus registered, assigned bus number 1 [ 3.385597] ohci-platform 1c1c400.usb: irq 26, io mem 0x01c1c400 [ 3.445210] ohci-platform 1c1c400.usb: init err (00000000 0000) [ 3.445219] ohci-platform 1c1c400.usb: can't start [ 3.445262] ohci-platform 1c1c400.usb: startup error -75 [ 3.445273] ohci-platform 1c1c400.usb: USB bus 1 deregistered [ 3.445308] ohci-platform: probe of 1c1c400.usb failed with error -75 [ 3.445417] ohci-platform 1c1d400.usb: Generic Platform OHCI controller [ 3.445441] ohci-platform 1c1d400.usb: new USB bus registered, assigned bus number 1 [ 3.445535] ohci-platform 1c1d400.usb: irq 28, io mem 0x01c1d400 [ 3.505209] ohci-platform 1c1d400.usb: init err (00000000 0000) [ 3.505218] ohci-platform 1c1d400.usb: can't start [ 3.505257] ohci-platform 1c1d400.usb: startup error -75 [ 3.505268] ohci-platform 1c1d400.usb: USB bus 1 deregistered [ 3.505299] ohci-platform: probe of 1c1d400.usb failed with error -75 [ 3.505459] usbcore: registered new interface driver usb-storage [ 3.545753] usbcore: registered new interface driver usbhid [ 3.545759] usbhid: USB HID core driver The above was with a clean git clone of Armbian ./compile.sh BUILD_DESKTOP=no RELEASE=jessie BRANCH=dev PROGRESS_DISPLAY=plain PROGESS_LOG_TO_FILE=yes BOARD=orangepih3 PROGRESS_DISPLAY=plain COMPRESS_OUTPUTIMAGE=no DEST_LANG=en_GB.UTF-8 EXTENDED_DEBOOTSTRAP=yes APT_PROXY_ADDR=nas-1.local:3142 FORCE_CHECKOUT=yes PROGRESS_LOG_TO_FILE=yes

That looks good :-) So am I right in thinking that anything I put into userpatches/overlay on the host will appear in /tmp/overlay inside the chroot therefore allowing customize_image.sh to copy things as required?

Is it worth starting an Armbian wiki? It seems that for questions such as how do I write the raw file to an SD card, how do I disable LEDs on my whicheverPi, why doesn't my orange pi pc boot this work in progress image and so forth could be answered there. Hopefully it would reduce the Forum traffic for such things, but also when they do pop up a simple one line response with the wiki link is all that's required. I realise that there is the Armbian documentation, which is great, but only really covers how to use Armbian to build images. Equally there is gold mine of information in the forum, but it's sometimes buried a few pages deep or only comes to the surface with a particular search string. It sometimes takes me a while to find my own posts and I at least know that they are there. If the wiki did go ahead then I see the forum as a discussion place, solving problems discussing new boards etc. with all of the noise that goes with it, with the wiki presenting a polished concise answer to most questions. A while ago (before I even knew about Armbian) I started a banana pi wiki here https://bananapi.widgethub.co.uk/ but it never went very far, partly because it didn't have the visitor traffic to warrant putting the work in. In the short term at least I would be happy to start writing some of the articles and getting the basic structure together, as long as it has the backing of everyone else. On the flip side, there is already a wiki at sunxi. Personally I find the linux-sunxi site a little hard to navigate, but equally there is alot of information there so simply coming up with a duplicate of that doesn't see like a good use of resources. Any thoughts?

You are certainly doing something very similar to me. I can see why you would want to be able to copy and it maybe that your situation requires it, but I thought I would just let you know how I tackled it and why. I have setup my own apt-repo (possibly overkill for your needs, but equally not hard to do) by following this article https://www.debian-administration.org/article/286/Setting_up_your_own_APT_repository_with_upload_support There were a few reasons for setting up the repo: It solves the getting files into armbian problem I have quite a few boards to manage so having one central location that is online 24/7 helps alot apt handles all of the install, upgrading, dependency tracking etc. so whenever I need to upgrade my boards I just run apt update; apt upgrade (which is actually done with a cron script for auto upgrades) I agree that it is alot of effort and if you only have 1 board, in which case your suggestion of adding customize-distro.sh might be worth it, but as soon as you have multiple boards with different versions of your packages etc. apt may actually be your friend. FWIW I use customize-image.sh to pull in my repo key via wget and to add an extra file to /etc/apt/sources.d/ to include my repo.

Wow that is very cool, I like that alot!!

They both sound good to me, I'm not sure that I like the idea of swap files on SD cards anyway, I guess if you have a HDD attached it might be ok. That said, is it a good idea to have a small one as a get out of jail if you do use all of your physical RAM, or is the kernel pretty good at keeping a safe buffer anyway? I've not heard of zram/zswap so you may have to enlighten me there... Again, updating packages list is in my situation at least handled by my code when I need it, so it wouldn't worry me if it came out of first boot.

This may not be the right thread for it, but what in particular would you like to disable if you had the option. I know that my ideal Armbian would be one that creates the absolute minimum stable system with ethernet that I can them customise as I wish. Though Zador and I had a very brief discussion about this in another thread, where he did kindly make a change to the packaging so that it could be customised in lib.config. As Zador mentioned at the time though, one man's slimmed down is another man's bloated. Here is the other thread if you are interested http://forum.armbian.com/index.php/topic/714-armbian-lib-kernel-packaging-absolute-minimum-bootstrap/ for my own purposes as much as anything I have toyed with the idea of forking Armbian, writing my own version, considered suggesting a barebones branch. Some of those may have merit especially if Armbian is heading down an end user rather than dev route (as it appears to be with user creation etc.), but for the moment I'm fairly happy other than with the H3 regression grr

You might want to look here http://www.armbian.com/using-armbian-tools/ forvarious options, between lib.config and customize-image.sh you really shouldn't have to touch the Armbian code at all. I've been through the same process as you and learnt that almost everything can be customized. As you can see from the link, running compile.sh COMPRESS_OUTPUTIMAGE=no will prevent Armbian from shrinking your image. As for disabling change of root password and first login, I do exactly the same with customize-image.sh #Remove annoying user creation at login rm -f /root/.not_logged_in_yet echo -e "password\npassword" | (passwd root) Note that the above will change the root password to "password"