maemo.org - Talk

maemo.org - Talk (https://talk.maemo.org/index.php)
-   Troubleshooting (https://talk.maemo.org/forumdisplay.php?f=6)
-   -   Reboot loop / apr-get upgrade / 20 Jun 2006 firmware (https://talk.maemo.org/showthread.php?t=2492)

Kralisec 2006-07-16 20:16

Reboot loop / apt-get upgrade / 20 Jun 2006 firmware
 
Hi,

After a look to forum, I solve my mistral-beta problem
and add new repository to my list,
after this I application list give 1 update: ok, no problem.

But "apt-get install ssh" show me 19 applications to upgrade,
apt-get upgrade propose to begin with 17 applications,
I said yes and after a 3mn of download my 770 shutdown ...


Now I'm in infinite boot loop mode with about 30s time usable ...

Since I have enough time to copy a script from mmc card
and start it, I do a ps -ef > /media/mmc1/result.txt

this is result:

PID Uid VmSize Stat Command
1 root 484 S init [2]
2 root SWN [ksoftirqd/0]
3 root SW [watchdog/0]
4 root SW< [events/0]
5 root SW< [khelper]
6 root SW< [kthread]
22 root SW< [kblockd/0]
36 root SW< [omap_uwire]
42 root SW< [khubd]
93 root SW [pdflush]
94 root DW [pdflush]
96 root SW< [aio/0]
95 root SW [kswapd0]
171 root SW< [cqueue/0]
172 root SW< [kseriod]
205 root SW [mtdblockd]
239 root SW< [krfcommd]
248 root SW [mmcqd]
278 root 540 R < dsme -d -l syslog -v 4 -p /usr/lib/dsme/libst
283 root 540 S dsme -d -l syslog -v 4 -p /usr/lib/dsme/libst
284 root 540 S dsme -d -l syslog -v 4 -p /usr/lib/dsme/libst
285 root 312 S /usr/bin/bme
333 root DWN [jffs2_gcd_mtd4]
352 root 416 S < /sbin/udevd --daemon
565 nobody 652 S < /usr/sbin/dnsmasq -k -i lo -a 127.0.0.1 -z
579 messageb 1320 S < /usr/bin/dbus-daemon --system
584 root 1492 S < /sbin/mce
588 messageb 1468 D /usr/lib/gconf2/gconfd-2
594 root 3248 S < /usr/bin/Xomap -mouse /dev/input/mouse0 -mous
618 user 236 S < /usr/sbin/temp-reaper
621 user 1128 S < /usr/bin/dbus-daemon --session
625 user 2628 S < /usr/lib/sapwood/sapwood-server
629 user 2068 S < /usr/bin/matchbox-window-manager -theme defau
640 root 1112 S /usr/sbin/dsp_dld -p --disable-restart -c /li
659 root 1128 S < /usr/sbin/hcid -n
661 root 492 S < /usr/sbin/sdpd -n
690 root 1016 S < /usr/bin/bme-dbus-proxy -N
707 root 1364 S /usr/bin/osso_hss
712 root 632 S /usr/bin/esd
733 root 1220 S < /usr/bin/btcond -l
751 root 1500 S < /usr/sbin/ke-recv
757 root 1288 S < /usr/sbin/wlancond
760 nobody 1056 S < /usr/bin/btpin -l
763 root 1484 S < /usr/sbin/icd
784 user 2112 S /usr/libexec/dbus-vfs-daemon
786 user 1844 S < /usr/bin/maemo-launcher --daemon --quiet
804 user 2112 S /usr/libexec/dbus-vfs-daemon
805 user 2112 S /usr/libexec/dbus-vfs-daemon
838 user 356 S < /usr/bin/maemo_af_desktop
839 user 8748 S < /usr/bin/maemo_af_desktop
845 user 356 S < /usr/bin/osso-connectivity-ui-conndlgs
846 user 4072 S < /usr/bin/osso-connectivity-ui-conndlgs
849 user 356 S /usr/bin/hildon-input-method
851 user 7132 S /usr/bin/osso-xterm
854 user 6172 S /usr/bin/hildon-input-method
857 user 600 S < /usr/bin/clipboard-manager
861 root 1384 S /usr/bin/ias
888 systemui 5040 S < /usr/bin/systemui
889 user 612 S gnome-pty-helper
892 user 708 S sh
893 user 7132 S /usr/bin/osso-xterm
894 user 7132 S /usr/bin/osso-xterm
912 root 1740 S /usr/bin/osso_alarm
919 root 68 S /usr/bin/retutime -s
920 user Z < [osso-media-serv]
921 user 528 S /bin/sh ./t
922 user 620 R ps -ef



Is'it some kill I can do to stop the expected shutdown ?

or someone can send me a script I can start to see mode deeply what arrive ...


without any information, I will flash back to original final 2006,
but it's also time to see what arrive ...


Any idea ???

Thank you,
Laurent.

Kralisec 2006-07-17 19:20

Hi,

Someone can post his own "ps -ef" result from a fresh booted 770 that I can compare ?
(from 30 Jun 2006)


Thanks,
Laurent.

HardCoder 2006-07-17 19:59

1 Attachment(s)
I had just the same problem i couple of days ago...
I just reflashed the latest firmware to solve the problem. Thats the price to pay sometimes when living on the bleeding edge. :)

Kralisec 2006-07-18 19:48

Quote:

Originally Posted by HardCoder
I had just the same problem i couple of days ago...
I just reflashed the latest firmware to solve the problem. Thats the price to pay sometimes when living on the bleeding edge. :)

Yea ... I know ... but I expect to be more able to trace into the problem :)

Thank you for your list, but that didn't help me,


But I have very very big problem with reflash I didn't expect,
solution was:
export USB_DEVFS_PATH=/dev/bus/usb

I see solution http://l3x.net/imwiki/Wiki.jsp?page=Main

I will a make a post with better explanation.

Good day,
Laurent.

hallgreng 2006-07-18 20:51

I had this problem yesterday.
I believe it is the maemo-launcher update that causes the problem, mainly because it cannot upgrade properly (as the table reboots when it senses maemo-launcher not running).

flash your 770 like so:
$./flasher-2.0 --set-rd-flags=no-lifeguard-reset

That will keep the 770 from rebooting when something critical has crashed or has been shut down for an upgrade.
You can then use --clear-rd-flags=no-lifeguard-reset to put it back to normal, but I havent at this time.

Kralisec 2006-07-18 21:32

Now I have more information and I can reproduce my crash/reboot :(

- I flash with 30 June
- add repository maemo-hackers.org/apt mistral main (to install xterm)
- install xterm
- activate R&D mode (to modify /usr/bin/gainroot)
- disable R&D mode
- add some repository
- apt-get update
- apt-get upgrade
upgrade propose 14 update: I said Yes, system crash with libtelepathy install.

this my repository list I copy from some post: (I don't remeber where)

TAKE CARE OR YOU WILL CRASH YOUR SYSTEM, YOU HAVE BEEN WARNED !!!


deb http://repository.maemo.org/contrib/ mistral free non-free
deb http://repository.maemo.org/ mistral free non-free
deb http://only.mawhrin.net/fbreader/maemo/ mistral user
deb http://www.kernelconcepts.de/~fuchs/nokia770/ex_2006/ ./
deb http://www.mulliner.org/nokia770/repository/ maemo2 free
deb http://maemo.org.br/platform/apt/ python python
deb http://zeus.rm-fr.net/~skyhusker/maemo/ mistral user
deb http://www.oakcourt.dyndns.org/maemo/ ./


Thank you for your tips hallgreng,
I will make a new try with it and see if I can have libtelepathy version


asap,
Laurent.

hallgreng 2006-07-18 21:40

Please post me the output of
$dpkg -l | grep telepathy

my output from a working tablet is as follows:
Code:

# dpkg -l | grep telepathy
ii  libtelepathy                                  0.0.14-0osso1maemo1          Telepathy client library
ii  telepathy-feed                                0.10                        Telepathy presence feed for Galago
ii  telepathy-gabble                              0.1.41-0osso1                Telepathy Gabble
ii  telepathy-voip-engine                          0.1.12-0osso1                Telepathy voice over IP engine

If you have an earlier version, then it should be upgradable (albeit with a little hassle and that flasher flag).

I am beginning to wonder if I should apt-get from the command line at all any more. I am a hardcore Debian user, so I prefer to use apt command-line tools to do updates, but I have run into problems with maemo-launcher before, apparently libtelepathy (that was in the list of upgraded packages prior to my problems yesterday), and I needed to enable development flags in the firmware to fix them.
I am not against doing heavy lifting, but the 770 is one of those devices that I like to "just work" and not have to fool around with it to do something.

Kralisec 2006-07-18 23:18

Next ...
 
This is my libtelepathy version before I try to upgrade:
Code:

home/user # dpkg -l | grep telepathy
ii  libtelepathy                              0.0.14-0osso1        Telepathy client library
ii  telepathy-feed                        0.10                        Telepathy presence feed for Galago
ii  telepathy-gabble                    0.1.41-0osso1        Telepathy Gabble
ii  telepathy-voip-engine            0.1.12-0osso1        Telepathy voice over IP engine

apt-get update & apt-get upgrade before I add my previous repository list
Code:

# apt-get update
Get:1 http://maemo-hackers.org mistral Release.gpg [189B]
Get:2 http://catalogue.tableteer.nokia.com mistral Release.gpg [189B]
Ign http://catalogue.tableteer.nokia.com mistral Release.gpg
Hit http://maemo-hackers.org mistral Release
Ign http://maemo-hackers.org mistral Release
Hit http://catalogue.tableteer.nokia.com mistral Release
Hit http://maemo-hackers.org mistral/main Packages
Hit http://catalogue.tableteer.nokia.com mistral Release
Hit http://catalogue.tableteer.nokia.com mistral/user Packages
Hit http://catalogue.tableteer.nokia.com mistral/user Packages
Fetched 190B in 7s (26B/s)
Reading package lists... Done
W: GPG error: http://maemo-hackers.org mistral Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 3A312B21B9E81572
W: You may want to run apt-get update to correct these problems
/ #


/ # apt-get upgrade
Reading package lists... Done
Building dependency tree... Done
The following packages have been kept back:
  libosso-gnomevfs2-0 libosso-gnomevfs2-common
0 upgraded, 0 newly installed, 0 to remove and 2 not upgraded.
/ #

apt-get update & apt-get upgrade after I add my previous repository list
Code:

/home/user # apt-get update
Get:1 http://maemo-hackers.org mistral Release.gpg [189B]
Ign http://only.mawhrin.net mistral Release.gpg
Ign http://repository.maemo.org mistral Release.gpg
Ign http://www.kernelconcepts.de ./ Release.gpg
Ign http://repository.maemo.org mistral Release.gpg
Hit http://maemo-hackers.org mistral Release
Ign http://maemo.org.br python Release.gpg
Ign http://maemo-hackers.org mistral Release
Ign http://www.mulliner.org maemo2 Release.gpg
Ign http://zeus.rm-fr.net mistral Release.gpg
Get:2 http://www.oakcourt.dyndns.org ./ Release.gpg [189B]
Get:3 http://catalogue.tableteer.nokia.com mistral Release.gpg [189B]
Ign http://catalogue.tableteer.nokia.com mistral Release.gpg
Get:4 http://repository.maemo.org mistral Release [228B]
Hit http://maemo-hackers.org mistral/main Packages
Ign http://zeus.rm-fr.net mistral Release
Get:5 http://www.oakcourt.dyndns.org ./ Release [730B]
Hit http://catalogue.tableteer.nokia.com mistral Release
Ign http://www.oakcourt.dyndns.org ./ Release
Ign http://www.kernelconcepts.de ./ Release
Get:6 http://only.mawhrin.net mistral Release [397B]
Get:7 http://repository.maemo.org mistral Release [248B]
Get:8 http://www.mulliner.org maemo2 Release [97B]
Ign http://maemo.org.br python Release
Get:9 http://zeus.rm-fr.net mistral/user Packages [611B]
Hit http://catalogue.tableteer.nokia.com mistral Release
Get:10 http://www.oakcourt.dyndns.org ./ Packages [2079B]
Get:11 http://only.mawhrin.net mistral/user Packages [991B]
Get:12 http://repository.maemo.org mistral/free Packages [6780B]
Get:13 http://repository.maemo.org mistral/non-free Packages [20B]
Hit http://catalogue.tableteer.nokia.com mistral/user Packages
Get:14 http://www.kernelconcepts.de ./ Packages [5178B]
Get:15 http://repository.maemo.org mistral/free Packages [84.0kB]
Hit http://catalogue.tableteer.nokia.com mistral/user Packages
Get:16 http://maemo.org.br python/python Packages [4355B]
Get:17 http://www.mulliner.org maemo2/free Packages [1309B]
Get:18 http://repository.maemo.org mistral/non-free Packages [10.4kB]
Fetched 118kB in 37s (3167B/s)
Reading package lists... Done
W: GPG error: http://maemo-hackers.org mistral Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 3A312B21B9E81572
W: GPG error: http://www.oakcourt.dyndns.org ./ Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 86E33E67AD9AE76A
W: You may want to run apt-get update to correct these problems
/home/user #


home/user # apt-get upgrade
Reading package lists... Done
Building dependency tree... Done
The following packages have been kept back:
  libosso-gnomevfs2-0 libosso-gnomevfs2-common
The following packages will be upgraded:
  certs clinkc0 gcc-3.4-base gst-plugins-farsight lessertunjo0 libcst libgalago1
  libgcc1 libjinglebase0 libjinglep2p0 libstdc++6 libtelepathy maemo-launcher
  osso-sounds-ui
14 upgraded, 0 newly installed, 0 to remove and 2 not upgraded.
Need to get 2345kB of archives.
After unpacking 1138kB of additional disk space will be used.
Do you want to continue [Y/n]? y
WARNING: The following packages cannot be authenticated!
  libgcc1 gcc-3.4-base libstdc++6 libcst certs clinkc0 libjinglebase0 libjinglep2p0
  gst-plugins-farsight lessertunjo0 libgalago1 libtelepathy maemo-launcher
  osso-sounds-ui
Install these packages without verification [y/N]?


This is update downloaded:
Get:12 http://repository.maemo.org mistral/free libtelepathy 0.0.14-0osso1maemo1 [20.8kB]


and still crash :( :( :(

But reboot looks stable :) and telepathy is updated to:
Code:

iiU  libtelepathy                    0.0.14-0osso1maemo1 Telepathy client library
ii  telepathy-feed                0.10                                  Telepathy presence feed for Galago
ii  telepathy-gabble            0.1.41-0osso1                Telepathy Gabble
ii  telepathy-voip-engine    0.1.12-0osso1                Telepathy voice over IP engine


Final:
Code:

/ # apt-get upgrade
E: dpkg was interrupted, you must manually run 'dpkg --configure -a' to correct the problem.
/ # apt-get --configure -a
E: Command line option --configure is not understood
/ # dpkg -l | grep telepathy >> /media/mmc1/res1
/ # dpkg -l | grep telepathy
iU  libtelepathy                                  0.0.14-0osso1maemo1          Telepathy client library
ii  telepathy-feed                                0.10                        Telepathy presence feed for Galago
ii  telepathy-gabble                              0.1.41-0osso1                Telepathy Gabble
ii  telepathy-voip-engine                          0.1.12-0osso1                Telepathy voice over IP engine
/ # dpkg --configure -a
Setting up libcst (1.6.34maemo1) ...

Setting up libjinglebase0 (0.3.0-0osso1.2maemo1) ...

Setting up libjinglep2p0 (0.3.0-0osso1.2maemo1) ...

Setting up libgalago1 (0.5.0-0osso3maemo1) ...

Setting up libtelepathy (0.0.14-0osso1maemo1) ...

Setting up clinkc0 (1.0-27.1) ...

Setting up lessertunjo0 (0.1.2-3maemo1) ...

Setting up certs (1.5.4maemo1) ...
gconftool-2: I've been haxored to use xml::/etc/gconf/gconf.xml.defaults as the config source.
Attached schema `/schemas/apps/osso/certman/filename' to key `/apps/osso/certman/filename'
Installed schema `/schemas/apps/osso/certman/filename' for locale `C'

Setting up gst-plugins-farsight (0.10.1-0osso2maemo1) ...
/ # apt-get upgrade
Reading package lists... Done
Building dependency tree... Done
The following packages have been kept back:
  libosso-gnomevfs2-0 libosso-gnomevfs2-common
The following packages will be upgraded:
  maemo-launcher osso-sounds-ui
2 upgraded, 0 newly installed, 0 to remove and 2 not upgraded.
1 not fully installed or removed.
Need to get 0B/1330kB of archives.
After unpacking 868kB of additional disk space will be used.
Do you want to continue [Y/n]?
WARNING: The following packages cannot be authenticated!
  maemo-launcher osso-sounds-ui
Install these packages without verification [y/N]? y

/bin/sh: /usr/sbin/dpkg-preconfigure: not found
(Reading database ... 10506 files and directories currently installed.)
Preparing to replace maemo-launcher 0.17-1.1 (using .../maemo-launcher_0.17-1.1sdk1_armel.deb) ...
Unpacking replacement maemo-launcher ...
Preparing to replace osso-sounds-ui 1.2-4 (using .../osso-sounds-ui_1.2-4sdk1_all.deb) ...
Unpacking replacement osso-sounds-ui ...
Setting up maemo-launcher (0.17-1.1sdk1) ...

Configuration file `/etc/init.d/maemo-launcher'
 ==> File on system created by you or by a script.
 ==> File also in package provided by package maintainer.
  What would you like to do about it ?  Your options are:
    Y or I  : install the package maintainer's version
    N or O  : keep your currently-installed version
      D    : show the differences between the versions
      Z    : background this process to examine the situation
 The default action is to keep your current version.
*** maemo-launcher (Y/I/N/O/D/Z) [default=N] ? Y
Installing new version of config file /etc/init.d/maemo-launcher ...
Starting Maemo Launcher: maemo-launcher start failed.

Setting up osso-sounds-ui (1.2-4sdk1) ...
/ #


Now system looks good now :)
but I can't go out from R&D --set-rd-flags=no-lifeguard-reset mode
cause reboot loop arrive :( :( :(


hallgreng: did you test to go out from R&D to see what appends ?
Cause it's not just an install point problem, it's stability system problem now I guess ...


time for a bug report ?


Good day,
Laurent.

hallgreng 2006-07-19 00:03

Looks like your situation was the same as mine, I had apt asking about maemo-launcher during upgrade, though I opted to keep the original config file.

I did not disable R&D mode nor have i cleared the no-lifeguard-reset flag.
I may do that tomorrow if necessary to test.

I looked through dmesg and around the end of bootup and this is all that looks interesting:
Code:

~ $ dmesg | grep -A 11 36.660
[  36.660705] aic23 powering up
[  36.664642] aic23_init_power() done
[  43.419555] aic23 powering down
[  50.039581] aic23 powering up
[  50.043670] aic23_init_power() done
[  50.045440] omapdsp: mmap info: vmadr = 40016000, padr = 12a10000, len = 2000
[  52.419555] aic23 powering down
[  67.910156] aic23 powering up
[  67.914123] aic23_init_power() done
[  68.450836] omap-hw-nand: DMA timeout after 1148 ms, max. seen latency 1148 ms
[  70.974273] aic23 powering down
[  326.545074] Loading 3825.arm firmware

Namely the "aic23 powering up and down (I dont know what this is or if it is normal behavior)
and the DMA timeout.
Does anyone have any knowledge of these two things? Due to the nature of the device, logs are sparce. dmesg is the best I can think of to gather information in this case...

spycedtx 2006-07-19 01:54

same issue here
 
I did the exact same steps.. got stuck in the reboot loop once I'd installed (and forced the rest of the install to finish). Popped the battery, rd mode to no lifeguard, and the reboot cycle goes away.. and returns again if I take out of R&D/no lifeguard.

I don't know if it's a symptom of R&D mode ( my understanding is that it doesn't cause any performace differences), or maybe I'm just imagining things.. but the overall system suddenly feels speedier since doing this minor upgrade.

hallgreng 2006-07-19 22:23

Well, I noticed today that I cannot play video or audio files any more, so I suppose the problem lies with the aic23 powering up and down.
I think that tomorrow I am going to go ahead and flash the official 2006 root directory onto my 770 and avoid using apt on the command line to upgrade...

spycedtx 2006-07-20 08:06

ditto on dead music/video
 
Quote:

Originally Posted by hallgreng
Well, I noticed today that I cannot play video or audio files any more, so I suppose the problem lies with the aic23 powering up and down.
I think that tomorrow I am going to go ahead and flash the official 2006 root directory onto my 770 and avoid using apt on the command line to upgrade...


I did the upgrade on telepathy and such as posted in the prior messages. This, of course, required using apt-get and fixing the install a few times (as other people noted).

And now that I've mounted remote windows shares, I've noticed audio and video no longer works for me, either. I even did a 'shutdown -r now', and attempted to play a local audio/video before even making a network connection. Always get "unable to perform operation. try again".

I'm wondering what exactly "broke" in updating those libraries.

BigBob 2006-07-20 09:57

Quote:

Originally Posted by spycedtx
I did the upgrade on telepathy and such as posted in the prior messages. This, of course, required using apt-get and fixing the install a few times (as other people noted).

And now that I've mounted remote windows shares, I've noticed audio and video no longer works for me, either. I even did a 'shutdown -r now', and attempted to play a local audio/video before even making a network connection. Always get "unable to perform operation. try again".

I'm wondering what exactly "broke" in updating those libraries.

Hi,

Same for me, it's really annoying since i use my time to constantly repair my toy.

That mean we can't use this device properly.

That really sux ...

Kralisec 2006-07-20 13:58

Same things for me: no way to play Discovery internal video,

this is my dmesg | grep aic
Code:

[  22.237487] aic23 powering up
[  22.241394] aic23_init_power() done
[  25.046630] aic23 powering down
[  32.479309] aic23 powering up
[  32.483184] aic23_init_power() done
[  39.046600] aic23 powering down
[  45.193847] aic23 powering up
[  45.197814] aic23_init_power() done
[  47.579681] aic23 powering down
[  66.888153] aic23 powering up
[  66.892089] aic23_init_power() done
[  69.960662] aic23 powering down


dmesg
Code:

ialized vram=1048576 pixclock 21940 kHz hfreq 24.9 kHz vfreq 51.0 Hz
[    1.966705] OMAP Watchdog Timer: initial timeout 60 sec
[    1.967407] Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
[    1.977020] loop: loaded (max 8 devices)
[    1.978271] PPP generic driver version 2.4.2
[    1.978881] PPP Deflate Compression module registered
[    1.979003] PPP BSD Compression module registered
[    1.979095] tun: Universal TUN/TAP device driver, 1.6
[    1.979187] tun: (C) 1999-2004 Max Krasnyansky <maxk@qualcomm.com>
[    1.981994] Clock enabled to MCBSP1 & 3
[    1.982116] TLV320AIC23 I2C version 1.9 (02-Mar-2006)
[    1.982238] Retu power button driver initialized
[    1.982330] retu: Registering interrupt 0 for device PwrOnX
[    1.983154] input: retu-pwrbutton as /class/input/input0
[    1.984344] retu: Registering interrupt 2 for device RTCS
[    1.984588] retu: Registering interrupt 5 for device RTCA
[    1.985870] Retu watchdog driver initialized
[    1.986358] omap-hw-nand: OMAP NAND Controller rev. 1.1
[    1.986541] NAND device: Manufacturer ID: 0xec, Chip ID: 0xa1 (Samsung NAND 128MiB 1,8V 8-bit)
[    1.986724] omap-hw-nand: using PSC values 2, 2, 3
[    1.986846] Scanning device for bad blocks
[    1.986968] Bad eraseblock 0 at 0x00000000
[    2.022918] 5 cmdlinepart partitions found on MTD device omap-nand
[    2.023071] Creating 5 MTD partitions on "omap-nand":
[    2.023193] 0x00000000-0x00020000 : "bootloader"
[    2.024780] 0x00020000-0x00080000 : "config"
[    2.026275] 0x00080000-0x00280000 : "kernel"
[    2.027801] 0x00280000-0x00480000 : "initfs"
[    2.029235] 0x00480000-0x08000000 : "root"
[    2.030700] udc: OMAP UDC driver, version: 4 October 2004
[    2.031555] udc: OMAP UDC rev 6.1, Mini-AB
[    2.031677] udc: hmc mode 16, tahvo-usb transceiver
[    2.031829] udc: fifo mode 0, 1832 bytes not used
[    2.032623] mice: PS/2 mouse device common for all mice
[    2.033203] OMAP Keypad Driver
[    2.034454] input: omap-keypad as /class/input/input1
[    2.035705] ads7846 spi2.0: touchscreen, irq 175
[    2.036926] input: ADS784x Touchscreen as /class/input/input2
[    2.041442] Advanced Linux Sound Architecture Driver Version 1.0.11rc2 (Wed Jan 04 08:57:20 2006 UTC).
[    2.042694] ALSA device list:
[    2.042785]  No soundcards found.
[    2.042877] Netfilter messages via NETLINK v0.30.
[    2.043487] NET: Registered protocol family 2
[    2.101745] IP route cache hash table entries: 1024 (order: 0, 4096 bytes)
[    2.101898] MMC2: Command timeout, CMD0
[    2.103271] TCP established hash table entries: 4096 (order: 2, 16384 bytes)
[    2.103637] TCP bind hash table entries: 4096 (order: 2, 16384 bytes)
[    2.103973] TCP: Hash tables configured (established 4096 bind 4096)
[    2.104064] TCP reno registered
[    2.104309] ip_tables: (C) 2000-2006 Netfilter Core Team
[    2.148773] MMC2: Command timeout, CMD0
[    2.148895] MMC2: Command timeout, CMD0
[    2.171569] TCP bic registered
[    2.171752] NET: Registered protocol family 1
[    2.171905] NET: Registered protocol family 17
[    2.172027] Bluetooth: L2CAP ver 2.8
[    2.172119] Bluetooth: L2CAP socket layer initialized
[    2.172302] Bluetooth: RFCOMM socket layer initialized
[    2.172454] Bluetooth: RFCOMM TTY layer initialized
[    2.172546] Bluetooth: RFCOMM ver 1.7
[    2.172637] Bluetooth: HIDP (Human Interface Emulation) ver 1.1
[    2.172729] Disabling unused clock "uart3_ck"...  done
[    2.172851] Disabling unused clock "uart1_ck"...  done
[    2.172943] FIXME: Clock "tc2_ck" seems unused
[    2.173034] Skipping reset check for DSP domain clock "dsptim_ck"
[    2.173126] Skipping reset check for DSP domain clock "dspxor_ck"
[    2.173248] Skipping reset check for DSP domain clock "dspper_ck"
[    2.173339] FIXME: Clock "ck_dpll1out" seems unused
[    2.174041] Bootup reason: pwr_key
[    2.174194] OMAP GPIO switch handler initializing
[    2.177398] headphone (GPIO 14) is now disconnected
[    2.177551] speaker (GPIO 58) is now disconnected
[    2.177642] prot_shell (GPIO 5) is now open
[    2.177734] bat_cover (GPIO 12) is now closed
[    2.178009] OMAP Random Number Generator ver. 40
[    2.178466] 128 words of entropy generated
[    2.224029] mmcblk0: mmc0:0001 000000 1003520KiB
[    2.224670]  mmcblk0: p1 p2 p3
[    2.453491] Empty flash at 0x001cd7c8 ends at 0x001cd800
[    2.459106] VFS: Mounted root (jffs2 filesystem).
[    2.459594] Freeing init memory: 92K
[    3.839263] tahvo: Registering interrupt 7 for device
[    3.840087] retu: Registering interrupt 8 for device
[    3.842346] retu: Registering interrupt 1 for device
[    3.842803] tahvo: Registering interrupt 1 for device
[    4.205139] Selected PERIPHERAL mode: no gadget driver, powering off.
[    5.681823] umac: module license 'Proprietary' taints kernel.
[    5.901336] CX3110x chip variant: STLC4370
[    6.021453] CX3110x: firmware version: 2.13.0.0.a.13.9
[    6.021667] Loaded CX3110x driver, version 0.8
[    9.762420] JFFS2 notice: (371) read_dnode: header CRC failed on node at 0x7ae97f0: read 0xffffffff, calculated 0x5dc85953
[  20.651489] omapdsp: mapping in ARM MMU, v=0xe0fff000, p=0x13c4a000, sz=0x1000
[  20.690704] omapdsp: mapping in ARM MMU, v=0xe0100000, p=0x12900000, sz=0x100000
[  20.690979] omapdsp: mapping in ARM MMU, v=0xe0200000, p=0x12800000, sz=0x100000
[  20.691314] omapdsp: mapping in ARM MMU, v=0xe0028000, p=0x13167000, sz=0x1000
[  20.691467] omapdsp: frame buffer export
[  20.691528] omapdsp: mapping in ARM MMU, v=0xe0300000, p=0x13d00000, sz=0x100000
[  20.691711] omapfb_notifier_cb(): event = READY
[  20.691772] omapfb_register_client(): success
[  20.692047] omapdsp: mapping in ARM MMU, v=0xe0400000, p=0x12f00000, sz=0x100000
[  20.692260] omapdsp: mapping in ARM MMU, v=0xe0500000, p=0x12e70000, sz=0x10000
[  20.692352] omapdsp: mapping in ARM MMU, v=0xe0510000, p=0x12e60000, sz=0x10000
[  20.692443] omapdsp: mapping in ARM MMU, v=0xe0520000, p=0x12e50000, sz=0x10000
[  20.692504] omapdsp: mapping in ARM MMU, v=0xe0530000, p=0x12e40000, sz=0x10000
[  20.692596] omapdsp: mapping in ARM MMU, v=0xe0540000, p=0x12e30000, sz=0x10000
[  20.692687] omapdsp: mapping in ARM MMU, v=0xe0550000, p=0x12e20000, sz=0x10000
[  20.692779] omapdsp: mapping in ARM MMU, v=0xe0560000, p=0x12e80000, sz=0x10000
[  20.692871] omapdsp: mapping in ARM MMU, v=0xe0570000, p=0x12e90000, sz=0x10000
[  21.109436] omapdsp: IPBUF configuration
[  21.109466]            512 words * 16 lines at 0xe0200000.
[  21.109710] omapdsp: found 4 task(s)
[  21.109985] omapdsp: task 0: name pcm0
[  21.110565] omapdsp: taskdev pcm0 enabled.
[  21.119293] omapdsp: task 1: name pcm1
[  21.119812] omapdsp: taskdev pcm1 enabled.
[  21.128814] omapdsp: task 2: name avsync
[  21.129333] omapdsp: taskdev avsync enabled.
[  21.138305] omapdsp: task 3: name audiopp
[  21.138916] omapdsp: taskdev audiopp enabled.
[  21.558776] hci_cmd_task: hci0 command tx timeout
[  21.558868] brf6150: Frame for non-running device
[  22.237487] aic23 powering up
[  22.241394] aic23_init_power() done
[  22.664733] omapdsp: mmap info: vmadr = 40016000, padr = 12f30000, len = 2000
[  25.046630] aic23 powering down
[  32.479309] aic23 powering up
[  32.483184] aic23_init_power() done
[  39.046600] aic23 powering down
[  45.193847] aic23 powering up
[  45.197814] aic23_init_power() done
[  45.199127] omapdsp: mmap info: vmadr = 40016000, padr = 12f10000, len = 2000
[  47.579681] aic23 powering down
[  66.888153] aic23 powering up
[  66.892089] aic23_init_power() done
[  69.960662] aic23 powering down
[  91.588378] omap-hw-nand: DMA timeout after 1187 ms, max. seen latency 1187 ms
[  93.730346] omap-hw-nand: DMA timeout after 1234 ms, max. seen latency 1234 ms


- I flash back to original 30 June 2006,
- install xterm with maemo-hackers.org/apt mistral main repository,
- update manually libtelepathy only with
dpkg -i ./libtelepathy_0.0.14-0osso1maemo1_armel.deb

Now internal Discovery video is ok, reboot is stable without R&D,
still powering up / powering down with aic like quoted in beginning, (dunno if it's important ...)


result: libtelepathy alone is not the problem :(


Good day,
Laurent.

Kralisec 2006-07-20 16:22

Quote:

Originally Posted by hallgreng
Looks like your situation was the same as mine, I had apt asking about maemo-launcher during upgrade, though I opted to keep the original config file.

I did not disable R&D mode nor have i cleared the no-lifeguard-reset flag.
I may do that tomorrow if necessary to test.

I looked through dmesg and around the end of bootup and this is all that looks interesting:
Code:

~ $ dmesg | grep -A 11 36.660
[  36.660705] aic23 powering up
[  36.664642] aic23_init_power() done
[  43.419555] aic23 powering down
[  50.039581] aic23 powering up
[  50.043670] aic23_init_power() done
[  50.045440] omapdsp: mmap info: vmadr = 40016000, padr = 12a10000, len = 2000
[  52.419555] aic23 powering down
[  67.910156] aic23 powering up
[  67.914123] aic23_init_power() done
[  68.450836] omap-hw-nand: DMA timeout after 1148 ms, max. seen latency 1148 ms
[  70.974273] aic23 powering down
[  326.545074] Loading 3825.arm firmware

Namely the "aic23 powering up and down (I dont know what this is or if it is normal behavior)
and the DMA timeout.
Does anyone have any knowledge of these two things? Due to the nature of the device, logs are sparce. dmesg is the best I can think of to gather information in this case...

Nop for the powering up / powering down ...

fresh flash: only install xterm, dmesg | grep aic
Code:

[  26.793823] aic23 powering up
[  26.797851] aic23_init_power() done
[  29.591705] aic23 powering down
[  36.857269] aic23 powering up
[  36.861114] aic23_init_power() done
[  43.424804] aic23 powering down
[  49.971679] aic23 powering up
[  49.975769] aic23_init_power() done
[  52.365051] aic23 powering down
[  81.095062] aic23 powering up
[  81.099121] aic23_init_power() done
[  84.185424] aic23 powering down

But no DMA timeout ...

Do you know how to clear "no-lifeguard-reset flag" ?
=> go out from R&D: 770 reboot loop,
go back to R&D: still reboot
no-lifeguard-reset flag set: no reboot loop

=> I expect that to clear this flag: going out from R&D is enough ...
but I don't find an express was to do this.



Good day,
Laurent.

spycedtx 2006-07-20 18:21

aic23 or something else?
 
I've noticed the powerup/down messages a lot in dmesg, even before I noticed a problem with the video / audio. And, that makes me wonder something else.. perhaps the audio/video was broken before this update. Maybe installing Gizmo managed to break it.


I know this is the price we pay for playing on the edge, but I dread the thought of reflashing at this point, and having to reinstall all the apps, all over again.

I was cruising through dmesg -l output, and couldn't help but wonder if maybe something in the gstreamer* is what's been borked, even if -C doesn't report anything broken.

hallgreng 2006-07-20 19:08

Quote:

Originally Posted by Kralisec
Do you know how to clear "no-lifeguard-reset flag" ?
=> go out from R&D: 770 reboot loop,
go back to R&D: still reboot
no-lifeguard-reset flag set: no reboot loop

=> I expect that to clear this flag: going out from R&D is enough ...
but I don't find an express was to do this.

I believe the syntax is:
$./flasher-2.0 --clear-rd-flags=no-lifeguard-reset

I think that today I will spend some time updating my personal 'make dirs, symlinks, and file edits the way i like them after a fresh install'-script and then flash back the official 06 root dir.

Like BigBob said, this is really hurting my ability to use this device.
On the bright side, at least it is easy to go back to a working state! =]

Kralisec 2006-07-20 22:10

Quote:

Originally Posted by hallgreng
I believe the syntax is:
$./flasher-2.0 --clear-rd-flags=no-lifeguard-reset

Very Good ! Exactly what I was looking for ... Thank you !!!


