Page 35 of 55

Re: Rooting Toon (or boxx)

Posted: Wed Jul 25, 2018 10:28 am
by Toonz
mAiden wrote:Other error i got..
Inittab is fixed, thanks guys!

Code: Select all

INIT: Id "qtqt" respawning too fast: disabled for 5 minutes
Looks like the install wasn't completed. You could try to run the update script from TheHogNL again, forcing a reinstall of version 4.18.8.
New resource file is available so should work...

Re: Rooting Toon (or boxx)

Posted: Wed Jul 25, 2018 10:41 am
by mAiden
Toonz wrote:
mAiden wrote:Other error i got..
Inittab is fixed, thanks guys!

Code: Select all

INIT: Id "qtqt" respawning too fast: disabled for 5 minutes
Looks like the install wasn't completed. You could try to run the update script from TheHogNL again, forcing a reinstall of version 4.18.8.
New resource file is available so should work...
Hi Toonz,

This is what i got, when i try it:

Code: Select all

2018-07-25 08:38:42 (1.21 MB/s) - `update-rooted.sh' saved [19684/19684]

eneco-001-xxxxxx:~# sh update-rooted.sh -f 4.18.8
===================================================================================================================================================================
Welcome to the rooted Toon upgrade script. This script will try to upgrade your Toon using your original connection with Eneco. It will start the VPN if necessary.
Please be advised that running this script is at your own risk!

Version: 2.94  - ThehogNL - 19-7-2018

===================================================================================================================================================================

Only fixing files.
FIXING: Downloading resources.rcc TSC mod for this version 4.18.8.
FIXING: Now installing latest toonstore app. This fixes some files also.
Downloading http://files.domoticaforum.eu/uploads/Toon/apps//toonstore-1.1.4/toonstore_1.1.4-r0_qb2.ipk.
Package toonstore (1.1.5-r0) installed in root is up to date.
FIXING: Now installing latest busybox mod. This is necessary to enable console output again which is disabled in 4.10 by Eneco.
sh: 4: unknown operand
sh: 4: unknown operand
Custom busybox install not necessary for this firmware.
eneco-001-xxxxxx:~#
Other trys:

Code: Select all

eneco-001-xxxx:~# ./update-rooted.sh -f 4.18.8
-sh: ./update-rooted.sh: Permission denied
eneco-001-xxxx:~# chmod -x update-rooted.sh
eneco-001-xxxx:~# ./update-rooted.sh -f 4.18.8
-sh: ./update-rooted.sh: Permission denied
eneco-001-xxxxx:~#


Re: Rooting Toon (or boxx)

Posted: Wed Jul 25, 2018 11:23 am
by TheHogNL
First read what the options are doing: https://github.com/IgorYbema/update-rooted

You are trying -f with a version number. That is invalid. -f is for fixing after an install and doesn't need a version number. You probably mean -v 4.18.8 (force version 4.18.8).

And the permissions denied is because you are now running it as "./update-script.sh" instead of "sh update-script.sh" (and you did make the file executable).

And an empty /etc/inittab .. how did you manage to do that?? It looks you are doing stuff on linux without understanding what you are doing :-)

Re: Rooting Toon (or boxx)

Posted: Wed Jul 25, 2018 11:32 am
by TerrorSource
TheHogNL wrote:First read what the options are doing: https://github.com/IgorYbema/update-rooted

You are trying -f with a version number. That is invalid. -f is for fixing after an install and doesn't need a version number. You probably mean -v 4.18.8 (force version 4.18.8).

And the permissions denied is because you are now running it as "./update-script.sh" instead of "sh update-script.sh" (and you did make the file executable).

And an empty /etc/inittab .. how did you manage to do that?? It looks you are doing stuff on linux without understanding what you are doing :-)
Empty inittab can happen when an upgrade was unsuccessful and you did not check it before rebooting :)

@mAiden, maybe a better idea to improve you're linux skills a bit more before you do anything else on the Toon?

Re: Rooting Toon (or boxx)

Posted: Wed Jul 25, 2018 12:53 pm
by mAiden
I think my Linux skills are ok, with certificate for it.
I got a LPI certificate for it.

