> > Is this SSD formatted Ext3 or Ext4? > > Both are formatted Ext4. Next time it comes up and > fails, i'll try to get a console running to > capture the issue for clarity's sake. I hope you did a "finalize" ( (no lazy_init) when you created this Ext4 rootfs.by bodhi - Debian
rsavera, > Am I missing something obvious here? I feel like I > am. Nothing obviously wrong. What Debian version are you at? cat /etc/debian_version If it is too old, try upgrading Debian first.by bodhi - Debian
sudos, > Something to report... Even with adjusted envs to > take care of the larger-than-11MB initramfs with > 5.18.6, whenever I'm doing an update that requires > the rewrite of uInitrd as outlined in the kernel > thread, the M300 always wants to just not-boot > afterwards. This also happened previously on older > kernels as well, but only past 5.15. > &by bodhi - Debian
> I guess these kirkwood boxes uses 3.3v i/o, so you > can just use a raspberry pi's serial port and > connect them directly. No ca-42 cable needed. IIRC, the rasPi has the Prolific PL3xxx chip built in. The Mirabox also has it.by bodhi - Debian
chessplayer, I beleive all GF Home can be kwboot'ed (Some GF Net cannot). Haha, ninja'ed by David.by bodhi - Debian
FYI, I've tried reinstalling OpenWrt today on my NSA310S and having quite a bit of problem. Perhaps something has changed in sysupgrade process. Update: This image works https://downloads.openwrt.org/releases/22.03.0-rc6/targets/kirkwood/generic/openwrt-22.03.0-rc6-kirkwood-zyxel_nsa310s-squashfs-sysupgrade.binby bodhi - Rescue System
Nirmal, > Do you think it's possible to repair this via JTag > or UART ? Nobody has reported this issue. But for this Pogo V4, it is almost unbrickable. You need to run kwboot to boot it over UART and see the behavior. See here for an example booting with kwboot: https://forum.doozan.com/read.php?3,51739,51919#msg-51919 And pay attention to the power supply. QuoteTrby bodhi - uBoot
BTW, The work around is you can do this on your USB drives that don't have uEnv.txt. Assuming ithe other USB drive is mounted at /media/sdb1 mkdir media/sdb1/boot cd /media/sdb1/boot And then copy the uEnv.xt over to this drive.by bodhi - uBoot
chessplayer, > I just wanted to report an issue I just had on a > Pogo E02 concerning uEnv.txt. I wanted to use the > file to use systemd and it worked about three or > four times. However, since what I am testing is to > have the box act as a file server in the Windows > network with automount of the non-boot attached > devices, I have more than one USB drive attachby bodhi - uBoot
At the moment, you can only read the fan speed on those boxes, but not controlling it (it is automatically adjusted by hardware). Nobdy has found an i2c command to controlling the fan. Wiki thread: QuoteSensors & Fan & WOL OWFSPlug : 1-Wire File System + Motion Detection Software in a NAND Package OWFS & Motion Detection (external link) How to control GPIO in Linux usersby bodhi - Debian
rayknight, You should read the whole thread! I will boot up my Pogo V3 and see for myself. But it does make sense to me from what I've seen in this thread that OpenWrt can not write to it currenly.by bodhi - Rescue System
Hi chessplayer, > P.S.: These commands apparently bricked the > device again! Arghhh :) Let's regroup. I think we are trying too much to get around this limitation without examining the flash structure created by OpenWrt installation and sysupgrade. For now it's a caveat in the installation: that you can only read from the envs inside Openwrt, but can only update it iby bodhi - Rescue System
CodeJACK, Strange, you've used setenv before in stock u-boot as shown above. QuoteMarvell>> setenv usb_set_bootargs 'setenv bootargs console=ttyS0,115200 root=LABEL=rootfs rootdelay=10 earlyprintk=serial' Marvell>> setenv load_uimage 'ext2load usb 0:1 0x800000 /boot/uImage' Marvell>> setenv load_uinitrd 'ext2load usb 0:1 0x2100000 /boot/uIniby bodhi - uBoot
setenv ipaddr <your box static ip address> # choose some IP addr appropriate for your box setenv serverip <your router IP address> ping $serverip IP address such as 192.168.0.100 if your router is 192.168.0.1. No need to blank out these numbers. And the 100 in 192.168.0.100 could be any valid number range 2--255. setenv ipaddr 192.168.0.100 setenv serverip 192.168.0.1by bodhi - uBoot
Hi CodeJACK, QuoteMost everything looks good. Except, [ 5.683348][ T1] 3 fixed-partitions partitions found on MTD device orion_nand [ 5.690808][ T1] Creating 3 MTD partitions on "orion_nand": [ 5.696710][ T1] 0x000000000000-0x000000100000 : "u-boot" [ 5.703125][ T1] 0x000000100000-0x000000600000 : "uImage" [ 5.709603][ T1] 0xby bodhi - uBoot