I've play several times to know which package cause the problem,

After lots of test I see that:
Nokia 770 in normal mode, fresh flash, only xterm,
disable R&D, clear no-lifeguard-reset flag,

dpkg -i gst-plugins-farsight_0.10.1-0osso2maemo1_armel.deb goes to reboot loop ...

ok: I reinstall like before, go to 14 packages folders,
remove gst-plugins and do a "dpkg -i *.deb"
unexpected reboot, but looks stable after this,
do it another time "dpkg -i *.deb" to finish previous install, no reboot,
another dpkg -i *.deb" and ... reboot ! no logic for me ...

But after reboot: no loop mode, great I think ...

I go to gst-plugin to install it: dpkg request libstd :eek:
(do you remember ? not supposed to be installed ???)
ok ... but dpkg libstd request libgcc (supposely installed !)
and to finish dpkg -i libgcc request to install gcc-3.4 :confused:

After this 3 updates, I test to see result with reboot ... reboot loop mode come back :eek:
(remember: without gst-plugins)

ARGHHH ... it'enough for me ... I understood nothing :mad: :mad: :mad:
I stop playing, my 770 is ok me like that !





Bad day :mad:
Laurent.

shapr 2006-08-25 12:40

I have this same problem, would like a solution.
 
Two weeks ago I ssh'd into my 770 and did apt-get upgrade
After everything upgraded, I got a reboot loop.
I asked on #maemo and disable the lifeguard reboot

I didn't know that was the problem so I reflashed OS2006 today and did apt-get upgrade again...

The moral of this story is, apt-get on the command line is not supported by Nokia.

Sadly, I don't have time to reflash and reinstall everything again today.

hallgreng 2006-08-25 19:35

odd
 
I find this mostly baffling.
I understand how the application manager can do its own apt without killing stuff (just not showing these destructive packages and installing updates individually), but I dont understand WHY there are destructive packages in the stable repo!

BanditRider 2006-08-25 20:12

Just a guess here as I'm a Fedora guy. Maybe use update instead of upgrade.
In Fedora, yum update will check all installed apps for their latest stable version.
yum upgrade, will upgrade your system to the next version like Core 4 to Core 5
or Core 5 to Core 6. Probably not what you intended.

aleksandyr 2006-08-25 21:25

Because NOBODY is supposed to gain root permissions. Period. You need to use one a variety of hacks or exploits or R&D modes to do this, and it's simply assumed you know what you're doing at that point.

The primary problem is that attempting to upgrade a running application framework will fail, destroying the existing one, preventing any sort of bootup. In order to avoid this, it's necessary to use a serial console or apt-get over a non-running framework, such as updating one stored on an mmc chroot-ed in from one running on flash.

It's not NOT showing destructive packages, it's ONLY showing packages explicitly vetted for end-user use.

aflegg 2006-08-26 12:37

One of the packages with an upgrade in the repo (which is really designed for the SDK) is maemo-af-launcher. The lifeguard process on the device forces a reboot if this is killed.

It certainly needs a bit more sanity checking in its actions.

shapr 2006-08-26 14:35

Still, the problem is that even if you upgrade maemo-launcher via ssh with the lifeguard reboot turned off, once you disable developer mode and re-enable the lifeguard reset, the tablet is in a reboot loop again.

I googled to find the list of applications that were watchdogged by dsme, but I could not figure out what is prompting dsme to do the reboot.

I think I'll just give up on this and switch back to the OS2006 and install applications only via the appmanager.

Tobler 2006-12-04 13:40

SSH and RD-mode required to solve problem
 
I got my 770 rebooting always when I tried to install new programs. It's because dpkg tried to reconfigure maemo-launcher -program. If you try to install anything, package-management software will restart maemo-launcher service. And it will always reboot your 770 unless you are on R&D and no-lifeguard-reset -mode.

I had sshd running on my 770 (can't install it if your 770 is already on bad phase) beforehand. Then I changed mode:
Code:

$ ./flasher-2.0 --enable-rd-mode --set-rd-flags=no-lifeguard-reset
And when 770 was up and running, I connected to WLAN. Then I was able to ssh in and change to root. Run commands:
Code:

# dpkg --reconfigure -a
# apt-get upgrade

When apt-get was running, it told that it's stopping maemo-launcher. That caused 770 to reboot but not now because of no-lifeguard-reset -mode. But this stage I noticed that graphical UI disappeared on 770 screen. But it didn't matter because I had working ssh connection.
Finally apt-get started maemo-launcher again but graphical UI didn't show up. So I run reboot -command.
Finally it's time to disable RD mode:
Code:

$ ./flasher-2.0 --disable-rd-mode --clear-rd-flags=no-lifeguard-reset
So 770 goes up normally. And all updates and installations are working perfectly again.

BR, HenriP

debernardis 2008-08-17 14:50

Re: Reboot loop / apr-get upgrade / 20 Jun 2006 firmware
 
@Tobler: your post was precious - I was kept in a similar reboot loop with a borked apt - after updating a number of packages with apt-get. Entering R&D mode and disabling lifeguard allowed me to recover apt.


All times are GMT. The time now is 10:53.

vBulletin® Version 3.8.8