Jump to content

gleam

Members
  • Posts

    41
  • Joined

  • Last visited

Posts posted by gleam

  1. Oh no, sorry, MAC is the only thing working constant and correct. MAC is always the same, only the IPv6 is changing. IPv4 ist changing too bur this is matter of DHCP of a horrible router from Alcatel

    I have to live with. How are IPv6 are made and what is the programm used for it?

  2. Linux 2b8e 4.14.150-odroidxu4 #2 SMP PREEMPT Mon Oct 28 08:07:45 CET 2019 armv7l GNU/Linux

    DB 10.5 updates and upgraded

    High everybody :)

    AAMOE the starting procedure brings very often a new address for local use. As XU4 should be reachable by some other machines automatcally withour searching seconds an seconds I wonder if DHCP6 could help, when

    searching takes too much time and they already do have fixed IPv6 adresses and DHCP is disgusting.

    So, how the IPv6 ist made in Armbian?

    TIA 

  3. On 12/19/2019 at 5:15 AM, gleam said:

    AFAIK i really use often commands with systemctl and pure ip. Both are very useful. Especially "ip"for all problems with net.

    systemD is worth to learn about

    Ciao

    Well I looked at  https://linuxjourney.com/ carefully but by lack of all systemD commands it might be a little bit outdated, Routing is not worth to be mentioned as IPv6 never ist mentioned

  4. sudo reboot Reboot                      try  systemctl reboot

    sudo shutdown now Shutdown   try  systemctl poweroff

    systemctl restart gpm often needed with debian

       - " -       start, stop, ....  program instead of /etc/init.d/program startm restart, stop

    journalctl -b -p err Bootmeldungen, dazu nur die Fehlermeldungen wasaschaßis,
                weil da haddascho gestiefelt    --since-yesterday
    journalctl --since=2012-10-15 --until="2011-10-16 23:59:59"
    journalctl -u httpd --since=00:00 --until=9:30 
    journalctl /dev/sdc
    journalctl /usr/sbin/vpnc
    journalctl -o verbose -n
    journalctl -xe
    journalctl -b > /0/jounalctl-ge4-ox.txt journalctl -b > /7/journalctl_gd3_.txt

    systemctl status tahoe-lafs.service
    systemctl -t service      

    systemd-cgls | less?
    systemctl -a

    systemctl list-unit-files  | less
    systemctl -r
    systemctl --failed
    systemctl --now

    systemctl --message=ciao --halt
    systemctl poweroff  never --force
    systemctl start sshd  

    systemctl status /dev/sda = systemctl status dev-sda.device
    systemctl status /home = systemctl status home.mount  
    systemctl stop sshd@*.service
    systemctl kill crond.service

    systemctl stop sshd
    systemctl kill crond.service better than  killall -9 or kill
    systemctl enable example1 
    systemctl disable example1
    systemctl disable ntpd.service   '
    systemctl stop ntpd.service
    systemctl link /path/to/foo.service
    systemctl status networking.service
    systemctl list-units 
    'systemctl status bluetooth.service' and 'journalctl -xn'
    ln -s /dev/null /etc/systemd/system/ntpd.service  

    systemctl daemon-reload  

     

    BUT

    systemd-analyze plot > /tmp/plot.svg  fascinating!

  5.  

    As having to get along with old alcatel and debian 

    I come along with this script for DHCP is crap.

     

    ip link set dev enp4s0 down
    ip addr del 169.254.7.102/16 dev enp4s0
    ip addr add dev enp4s0 local 10.0.1.1/24 broadcast 10.0.0.31
    ip link set dev enp4s0 up

     

    For the other netcard I have to add

    ip route add default dev enp4s6 src 10.0.0.7 via 10.0.0.138

     

    now everthing works fine here. Never use alcatel!

  6. Hi Igor!

    I have another question as I have again and again starting from the very beginning. USB-sticks are kept in mind by kernel by /dev/disk/by-uuid allthough it is not inserted and so the kernel crashes. All the work ist gone. I only think about copying the SD-card at a point, when all works fine for not allways have to start from the very beginning. Wouldn't it be a better thing to divide to emmc and SD-card? OK, if kernel crashes it wouldn't help either. It just because it is so disgusting. Do you have a better idea?

    Odroid-XU4  Armbian-10 (I hate buster, it sounds like bastard :)

  7. One should try reportbug - vim-common - and if this ist reported

    version (2:8.0.0197-4+deb9u1) in debian stretch you are lost.

    Start with apt pinning for

    testing: 2:8.1.0549-1 and maybe it would help you.

    Vim in debian stretch will not be repaired.

  8. Does anybody ever get a solution for that problem?

    In VT trying to mark something via pointer this is not possible while editing in vim

    So I do "systemctl restart gpm" and pointer can mark text ........ BUT!

    vim switches to "-- VISUAL --' and cannot be used any longer!

    With arrow-keys one can go arround but all text is goin to be marked.

    You have to press the escape key for quite a long time.

    This is XAB    extraordinary   annoying   behavior

    I don't know if debian people are not willing to repair that or don's know how to do that.

    I did make several reports but just get stupid answers.

    Sorry for wanting to use: console AND vim AND gpm

  9. well, some annoying behaviors, extremly annoying, really

    do

    ip link set dev  enx001e06322b8e down

    and then

    ip addr show

    ip link is up again !!!!!!!!!!

    WHY? >8-0

    Well try that, open a file, insert

    ip --force link set dev  enx001e06322b8e down

    ip addr del 10.0.0.7/24 dev enx001e06322b8e

    ip addr add dev enx001e06322b8e local 10.0.0.1/24 broadcast 10.0.0.255

    ip -force link set dev enx001e06322b8e up

    close it and make chown "file" 755  and now it is better

    This is new in stretch ans idiotic.

    BUT! did you change  /etc/network/interfaces? I'm afraid it has to be changed too

    a don't want to reboot now.....

  10. something useful:

    Howto get these mmcblk-crap

    root@2b8e:/mnt# ls -lisa /dev/disk/by-uuid
    total 0
    12297 0 drwxr-xr-x 2 root root 160 Sep 27 20:58 .
     9262 0 drwxr-xr-x 7 root root 140 Jan  1  2000 ..
    13331 0 lrwxrwxrwx 1 root root  15 Sep 27 20:58 0000-3333 -> ../../mmcblk0p1
    12298 0 lrwxrwxrwx 1 root root  15 Sep 27 20:58 1334d04d-7ea0-44b3-9dba-fbc34e9e9ccd -> ../../mmcblk0p3
    11330 0 lrwxrwxrwx 1 root root  15 Sep 27 20:58 57f8f4bc-abf4-655f-bf67-946fc0f9f25b -> ../../mmcblk0p2
    14781 0 lrwxrwxrwx 1 root root  10 Sep 27 20:58 6afd7d22-8f78-4e09-9214-7df7421b7c06 -> ../../sdb1
     9291 0 lrwxrwxrwx 1 root root  15 Sep 27 20:58 7d408db8-5e88-4654-aaee-3e0d621c4fa1 -> ../../mmcblk1p1
    14750 0 lrwxrwxrwx 1 root root  10 Sep 27 20:58 D303-6887 -> ../../sda1

    don't forget:   mmcblk0p1 differs from mmcblk1p1

    and:

    find / .name mmcblk0p3 :-D  OKOK

    find / -name mmcblk0p2
    /sys/devices/platform/soc/12200000.mmc/mmc_host/mmc0/mmc0:0001/block/mmcblk0/mmcblk0p2
    /sys/class/block/mmcblk0p2

    Support isn't useful at all
     

  11. As mentioned I still use a memorycard with D9,5 , but did get e repaired eMMS with android from hardkernel.

    Now I made a look what's on to work with (file attached)

    Most annoying c&p using Debian stretch vim and gpm, please use nano  or D8 which is very old but works fine. Why old? Networkcard is still called eth. in D9 it is now end4s6 for instance an now en00(MAC) which is too long for many programs as sinfod.

    And one again, never try IPv6. going back to IPv4 because of sone problems, nothing will work, no net, no armbian-config so please adopt that for your needs:

    /etc/networking/interfaces

    nano /etc/network/interfaces

    # armbian-config created
    ##source /etc/network/interfaces.d/*

    # Local loopback
    auto lo
    iface lo inet loopback

    # Interface enx001e06322b8e

    auto enx001e06322b8e
    allow-auto enx001e06322b8e
    allow-hotplug enx001e06322b8e
    iface enx001e06322b8e inet static
            address 10.0.0.1
            netmask 255.255.255.0
            network 10.0.0.0
            broadcast 10.0.0.255
            gateway 10.0.0.138
            hardware ether  00:1e:06:32:2b:8e
            dns-nameserver 127.0.0.1 194.183.128.35 194.183.128.36
            dns-search at

    enx001e06322b8e I'm sure it is not yours

    Kernels reads this in the very beginning.

    to make changes use "ip". ip is strange but genious for it can be used for everything concerning nets.

    So get accustomed to it now.

    But just for the beginning:

    ip link set dev  enx001e06322b8e down
    ip addr del 10.0.0.7/24 dev enx001e06322b8e
    ip addr add dev enx001e06322b8e local 10.0.0.1/24 broadcast 10.0.0.255
    ip link set dev enx001e06322b8e up

     

    now as enx001e06322b8e is up now ask route -n and you might need:

    route add default dev enp4s0 gw 10.0.0.138 but that is ugly so use

    ip route add default dev enx001e06322b8e src 10.0.0.3/24 via 10.0.0.138

     

    I don's not what to repair but IPv6 is completly distroyed.

    1

  12. Well, things are going on. This time IPV6 <-> IPV4 and DHCP.

    As I mentioned my ADSL-modem is crap (ALCATEL). So switching to fixed address is not at all recommended. Here.

    Coming back to DHCP doesn't, work. No  IP no route, no relay (as the gateway should be called).

    Also IPv6-address disappeared. No SSH possible of course.

    And! eth0 is dead, enp4s1 ist dead, long live enx001234567890 where 1234567890 is the MAC

    My IPv4 looks like 10.0.0.x\24  thr gx is 10.0.0.138.

    Log in as root.

    Eine Zeile ist Memory usage:  9 % of 1996MB     IP:            nitchewo, null, niente

    you can put rthe following in a "script" without '#' and make it chmod "script" 755

    #ip link set dev enx001234567890 down
    #ip addr add dev enx001234567890 local 10.0.0.x/24 broadcast 10.0.0.255
    #ip link set dev enx001234567890 up
    #route add default dev enx001234567890 gw 10.0.0.138

    Then try ping 10.0.0.138  and ping or telnet 194.183.128.159 25  afterwards

    X you have to start with startx, also as root it works

     

    BEWARE: Don't try vim, gitfm, gpm,  please, that's crap from Debian!

    You only can use nano <wääähh> mc :-(  gpm is OK then

     

    Questions?

  13. OK, OK, OK, OK No problems! :-)

    Just experience. Mine. My XU4 is connected to VU7 Plus, nice. charming.

    The startscreen when starting from a very beginning: The letters are really very small.

    The VU7 is adjusted to 1200x720 but it is 1024x768

    Adjusting the size of the letters one should have some experience in doing that.

    Also in setting the keyboard.

    I don't have good experience without DHCP on XU4.

    Also not with IPv6 for the server doesn't support IPv6. (ADSL-Modem :->)

    https://wiki.odroid.com/accessory/emmc/recovery_xu4  is very good

    No problems.

    Working with Configuration utility one can get lost without reading the "book" simultanously.

    And here is some warning in "stretch' making stress: Copy&paste using vim in CLI is bad.

    One has always be prepared for "systemctl restart gpm" or vim.basic will block XU4!

    I always have a CLI with TOP to stop vim.basic.

    People at debian don't repaire it I don't get out why not.

    Is there anyone having experience with XU6 and Linphone?

     

    That's for the moment and all the best for the work on armbian!

     

  14. >systemctl disable NetworkManager.service

    Nice!

    Good hint!

    But it needs not to be disabled for it ist gone :-)

    >How inefficient is that !

    Nor so much because it can't manage IPv6 which i'm going to use.

    So I once will have DHCP never again. Never, never! 

  15. Good question .... I purged atop and still have top ... a miracle .... or something like that-

    But I got rid off network-manager and it was a genious idea!

    The DHCP of the alcatel-ADSL-mdem ist crap and the *****network-manager noticed each new IP got vrom the modem, didn't

    know which one to use and resigned managing.

    So I first stopped DHCP-client. a little bir better afterwards.

    And finally purged network-menager, although some programs recommended it.

    Every machine git its own static IP and now works perfect.

    Next switching to IPv6 because neither the modem nor the ISP uses it.

  16. Hi all! :-)

    To those who made the excellent work for my oroid!

    Well done, very well done!

    Well, just one point, maybe:

    Sep 08 13:47:55 odroidxu4 systemd[1]: Starting Atop process accounting daemon...
    Sep 08 13:47:55 odroidxu4 atopacctd[17567]: /run/pacct_shadow.d: File exists
    Sep 08 13:47:55 odroidxu4 systemd[1]: atopacct.service: PID file /run/atopacctd.pid not…ctory
    Sep 08 13:47:55 odroidxu4 systemd[1]: Failed to start Atop process accounting daemon.
    Sep 08 13:47:55 odroidxu4 systemd[1]: atopacct.service: Unit entered failed state.
    Sep 08 13:47:55 odroidxu4 systemd[1]: atopacct.service: Failed with result 'resources'.
    Hint: Some lines were ellipsized, use -l to show in full.

    But "top" allready works fine so this might be a hint.

    Another thing:

    Raspian doesn't us this network-manager and is nuch easier for configuration.

    Sorry for disturbing.

  17. 18 hours ago, tkaiser said:

    Just as a reference: a new Hardkernel eMMC module is to be released/shipped next week: https://wiki.odroid.com/accessory/emmc/reference_chart#new_orange_emmc_module

     

    The needed kernel quirk is already in our 4.9 kernel (just checked it) and I would believe the u-boot we're using is also ready....

     

    First of all: Thank you. Thank you for your engagement!

    Well then, I think I begin understanding.

    From 0-2048 there is the place for HardkernelBIOS, or Odroid-MBR. No, not MBR, because

    first is to check  the medium via BIOS, a medium being bootable. Copying an MBR to

    0-2048 cannot help because no medium for booting is known, cannot be choosen, there is

    nothing being able to choose. Also not by the tiny switch. So 0-2048 never ever should be writeable, only for one time. And the emmc never ever could be repaired? Only when be

    placed in the odoid? And it is lost for everything? Forever? With a dirty trick?

     

×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines