Jump to content

TCB13

Members
  • Posts

    67
  • Joined

  • Last visited

Everything posted by TCB13

  1. You always need and SD card / eMMC to start the boot, then it switches to the system on the NVMe disk. Here the general steps on how to set it up: parted -l parted /dev/nvme0n1 mklabel msdos parted -a opt /dev/nvme0n1 mkpart primary ext4 0% 100% armbian-config # -> system -> install -> boot from (eMMC/SD) run from NVMe. reboot cd /media/mmcboot rm -rf bin etc home lib media mnt opt proc root run sbin selinux srv sys usr var dev tmp Enjoy.
  2. TCB13

    NanoPI M4

    @williamv unfortunately I can't try the new kernel right now. But for for future reference: Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 Output is in kBytes/sec random random bkwd record stride kB reclen write rewrite 133470 128882 43201 75881 freread 102400 16 159788 222536 251343 252133 141357 215236 102400 512 530729 557813 497080 500936 469576 566763 102400 1024 572502 593950 518543 515461 502492 586127 102400 16384 635756 666508 591797 636189 613437 652484 NanoPi M4v2 + 8A custom power supply + Samsung NVMe 250 GB EVO 960 PRO (Samsung Electronics NVMe SSD Controller SM961/PM961). Kernel 4.4.213-rk3399.
  3. TCB13

    NanoPI M4

    hm.. not much better: dd if=/dev/zero of=test bs=100MB count=100 conv=fdatasync 100+0 records in 100+0 records out 10000000000 bytes (10 GB, 9.3 GiB) copied, 31.697 s, 315 MB/s My previous command gives perfectly good results on my desktop.
  4. TCB13

    NanoPI M4

    I was expecting something between 500-700 MB/s from my M4v2.
  5. TCB13

    NanoPI M4

    My results with a NanoPi M4v2 + 8A custom power supply + Samsung NVMe 250 GB EVO 960 PRO: $ dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB, 1.0 GiB) copied, 2.72603 s, 394 MB/s Kernel 4.4.213-rk3399 Slow isn't it ? I wonder what can be done to improve this.
  6. Maybe the problem is the kernel not the speed?
  7. @nobody @Noboot as we discussed on the other thread the issue seems to only happen in higher CPU frequencies. Now, the question I still have is, what power supplies are you using guys? I'm asking this because the issue might be lack of enough power to run at that speeds.
  8. With the 5.x kernel by default the CPU seems to upscale up to 2.02GHz, with the 4.x kernel it only goes up to 1.8Ghz. If I understood you correctly, you think that since my 5.x kernel uses a higher clock it just crashes due to lack of power. That's interesting and would mean that the new kernel isn't broken at all. It would also explain why it reboots without logs. What power supply are you using? and what would you recommend to run at 2.2 without crashes? What do you think about this @piter75?
  9. Yes I will, but only in a few days. I'll report as soon as I have new information.
  10. @piter75 and @NicoD as you guys suggested I decided to try the 4.x image and it works just fine! Look like there are bugs in the 5.x image. Thank you. PS: I tried to downgrade the kernel using armbian-config and my system never booted again had to do a fresh install. I've noticed that the CPU in the 5.x image was running up to 2.02Ghz on the big cores, with the 4.x image seems to be a cap at 1.80Ghz.
  11. Hi, About the Ethernet: as expected I was able to fix it by disabling tcp offload. Now about the hard drive: It's a SATA hard drive connected via a USB 3 ASM1051E SATA bridge. The disk has its own power supply and if I connect it to my old M4 (the original one) I'm able to copy GB of data without issues, however in the M4v2 the board restarts itself. I guess since it works fine on the old M4 it is safe to assume that the issue is on the new board and/or kernel. The only difference here is that the old board is runing a 4.X kernel while the new one is running 5.X.
  12. I've also notice some random reboots / maybe crashes without logs: As you can see it was running fine without any messages at 1:38 am and them it later rebooted by itself at 2:17 pm without logging anything.
  13. No I didn't build it myself but thank you for the update.
  14. @Nobby42 maybe your issue is somehow related to mine here: Does it also happen if you just try to use cp?
  15. @Igor I just noticed that you had some info about this problem before here: Is there any fix for this? I never had with this issue with a original board, using the same USB3 ASM1051E SATA 6Gb/s bridge What about you @NicoD , since you're always running nice tests, have you experienced such issues?
  16. Another thing I noticed: TCP offloading is still broken in all boards as discussed previously here: And here: https://github.com/MichaIng/DietPi/issues/2028 To make the network work properly under load and some protocols like HTTPS and NFS I have to run: ethtool -K eth0 rx off tx off after the network is up. This issue seems to affect both the original M4 and the v2 with all kernels.
  17. Hello guys, V2 buster image, kernel 5.4 from an NVMe drive. It works mostly fine but I've been having a consistent issue with copying directories with large files from a USB 3 drive to the internal NVMe. After `cp -r` the green LED of the board stops flashing and it reboots by itself. Unfortunately I can't get logs after the reboot so I don't know how to post more info about this. I never had this issue with the original board with the same USB3 ASM1051E SATA 6Gb/s bridge.
  18. @Igor is there an estimate on when the NanoPi M4v2 image will show on the downloads page? Thank you in advance.
  19. @martinayotte I decided to switch to the Debian kernel and the exact same config worked right after a reboot. Although Ubuntu is the OS that brags about using Apparmor by default looks like on ARM Debian works much better.
  20. Thanks for the answer, here is the result: root@nanopineo2:~# grep APPARMOR /boot/config-* CONFIG_SECURITY_APPARMOR=y CONFIG_SECURITY_APPARMOR_BOOTPARAM_VALUE=0 CONFIG_SECURITY_APPARMOR_HASH=y CONFIG_SECURITY_APPARMOR_HASH_DEFAULT=y # CONFIG_SECURITY_APPARMOR_DEBUG is not set # CONFIG_DEFAULT_SECURITY_APPARMOR is not set CONFIG_SECURITY_APPARMOR is set to y. Shouldn't it work out of the box with my changes to /boot/armbianEnv.txt ?
  21. Links above seem dead. I tried to get Apparmor to run in a Nanopi Neo 2 (18.04.1 LTS 4.19.13-sunxi64) as: apt install apparmor echo "extraargs=apparmor=1 security=apparmor" >> /boot/armbianEnv.txt update-initramfs -u reboot However after the reboot I still get: root@nanopineo2:~# service apparmor status ● apparmor.service - AppArmor initialization Loaded: loaded (/lib/systemd/system/apparmor.service; enabled; vendor preset: enabled) Active: inactive (dead) Condition: start condition failed at Sat 2019-01-26 16:10:54 UTC; 4min 55s ago └─ ConditionSecurity=apparmor was not met Docs: man:apparmor(7) http://wiki.apparmor.net/ I'm not much versed in this, do I need to compile a new kernel to enable AppArmor or, in theory my changes were enough? Thank you.
×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines