[Announce] CSSU Testing thread - Page 175 - maemo.org

merlin1991 | # 1741 | 2015-01-01, 22:54 | Report

Originally Posted by sLumPia View Post

Where can I get the nokiamaps-navigation-provider_0.200-2+0m5_armel.deb file?

I is right there in the repo http://repository.maemo.org/communit...tion-provider/ maybe try running apt-get update like it suggests.

Also I just released 21.2011.38-1Tmaemo10.1
There were two issues with the last release, the hildon-desktop dependency wasn't right and also fmtx-middleware was in the changelog but not actually in the repo.

Changelog for this version in the ususal place:
http://wiki.maemo.org/Community_SSU/...log#Tmaemo10.1

The Following 6 Users Say Thank You to merlin1991 For This Useful Post:
Amboss, Estel, klinglerware, panjgoori, Tiran, Wikiwide

pichlo | # 1742 | 2015-01-01, 22:58 | Report

While we are talking about hildon-desktop, would it be too much to ask to remove the User Guide desktop icon? It is only a dead web link anyway...

The Following 6 Users Say Thank You to pichlo For This Useful Post:
Amboss, Estel, marmistrz, rotoflex, sLumPia, Wikiwide

crail | # 1743 | 2015-01-02, 02:01 | Report

Originally Posted by Estel View Post

Sure, it will be done when it will be done, without any doubts

Seriously though, are you waiting for some particularly important hotfix?

not particularly important but can you fix coloured text in modest please?

The Following User Says Thank You to crail For This Useful Post:
Wikiwide

sLumPia | # 1744 | 2015-01-02, 02:15 | Report

Originally Posted by merlin1991 View Post

I is right there in the repo http://repository.maemo.org/communit...tion-provider/ maybe try running apt-get update like it suggests.
Also I just released 21.2011.38-1Tmaemo10.1
There were two issues with the last release, the hildon-desktop dependency wasn't right and also fmtx-middleware was in the changelog but not actually in the repo.
Changelog for this version in the ususal place:
http://wiki.maemo.org/Community_SSU/...log#Tmaemo10.1

I have try a few of "apt-get update" with the same result
But it fixed after I run "apt-get update --fix-missing" this morning.
Thank you

The Following 2 Users Say Thank You to sLumPia For This Useful Post:
Amboss, Wikiwide

pali | # 1745 | 2015-01-02, 08:10 | Report

Originally Posted by peterleinchen View Post

pali, but I do have gnu-coreutils and other gnu packages installed? Okay together with busybox-power.
Could you elaborate that a bit more, please?

For sure I would never ever do a dist(/deselect)-upgrade...

Do you have package coreutils or coreutils-gnu?

http://maemo.org/packages/view/coreutils/
http://maemo.org/packages/view/coreutils-gnu/

These are different packages and dist-upgrade needs first one.

The Following User Says Thank You to pali For This Useful Post:
Wikiwide

peterleinchen | # 1746 | 2015-01-02, 08:35 | Report

Originally Posted by pali View Post

Do you have package coreutils or coreutils-gnu?
These are different packages and dist-upgrade needs first one.

Ah, I see.
I have coreutils-gnu. So that explains!?

But when I do a dist-upgrade -s, I am not asked/informed about coreutils package:

Originally Posted by

~# apt-get upgrade -s
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following packages have been kept back:
hildon-initscripts libcurl3 libgcc1 libsdl-mixer1.2
libstdc++6 libxau6
0 upgraded, 0 newly installed, 0 to remove and 6 not upgraded.
delite:~# apt-get dist-upgrade -s
Reading package lists... Done
Building dependency tree
Reading state information... Done
Calculating upgrade... Done
The following packages will be REMOVED:
cell-modem-ui diffutils-gnu git grep-gnu
The following NEW packages will be installed:
diffmo gcc-4.6-base libkeyutils1 libkrb53 libvorbisfile3
The following packages have been kept back:
hildon-initscripts libcurl3 libgcc1 libsdl-mixer1.2
libstdc++6 libxau6
WARNING: The following essential packages will be removed.
This should NOT be done unless you know exactly what you are doing!
grep-gnu
0 upgraded, 5 newly installed, 4 to remove and 6 not upgraded.

--
but before we go into total OT, keep in mind this device is running on 'stock' PR1.3...

The Following 3 Users Say Thank You to peterleinchen For This Useful Post:
Amboss, handaxe, Wikiwide

Mara | # 1747 | 2015-01-02, 16:41 | Report

Originally Posted by Mara View Post

I have been looking around the config files etc. to see if I can pinpoint where the issue is. So far no success.

In the etc/fstab I have /dev/mmcblk0p1 /home/user/MyDocs vfat defaults 0 2

The fstab is generated by the script etc/event.d/rcS-late. I looked into this script and didn't notice anything unusual there.

The config file for user's filesystems seems to be the etc/default/mount-opts. I have there lines for "home-opts=" which I don't think applies for MyDocs case. The second line says: fat-opts="noauto,nodev,noexec,nosuid,noatime,nodiratim e,utf8,uid=29999,shortname=mixed, dmask=000,fmask=0133,rodir". I did try changing the fmask to 000 and rodir to rwdir, after which I mounted MyDocs, but it did not help. (Then I changed them back to original.)

There is another config file /etc/default/mount-opts.dpkg-dist, but there I have pretty much the same options except the fmask=0000, which should work fine?

The file etc/profile file I have line "umask 022" but when I changed it to "umask 000" it did not fix the problem. (Then I changed it back.)

The simplest workaround at this point is to umount /dev/mmcblk0p1 and then remount it using this command:
mount /dev/mmcblk0p1 -o dmask=000,fmask=000

In other words the mount command finds the all the correct default options from somewhere. It only needs manual overrides for dmask and fmask values.

Maybe this gives enough details for someone with detailed file system experience to pinpoint where the issue might be?

EDIT: Before someone points out the space before the last "e" in "nodiratime", it seems to be created by this forum: In post edit mode the space doesn't exists.

Little bit more to this:

I edited the /etc/event.d/rcS-late script line 100:
original: [ -z "$mydocs_opts" ] && mydocs_opts=defaults
modified: [ -z "$mydocs_opts" ] && mydocs_opts="defaults,dmask=000,fmask=000"

After this the fstab file does show:
/dev/mmcblk0p1 /home/user/MyDocs vfat defaults,dmask=000,fmask=000 0 2

After fresh boot MyDocs is not mounted using these parameters... The etc/mtab shows:
/dev/mmcblk0p1 /home/user/MyDocs vfat rw 0 0
so it is read-only.

Once I manually umount and remount it back:
mount /dev/mmcblk0p1

The /etc/mtab shows:
/dev/mmcblk0p1 /home/user/MyDocs vfat defaults,dmask=000,fmask=000 0 0
and I do have read/write access.

So to summarize the very first time the MyDocs is mounted after bootup does not use the fstab information. (Or the rc-S-late did not yet run by that time?) Only later phases the mount command does use fstab for its parameters.

I also noticed in Conky that there is fsck.vfat process that runs a while after bootup. MyDocs gets mounted after the fsck had finished but doesn't seem to use the fstab?

I guess the question now is that if anyone knows where is the script responsible to run the fsck.vfat after boot? I would like to look into that and see how the MyDocs gets first mounted after device bootup.

The Following 4 Users Say Thank You to Mara For This Useful Post:
Amboss, Estel, rotoflex, Wikiwide

rotoflex | # 1748 | 2015-01-03, 01:25 | Report

I can no longer access my SD card or anything other than the root directory. File Manager just shows Bluetooth as an available system (& no other file systems), & Filebox simply crashes when it tries to start.

Connection via USB asks for the usual Mass Storage or PC Mode, but after choosing Mass Storage both my Linux & W7 computers do not show the N900 file system or the SD card as availabilities. Then, when the USB is disconnected, the "Disconnect charger to save power" message from charger-only connection comes up. Subsequent reconnections via USB to a Linux or W7 computer then only show the "Charging" message, & not the "Mass Storage/PC Mode" selection screen until a power down & re-start - then the sequence repeats.

These updates have been installed since last access via FileBox or USB:
2nd CSSU-testing update
Modrana update
Enhanced Busybox update

I think the CSSU-testing update would be most likely to cause this problem.

Last edited by rotoflex; 2015-01-03 at 01:29. Reason: Corrected filebox/file manager misidentification

The Following 4 Users Say Thank You to rotoflex For This Useful Post:
Amboss, burwat292596, Estel, Wikiwide

Wikiwide | # 1749 | 2015-01-03, 02:14 | Report

Originally Posted by rotoflex View Post

I can no longer access my SD card or anything other than the root directory. File Manager just shows Bluetooth as an available system (& no other file systems), & Filebox simply crashes when it tries to start.

Connection via USB asks for the usual Mass Storage or PC Mode, but after choosing Mass Storage both my Linux & W7 computers do not show the N900 file system or the SD card as availabilities. Then, when the USB is disconnected, the "Disconnect charger to save power" message from charger-only connection comes up. Subsequent reconnections via USB to a Linux or W7 computer then only show the "Charging" message, & not the "Mass Storage/PC Mode" selection screen until a power down & re-start - then the sequence repeats.

These updates have been installed since last access via FileBox or USB:
2nd CSSU-testing update
Modrana update
Enhanced Busybox update

I think the CSSU-testing update would be most likely to cause this problem.

MMC inside N900 gets mounted by /usr/sbin/mmc-mount or something like this.

Don't be surprised that it starts fsck.vfat -a /dev/mmcblk0p1

However, if your memory is seriously wrong, it will not get repaired by this.

Try to run, as root:

umount /dev/mmcblk0p1
fsck.vfat -avl /dev/mmcblk0p1

To see which files are wrong.

If you want to repair it, run fsck to completition. You may need to keep charging N900, to prevent shutdown-during fsck.

Since for me shutdown-during-fsck was inevitable, I used switch -w to cure the filesystem.

After the filesystem gets repaired, reboot, and it should auto-mount. It may take several minutes, though, to assure itself that the filesystem is fine.

Best wishes.

The Following 4 Users Say Thank You to Wikiwide For This Useful Post:
Amboss, burwat292596, Estel, pichlo

burwat292596 | # 1750 | 2015-01-03, 11:42 | Report

My symtoms are somewhat the same, so i set off fsck over two hours ago, and it is now renaming files in it's third thousand. I do not know how long it should take given 26gb of music, golf and snooker. The only thing that worries me is that the memory view in file manager showed 1.3gb free and that looked like a third of the available memory on the little blue bar when should show approx 1.3 against 26.8 total. Am I in more trouble than I think?

Originally Posted by Wikiwide View Post

MMC inside N900 gets mounted by /usr/sbin/mmc-mount or something like this.

Don't be surprised that it starts fsck.vfat -a /dev/mmcblk0p1

However, if your memory is seriously wrong, it will not get repaired by this.

Try to run, as root:

umount /dev/mmcblk0p1
fsck.vfat -avl /dev/mmcblk0p1

To see which files are wrong.

If you want to repair it, run fsck to completition. You may need to keep charging N900, to prevent shutdown-during fsck.

Since for me shutdown-during-fsck was inevitable, I used switch -w to cure the filesystem.

After the filesystem gets repaired, reboot, and it should auto-mount. It may take several minutes, though, to assure itself that the filesystem is fine.

Best wishes.

The Following 3 Users Say Thank You to burwat292596 For This Useful Post:
Amboss, Estel, Wikiwide