tinycore_2.5rc2

[roberts]

The Second Release Candidate of v2.5 (tinycore_2.5rc2.iso), is now posted and ready for testing.
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/2.x/release_candidates/

tinycore_2.5rc2.iso
tinycore_2.5rc2.iso.md5.txt

Change log for v2.5rc2:

* Updated tce-setup with single end call to depmod and ldconfg, and startup scripts.
* Updated tce-load with flag to skip depmod, ldconfig, startup script execution while booting.
* Updated tce-update now supports query mode and user selectable deployment.
* Updated wbar.sh for file test of .wbar

Notes:

The first two changes above allow for much improved boot times.

tce-update now allows one to "check for updates" by use of the query option.
It also support three levels of batch extension update deployment.

The safest is to perform updates by rebooting with "base norestore
text" (this is as it was).

The second way is to exit to system prompt and type sudo tce-update, then use the 'force' option. This allows use of extension required network which can be most helpful while still minimizing conflicts with currently running (in use) extensions.

The third way is to run it from within an X desktop by way of using the 'force' option. Warning, there is no test for currently running extensions and the possibility of system hangs or corruption. YMMV.

[meo]

Hi Robert!

My regards to the tinycore team! I've just tried out the last cut and I haven't anything to complain about. It's hard to tell if it boots faster but I take you for your word. Thanks a lot!

Have fun making tinycore progress,
meo

[philip]

Typically from release to release, it is the tinycore.gz which you will need to replace on your current install. Now you have direct access.

Great idea. Thank you.

[jpeters]

I got a 45% increase in boot speed (67 seconds down to 37 seconds).  Quite remarkable, Robert.   

[roberts]

It was an on-going topic of discussion by the team. Jason came up with the idea, I just implemented it and posted a private preview. The boot speed we noticed is consistent with your findings. Thanks to Jason for another great idea now implemented in Core.

[bmarkus]

Extended HD partition is listed in fstab and has an /mnt entry, see /mnt/hda4 below:

Code: [Select]

# /etc/fstab
proc            /proc        proc    defaults          0       0
sysfs           /sys         sysfs   defaults          0       0
devpts          /dev/pts     devpts  defaults          0       0
tmpfs           /dev/shm     tmpfs   defaults          0       0
/dev/hda        /mnt/hda        auto     noauto,users,exec    0 0 # Added by TC
/dev/hda1       /mnt/hda1       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda2       /mnt/hda2       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda3       none            swap     defaults             0 0 # Added by TC
/dev/hda4       /mnt/hda4       auto     noauto,users,exec    0 0 # Added by TC
/dev/hda5       /mnt/hda5       ext3     noauto,users,exec,relatime 0 0 # Added by TC
Doesn't harm, but definitely not needed and can be misleading.

/dev/hda like entry was already discussed in the summer; Curaga (if I remember well) had a comment on certain badly partitioned MP3 players which can't be mounted on a usual way as /dev/sda1 only as /dev/sda However it is rare exception, would be better to remove these entries too.

[jpeters]

Extended HD partition is listed in fstab and has an /mnt entry, see /mnt/hda4 below:

Code: [Select]

[font=Courier]
# /etc/fstab
proc            /proc        proc    defaults          0       0
sysfs           /sys         sysfs   defaults          0       0
devpts          /dev/pts     devpts  defaults          0       0
tmpfs           /dev/shm     tmpfs   defaults          0       0
/dev/hda        /mnt/hda        auto     noauto,users,exec    0 0 # Added by TC
/dev/hda1       /mnt/hda1       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda2       /mnt/hda2       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda3       none            swap     defaults             0 0 # Added by TC
/dev/hda4       /mnt/hda4       auto     noauto,users,exec    0 0 # Added by TC
/dev/hda5       /mnt/hda5       ext3     noauto,users,exec,relatime 0 0 # Added by TC[/font]
Doesn't harm, but definitely not needed and can be misleading.

