NAS542 bootloop after Debian/OMV (get a new firmware) (solved)

Triu

Yes, found out that the main reason is 540 is for America and 542 for the European market. 540 came out 2015 and the other one year after, but don't think Zyxel will drop support for that one earlier than the other, cause as you mentioned, they are the same.

Tried to get a 542 firmware on it but now it is refused. So i will let it as it is.

I'm a little bit scared about the last days and happy that it works again.

Mijzelf

Yes, found out that the main reason is 540 is for America and 542 for the European market.

I think it's a bit different. The 540 was put on the market supporting volumes up to 2048TiB. With the update to firmware 5.10 that was dropped to 16TiB, I have written about the internals here.

I think they got a lot of complaints about the 540 not longer supporting the advertised volume size, and they took it from the market, and introduced the 542 instead.

As far as I know the 542 is identical to the 540 in all measures, except for the harddisk sledges, which changed color from orange to grey.

Triu

Found it on youtube, commented by Zyxel in the third answer:
https://www.youtube.com/watch?v=O--SPdrQu68
Also the power supply seems to be different.
On my 542 it is 19V, on 540 it should be 12V, found here:
http://naschart.com/compare/zyxel-nas540-nas540-151,zyxel-nas542-nas542-149

But anyhow, as long as the firmware has the same functionality, everything is good.

Triu

Unfortunately....i pushed the reset button to do a factory reset (3 beeps). After that it didn't come up, only the anoying beeber which doesn't stop. I connected with serial again and saw that it has problems with a corrupt NAND Flash.

So i tried to do the same as before, loaded the uImage.520ABAG1C0.bin but with the same result.

(Serial Log is attached)

Mijzelf, i hope you can help me again.

Maybe i really should try to get a NAS542 firmware on it to have no problems with the restet button?

Mijzelf

It has worked for some weeks? Strange. A factory reset doesn't touch mtdblock5. Here is the reset script (/usr/local/bin/reset_and_reboot.sh):

#!/bin/sh<br>if [ -e /etc/zyxel/pkg_conf/status ]; then<br>&nbsp;&nbsp;&nbsp; cp -a /etc/zyxel/pkg_conf/status /i-data/.system/zy-pkgs/status.packageinstinfo<br>&nbsp;&nbsp;&nbsp; sync<br>&nbsp;&nbsp;&nbsp; sleep 1<br>fi<br>rm -rf /i-data/.media/twonkymedia<br>rm -rf /etc/zyxel/conf<br>rm -rf /etc/zyxel/webdav<br>rm -rf /etc/zyxel/storage/*<br>umount /etc/zyxel<br>ubirmvol /dev/ubi3 -N ubi_config<br>ubidetach -m 3<br>rm -rf /etc/zyxel/<br>#remove desktop config by hazel<br>[ -e "/i-data/.system/guicfg" ] && rm -rf /i-data/.system/guicfg<br>reboot<br><br>

Nor does it switch to 'the other half' of the flash.

What concerns me in your log is the 'Dual images are currupt !!!' message. That means it has already switched to 'the other half' and rebooted. So both mtdblock5 and mtdblock7 doesn't seem to contain a valid rootfs.

When you don't upload uImage, how does is bootlog  (or in other words, is the kernel also corrupted in flash?)

Triu

Hello Mijzelf,

thank you for helping again.

I didn't use it alot the last weeks, but a regular start was no problem.

I played a little bit with all settings, cause of the Debian which was installed before,

i don't know everything on stock Zyxel NAS.

That was the reason i wanted to reset it.

Attached is the log when doing absolutely nothing.

Mijzelf

That looks the same. So the kernel in flash is still valid. You can get a shell in the end? Can you mount an usb drive?

Triu

The kernel is still valid? Hmmm. The changed start parameters from the past can't disturb this?

He is still loading it, cause i didn't make me a copy to set it back.

commandline: console=ttyS0,115200n8, init=/etc/preinit pcie_gen1_only=yes  , mac_addr=,, ip=dhcp root=ubi0:rootfs ubi.mtd=2,2048 rootfstype=ubifs rw noinitrd mtdparts=spi0.0:256k(uloader)ro,512k(barebox)ro,256k(env);comcertonand:10M(config),10M(kernel1),110M(rootfs1),10M(kernel2),110M(rootfs2),-(reserved) usb3_internal_clk=yes ip=dhcp root=ubi0:rootfs ubi.mtd=2,2048 rootfstype=ubifs rw noinitrd mtdparts=spi0.0:256k(uloader)ro,512k(barebox)ro,256k(env);comcertonand:10M(config),10M(kernel1),110M(rootfs1),10M(kernel2),110M(rootfs2),-(reserved) usb3_internal_clk=yes

For the USB stick, i think i can. I see it with "lsusb" and i know i can create a folder. The "mount" command is also there. What should i put on the stick?

Is there a way to disable this annoying buzzer?

Mijzelf

<quote>

The kernel is still valid? Hmmm. The changed start parameters from the past can't disturb this?

</quote>

The kernel is valid or not. The container has a checksum, which is actually checked, so for all practical purposes you can assume the kernel is valid when bootbox doesn't complain. And the compilebox&date is the same as in your previous log, so it's the same kernel.

The environment is not part of the kernel, but of bootbox environment. And that seems OK too, afaics. The most important part, the flash layout is in the kernel commandline.

<quote>

What should i put on the stick?

</quote>

mdtblock5 is called corrupt because it doesn't contain a sysdisk.img.gz. According to the log there is nothing wrong with the partition (no ubi errors), so you could try to copy sysdisk.img.gz on it.

The partition is mounted on /firmware/mnt/nand/. I uploaded the sysdisk.img.gz belonging to that kernel here: http://zyxel.diskstation.eu/Users/Mijzelf/NAS/542/

<quote>

Is there a way to disable this annoying buzzer?

</quote>

https://homeforum.zyxel.com/discussion/comment/5933/#Comment_5933

Triu

Great to turn off the buzzer.

But can't copy to destination, cause read only.

cp: can't create '/firmware/mnt/nand/sysdisk.img.gz.520ABAG1C0': Read-only file system<br>