Hi Clare,
Thanks for your sharing.
allen
clare wrote:
Hi Allen,
My device is GTA01, the "original card" is the 512M card that came
with it. THe other card is a MicroSD (not SDHC) of 2G size.
I had a Celtune Monster image on the "other card" and that is the one
that got zeroed.
I had been using the original card back at the end of last year and it
had on it a rootfs.tar.gz from that time. I put it back in and am
using it now, but have put on it a later filesystem. Again a Monster
but I have not added the wikipedia, there is not much space which is
why I got the bigger one.
But since I wrote that Joachim Steiger said:
"but afaik thats a upstream bug in the linux kernel."
means it hits everybody doing suspend/resume with rootfs on sdcards with
linux atm.
So I will repartition the card and avoid suspend/resume
regards,
clare
On Tue, 6 May 2008, Allen Lin wrote:
Hi Clare,
I am Allen from Openmoko Testing Team. Could you describe more
specific like what device you are using? GTA01? GTA02? What image
inside? What do you mean original SD card? That means you have
another one to boot? Thanks!
allen
clare wrote:
Hi Cel,
Just repeating the story as it happened: I was using 080422 monster
booted from the SD card, with a Nokia battery in the train reading
html files with "links". On approaching end of journey, I went to
the little section power management (?) and selected "dim then
lock." I had previously set gsm and bluetooth to off. Well I
probably left it a bit too long, as it was dead when I came back to
it. So took out the battery and charged that up. But subsequently it
won't boot from SD card, just circles round and round. saying each
time as far as I can tell
Bad MBR sector signature 0.0000
Bad partition,
bad magic number
cfdisk says it has 2G of free space.
I ran dd on it and there seems to be quite a lot of data on it
otherwise.
I put in the original SD card and was immediately able to boot from
that,
as it had a filesystem I think one of yours dated Nov 30, which does
phone calls very well.
It worries me though that the DIM then LOCK should have had such a
dramatic effect. What else might have caused loss of partition table?
thanks
clare
