Thanks Bodhi! I issued ethtool -K eth0 tso off after boot and I will see if it helps.by schnee - Debian
Hi, i noticed that the logs on one of my Pogo Plug V4 are spammed with mv643xx_eth_port mv643xx_eth_port.0 eth0: failed to linearize skb with tiny unaligned fragment There are several hundreds of these lines per second. After reboot everything works fine for a few hours, but after that this happens and only reboot fixes. I searched the net, but found nothing conclusive. I don't seeby schnee - Debian
Thanks. I already downloaded the kernel source from kernel.org and tried to compile just the driver, but failed. Will search for dkms to understand it and try again.by schnee - Debian
Hi, I am trying to add wifi to a pogoplug V4 with an old USB wifi. It seems it requires zd1211rw.ko module, but it is not avaialble. Is there a "howto" to compile a module? Thanksby schnee - Debian
I am not an expert, but looking at your logs, the flash has a bad block, so your total size is 128mb-bad flash block size. I would try to reduce last write with 1 block size and see if that helps. Hopefully nothing important is stored at the end of the flash. As I remember the typical block sizes are 512, 2,048 or 4,096 bytes. Reduce the data size with these values and see what happens.by schnee - Rescue System
The flash is 128MB in Hex 7FFFFFF The flashing command adds up 2240000+5dc0000-1=7FFFFFF (128MB) and -1 because the first byte is written to 2240000 What if you try to write 5dbffff bytes. Do you get the same error?by schnee - Rescue System
Being sad :(by schnee - Debian
For the pogo plug v4 i had similar errors when I forgot to specify the arcNumber and machig in uboot. Can this be a similar problem?by schnee - Rescue System
Thank you! I will wait till the next releaseby schnee - Debian
Sorry if i was not clear. I am trying to do the same as I did with the Pogo V4, describer here: https://forum.doozan.com/read.php?2,132241 The problem is when I try to mount the NAND as JFFS2 filesystem example mount -t jffs2 /dev/mtdblock1 I get the error that it is unknown file system. I am not sure how to add the support to the device.by schnee - Debian
Hello, I am trying to utilize the onboard FLASh on the Pogoplug V3 board to load the kernel from it. I tried it with the latest kernel (5.4.224) and also with previous ones, but it seems that JFFS2 file system support it missing. root@pogo:~/ ls -l /lib/modules/$(uname -r)/kernel/fs total 64 drwxr-xr-x 16 root root 4096 Nov 20 19:30 . drwxr-xr-x 11 root root 4096 Nov 20 19:31 .. drwxby schnee - Debian
I guess i should read the release notes next time more carefully.... I just read the red USB warning I added the module to /etc/initramfs-tools/modules, then regenerated the initrd file: cd boot update-initramfs -u mkimage -A arm -O linux -T ramdisk -C gzip -a 0x00000000 -e 0x00000000 -n initramfs-5.17.4-kirkwood-tld-1 -d initrd.img-5.17.4-kirkwood-tld-1 uInitrd Now everything works fiby schnee - Debian
Hi, Yesterday I upgraded one of my Pogo V4 box to the latest 5.17.4 kernel. When i tried to mount the mtd block devices i received an error, that it does not exists. Looking into the dev list they are indeed missing: root@server:~# uname -r 5.17.4-kirkwood-tld-1 root@server:~# ls /dev/m* /dev/mem /dev/mtd0 /dev/mtd0ro /dev/mtd1 /dev/mtd1ro When i downgrade the kernel to 5.15.5 eby schnee - Debian
I made a setup to boot my Pogoplug V4 kernel from NAND. The reason was that i wanted to attache a SATA multiplier, but when it is attached the Pogoplug cannot boot from it. It has to use either USB or SD card for the boot, which i don't like. This setup allowed me to boot from SATA with the multiplier attached. WARNING !!!! This is a destructive change! It will overwrite the content ofby schnee - Debian
Thanks for the suggestions echowarrior108 nad bodhi. Upgrading to Debian 11 resolved the issue: root@server:~# pip3 Usage: pip3 <command> Commands: install Install packages. download Download packages. uninstall Uninstall packages. freeze Output installed packages in requirements formby schnee - Debian
Thanks for the update. It seems that you are on a different Debian verison (bullseye<>buster). here are the same commands and results from my box: root@schtorrent:~# cat /etc/debian_version 10.11 root@server:~# uname -r 5.4.179-oxnas-tld-1 root@server:~# python3 --version Python 3.7.3 root@server:~# apt install python3-pip Reading package lists... Done Building dependency treeby schnee - Debian
bodhi Wrote: ------------------------------------------------------- > Hi Schnee, > > > Yes, i installed it with apt install > pip3 > > > I tried to search for this package and it is not > in Debian repo. Wondering if you have a different > repo added to your /etc/apt/sources.list ? > > https://packages.debian.org/search?suite=all&arch=any&amby schnee - Debian
bodhi Wrote: ------------------------------------------------------- > Schnee, > > How did you install PIP3? was it from Debian repo? Yes, i installed it with apt install python3-pip I installed flexget while the box was running a 4.x.x kernel Edit: corrected apt command pip3 -> python3-pipby schnee - Debian
HI, after a long time i did a clean install for one of he Pogo V3. I used Debian-4.14.180-oxnas-tld-1-rootfs-bodhi.tar.bz2 (17 May 2020). When i rebooted i realized that the SSH server takes very long time to start (4+ minutes): [....] Starting OpenBSD Secure Shell server: sshd[ 267.217536][ C0] crng init done After searching the net i found https://daniel-lange.com/archives/152-Opensby schnee - Debian
Thanks Bodhi! Unfortunatelly it did not help: root@schnee:~# pip3 Segmentation fault root@schnee:~# flexget Segmentation fault Here is the trace with the same error [ 248.669091] 8<--- cut here --- [ 248.672839] Unhandled fault: page domain fault (0x81b) at 0xb6efe000 [ 248.679854] pgd = 6eeb95c9 [ 248.683238] *pgd=669b4831, *pte=622c955f, *ppte=622c9c7e [ 248.690178by schnee - Debian
Thanks, will check with the next kernel. Meanwhile just to confirm goind back to 4.x.x kernel solves the issueby schnee - Debian
So i just built a new system from scratch (latest rootfs), still have the error with the new kernel and the dump is still the same. [13425.208517] 8<--- cut here --- [13425.212267] Unhandled fault: page domain fault (0x81b) at 0xb6f98000 [13425.219279] pgd = 7ee1738c [13425.222663] *pgd=67b6f831, *pte=62c0a55f, *ppte=62c0ac7e [13425.229601] Internal error: : 81b [#11] PREEMPT SMP ARMby schnee - Debian
Finally i found the SATA boot disk creating script. Here is a link https://mega.nz/file/YRVhzQBa#jnu1JbxtMf8JStWZ0_m70AudAjfUSi-fLqhmWyxGIhQby schnee - Debian
I did some test on another box and it seems to be kernel related. I moved to 5.4.101 and i started the get the segmentation faults. Upgraded to debian 10 and it did not help. oot@schcamus:~# cat /etc/debian_version 10.9 root@schcamus:~# pip Segmentation fault root@schcamus:~# uname -r 5.4.101-oxnas-tld-1 root@schcamus:~# Going back to kernel 4.14.198 fixes the issue root@schcamusby schnee - Debian
Hi, I have the latest kernel running on my Pogoplug V3 box. I tried both pip and pip3, but both generates segfault. root@scht:~# uname -r 5.4.101-oxnas-tld-1 root@sch:~# cat /etc/debian_version 9.13 root@sch:~# pip3 Segmentation fault root@sch:~# python --version Python 3.5.3 Here is the segmentation fault trace from the logs: [ 373.945011] 8<--- cut here --- [ 373.948761by schnee - Debian
bodhi, thank you for the quick fix. It works perfectly on my box.by schnee - Debian
I looked at the man page and was unable to find the network option you are using (but could be wrong). Do you see anything in dmesg?by schnee - Debian
bluzfanmr1 , this is how i added and it works fine: auto lo eth0 iface lo inet loopback iface eth0 inet dhcp hwaddress ether 52:3b:20:XX:XX:XXby schnee - Debian
bodhi, i checked bdinfo in uboot and the MAC address is the same what is in the env variables ethaddr 52:3b:20:9c:11:51 If i remove the ethaddr variable and reboot the board the MAC address remains empty. Unfortunately i got this board without the case and with modified firmware with the above MAC. i have no idea how to recover the Cloud Engine assigned MAC address. Anyway, i think iby schnee - Debian
Thanks bodhi! I do remember changing the ethaddr variable during the installation ~1 year ago, but i could be wrong. Anyway i don't have the case with the sticker so i have no idea what was the original MAC for this box. Is there any way to figure it out? With the 4.4.X kernel version this MAC address worked fine. You mentioned that you may have to fix this in DTS, so i feel that curreby schnee - Debian