Hi CodeJACK, Most everything looks good. Except, [ 5.683348][ T1] 3 fixed-partitions partitions found on MTD device orion_nand [ 5.690808][ T1] Creating 3 MTD partitions on "orion_nand": [ 5.696710][ T1] 0x000000000000-0x000000100000 : "u-boot" [ 5.703125][ T1] 0x000000100000-0x000000600000 : "uImage" [ 5.709603][ T1] 0x00000by bodhi - uBoot
Hi Trond, > For the bulk of MVEBU boxes, your current version > is 5.18.6. Do you intend to keep Mirabox and > RN-102 current in the future? Yes, I will. I just forgot about it because nobody mentioned this. > The stable Debian version is now bullseye rather > than buster. Is it safe to follow the standard > Debian > release upgrade procedure or should I startby bodhi - Debian
> So right now, I have the stock u-boot installed > and working fine booting from USB with your latest > Debian. Cool! Now some info gathering for new u-boot. Boot with serial console and interrupt it at count down. And, help printenv md.l 0xF1010000 8 md.l 0xF1010100 1 md.l 0xF1010140 1 And then boot with whatever envs you are using to boot USB rootfs. Log in to Debiaby bodhi - uBoot
You're in luck! here is the stock u-boot dump from DaMacc: https://forum.doozan.com/read.php?2,22623,25690#msg-25690by bodhi - uBoot
CodeJACK, > I wouldnt ask you to roll a new version for me but > would this be something I could attempt myself by > following docs or would I need a lot more > experience? The quickest way is to find this ix4-200d box stock u-boot mtd0 backup somewhere. Or somebody has this box and willing to post a serial boot log, and dump the mtd0 partition on flash to a file. You will beby bodhi - uBoot
I will release new uboot for DS116/RS816. So all old installation methods from raw sectors will be obsolete.by bodhi - Debian
> > Do you think there is any way sheevaplug could be > manipulated to run from nvram? Why sheevaplug? do you mean this ix4-200d kernel files can be stored in flash? Or do you mean perhaps I can roll new u-boot for this box, since it is so close to the sheevaplug? I forgot the previous conversations (I have too many things going on) if you tried the sheevaplug u-boot, please liby bodhi - uBoot
> No joy with the "ide part" > > printenv below: > Marvell>> ide part > > no IDE devices available I forgot, also try scsi reset However, we don't really care much to find out, if you are going boot with USB kernel and HDD rootfs.by bodhi - uBoot
QuoteConclusion: The /boot-directory did not survive. Thus, if the file absolutely has to be in /boot, it should be symlinked. However, since the uBoot-env will have to be customized anyway, why not just read it from /etc/config right away for the OpenWrt boot? No it does not have to be in /boot. /etc/config is fine! Have you tried ubifsmount ubi0:mdt1by bodhi - Rescue System
> Is there any setenv commands I can run to "test" > gpt boot? ide reset ide part And it will show the capacity. But based on your log above, it already seems wrong. > I can boot from USB and all 4 x 6Tb are detected > but I'd rather move the install to disk so that I > can do OMV etc. So that is your fallback option. Create an Ext3 partititon on oby bodhi - uBoot
Other thought: we can also file a "bug" report at OpenWrt site to request for this OXNAS kernel processes the bootargs passed into by u-boot :)by bodhi - Rescue System
chessplayer, Sure, 1. Make sure the uEnv.txt will survive a sysupgrade. Most important to find out first! if it does not, then that will need to be a caveat for users to save the file and restore. 2. Boot with serial/neconsole. Interrupt u-boot and setenv mtdparts 'mtdparts=orion_nand:0x100000@0x0(u-boot),-@0x100000(ubi)' setenv partition 'nand0,0' ubi part ubby bodhi - Rescue System
This is my current thinking. A bit of experiement needed to script the u-boot envs during boot so that it can attach and read the ubi rootfs and load a text file from it. It sounds overkilled, but this is probably the only way, given the limitation of OpenWrt put on us. And we dont want to recompile its kernel. ========== Reference to the u-boot installation instruction: 2015.10 U-Bootby bodhi - Rescue System
chessplayer, What you see in Debian is a better implementation. We have new u-boot defines the envs location at 1MB in flash. The DTS has mtd0 as writeable. And the Debian kernel was configured to accept the bootargs (command line arguments). So everything works. On OpenWrt side, it's the other way around. The envs are at a different location. The MTDs are protected by read-only propeby bodhi - Rescue System
Hold on. By " with the calculated file I can read from, but not write to uBoot-env." Do you mean the envs from my u-boot installation?by bodhi - Rescue System