Inittab goes empty indeed due to a failed update.

I have fixed already, Toon start up normally. Thank you for the support.

Screenshot Toon:

http://prntscr.com/kasgt1

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 3:30 am
by Ierlandfan
He posted, got support, he solved it and in the end, that's the whole meaning of this board.
Thanx for helping him!

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 9:37 am
by Toonz
mAiden wrote:I have fixed already, Toon start up normally. Thank you for the support.
Good to hear you're up and running again :) .
As we have heard about a few similar cases before I would like to understand why these updates fail (in this case a genuine firmware update).
Not enough disk space before the update starts? Does the fact that a Toon is rooted is relevant?
Has anyone been able to look into the root cause?

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 9:38 am
by TheHogNL
mAiden wrote: Inittab goes empty indeed due to a failed update.
Update doesn't do anything with inittab :-)

But nice that it is working now!

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 11:19 am
by Toonz
Question remains why the inittab file can get empty in some cases....

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 12:27 pm
by aabre
Toonz wrote:Question remains why the inittab file can get empty in some cases....
maybe mAiden is not very familiar with the vi editor ?

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 12:40 pm
by TerrorSource
aabre wrote:
Toonz wrote:Question remains why the inittab file can get empty in some cases....
maybe mAiden is not very familiar with the vi editor ?
Maybe true but he isn't the first one with this issue after updating. already seen 3-4 persons before who had exactly the same issue with inittab.

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 3:17 pm
by TheHogNL
TerrorSource wrote:
aabre wrote:
Toonz wrote:Question remains why the inittab file can get empty in some cases....
maybe mAiden is not very familiar with the vi editor ?
Maybe true but he isn't the first one with this issue after updating. already seen 3-4 persons before who had exactly the same issue with inittab.
It is possible that after (manually) editing the inittab file (to disable openvpn) you forget to 'sync' to the disk and reboot to soon. This will result in a corrupted or empty file.

However this is not part of the upgrade (the upgrade never touches the inittab) and only part of the manually rooting process.

Re: Rooting Toon (or boxx)

Posted: Thu Jul 26, 2018 4:22 pm
by mAiden
aabre wrote:
Toonz wrote:Question remains why the inittab file can get empty in some cases....
maybe mAiden is not very familiar with the vi editor ?
I do not like vi, indeed. Use much more often and prefer nano.
That is no excuse. inittab must be empty for a reason. Will view the log files on Saturday or Sunday. Maybe I find something in it.

Re: Rooting Toon (or boxx)

Posted: Sat Jul 28, 2018 2:37 pm
by mAiden
Find this in the update log.

Code: Select all

HCBv2 startables changed, writing /etc/inittab.
>> Done, scheduling reboot in 2 secs...
>> 2018-07-23 17:35:37 +0200 (pid 1453) executed
Location: /mnt/data/log/last-update.log

Is almost certain that this has destroyed my inittab.

Re: Rooting Toon (or boxx)

Posted: Sat Jul 28, 2018 9:20 pm
by TheHogNL
Impossible. This is part of the HCBv2 startup script which is called by the eneco update script just before the script is finished (this is what you see in the log).

The part of the HCBv2 script responsible:

Code: Select all

                                printline "startables changed, writing $INITTAB_FILE."
                                echo "$INITTAB_CONTENT_NEW" > $INITTAB_FILE.new

                                # file was created ok? (non-zero size)
                                if [ -s $INITTAB_FILE.new ]
                                then
                                        # force-rename removes old file, still more atomic than direct write
                                        mv -f $INITTAB_FILE.new $INITTAB_FILE

                                        # so repeated calls to it_writeifchanged() work:
                                        INITTAB_CONTENT_ONDISK="$INITTAB_CONTENT_NEW"
                                else
                                        printline "ERROR: Writing $INITTAB_FILE failed! Disk full/read-only?"
                                fi

This has a zero-file-size check in it. If the new created inittab.new file is zero, it will not overwrite the real inittab.

There must be another reason for this to happen.