7439Re: [nslu2-general] NSLU2 Major Problem
- Jun 24, 2008In message <86714D25-0E53-48B7-86EA-22D707EE96D6@...>, Eike Lang
>Now 99% certain that the hardware is OK again. Latest problem looks to
>Am 24.06.2008 um 20:54 schrieb Robert Hammond:
>[nslu2 faulty psu]
>> Yes indeed. Fault diagnosed as low voltage when the disk is plugged
>> So I followed a hint from the Wiki which infers changing a large
>> capacitor in the power supply and the voltage now seems stable at
>> exactly 5 volts, with or without load (will monitor this, seems to be
>> still stable after 2 hours running).
>Ah, I didn't see that hint back then. I just went ahead and bought a
>PSU with the appropriate rating which set me back $15'ish.
>> And my NSLU2 seems to boot up correctly.
>> But it is now broken in another way. I cannot see it over the
>I'm 99% sure the hardware is ok - until I found the source of my
>was trying to run my slug for extended periods of time with no adverse
>effects (except not working before the psu was replaced).
>I'm not sure the config tool will just any slug or only those in
>As a non-windows person I'd suggest putting the slug in upgrade mode and
>trying upslug2 in discover mode. After trying that you can power-cycle
>slug and nothing will be changed. If that works my guess would be that
>network config on the slug got borked.
>Before you try all this you should give the default IP of 192.168.1.77
>a spin (if you haven't done so already).
be corruption in the sysconf, but have managed to get it back to life
without re-setting the sysconf using many Wiki pages for reference. The
procedure used is posted below, perhaps may be of use to others who
come across this page.
1. Used a direct crossover cable connection from my XP box, also
involves changes to the XP network settings.
2. Checked I could ping redboot 192.168.0.1 to prove the ether port is
3. Because my NSLU2 seemed to fully boot I used the 20 second long
reset (button on back of unit), which set the IP back to default
22.214.171.124 followed by a re-boot.
4. Managed to enter the management web pages 126.96.36.199, everything
very strange here such as Chinese character set selected and strange
messages about broken Samba i.e. all settings total garbage apart from
network IP. Used the web page 'reset to factory defaults' button which
mended many things (had to do this twice, firstly seemed to fix the
Samba warnings, second time sorted out many other corrupted fields).
5. Re-configured many necessary network settings in the management web
6. Plugged the unit back into the router and re-booted disk less
6. Enabled Telnet and logged in as root/uNSLUng
7. Had to delete /var/run/usb_det.pid and re-boot again followed by
another telnet as above.
8. Hot plugged in the disk and waited for it to mount. With my disk it
normally takes at least 30 minutes to hot mount properly, (something to
do with the quota function/disk size/qty of files etc).
9. Followed the Wiki procedure to check for disk errors, a small
number of orphaned inodes fixed (always seems to be about five to fix
with the unslung firmware).
10. Re-booted with disk attached successfully and it seems to be back
with no apparent functionality loss or loss of data.
Unsure if it is totally back, suppose time will tell.
In the UK, Maplin's seem to have a compatible power supply for less than
£10, if it looks to be going down again I will instantly bin it and go
for a new one hopefully before any damage to NSLU2 or disk files.
Also I suppose the only way to guarantee a corruption free sysconf is to
follow the Wiki procedure to wipe and re-generate it but I prefer to
keep anything to do with reboot access/flashing firmwares back as a last
- << Previous post in topic