/dev/hda like entry was already discussed in the summer; Curaga (if I remember well) had a comment on certain badly partitioned MP3 players which can't be mounted on a usual way as /dev/sda1 only as /dev/sda However it is rare exception, would be better to remove these entries too.

don't know if this is similar, but if I mount my thumb drive (sda1) fstab includes sda2 for the partition:

Code: [Select]


/dev/sda        /mnt/sda        auto     noauto,users,exec    0 0 # Added by TC
/dev/sda1       /mnt/sda1       vfat     noauto,users,exec,umask=000 0 0 # Added by TC
/dev/sda2       /mnt/sda2       vfat     noauto,users,exec,umask=000 0 0 # Added by TC


[jls]

I would change the info messages in tce-update from : tce-update to: sudo tce-update.
Infact it's written that is suggested to exit from X and run tce-update, but this is wrong since the correct command is sudo tce-update

[roberts]

Extended HD partition is listed in fstab and has an /mnt entry, see /mnt/hda4 below:

Code: [Select]

[font=Courier]
# /etc/fstab
proc            /proc        proc    defaults          0       0
sysfs           /sys         sysfs   defaults          0       0
devpts          /dev/pts     devpts  defaults          0       0
tmpfs           /dev/shm     tmpfs   defaults          0       0
/dev/hda        /mnt/hda        auto     noauto,users,exec    0 0 # Added by TC
/dev/hda1       /mnt/hda1       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda2       /mnt/hda2       ext3     noauto,users,exec,relatime 0 0 # Added by TC
/dev/hda3       none            swap     defaults             0 0 # Added by TC
/dev/hda4       /mnt/hda4       auto     noauto,users,exec    0 0 # Added by TC
/dev/hda5       /mnt/hda5       ext3     noauto,users,exec,relatime 0 0 # Added by TC[/font]
Doesn't harm, but definitely not needed and can be misleading.

/dev/hda like entry was already discussed in the summer; Curaga (if I remember well) had a comment on certain badly partitioned MP3 players which can't be mounted on a usual way as /dev/sda1 only as /dev/sda However it is rare exception, would be better to remove these entries too.

don't know if this is similar, but if I mount my thumb drive (sda1) fstab includes sda2 for the partition:

Code: [Select]


/dev/sda        /mnt/sda        auto     noauto,users,exec    0 0 # Added by TC
/dev/sda1       /mnt/sda1       vfat     noauto,users,exec,umask=000 0 0 # Added by TC
/dev/sda2       /mnt/sda2       vfat     noauto,users,exec,umask=000 0 0 # Added by TC


While there is still some debate whether to have too many (no harm) or not enough in certain (but rare) situations. I will try out an alternate build, but will do so for the next rc1 cycle, not  in our current rc cycle.

[roberts]

I would change the info messages in tce-update from : tce-update to: sudo tce-update.
Infact it's written that is suggested to exit from X and run tce-update, but this is wrong since the correct command is sudo tce-update

Done.

[bmarkus]

While there is still some debate whether to have too many (no harm) or not enough in certain (but rare) situations. I will try out an alternate build, but will do so for the next rc1 cycle, not  in our current rc cycle.

Thanks, that's fine. Problem with /dev/hda4 in the above example is that it is a misleading information, as it is just a place holder for real partions like /dev/hda5 which may be not so trivial for ordinary users not very familiar with LINUX and with certain applications (I do not have example).

[JoXo009]

Starting a terminal and typing 'tce-update' results in a scary warning text.

Why not offering a working solution instead of making people feel bad who can't use command line.

The query feature probably could be offered without problems without leaving the running windows manager.

And a GUI solution for the update feature maybe is possible by creating a one time flag, resulting in auto-update during next reboot. So that the user could be adviced, 'Do you want to set a flag, that next reboot an auto-update is executed?

After setting that flag the user just would reboot and the extension update would be executed automatically. The auto-update should end with an auto-reboot bringing back the interface and possibly showing some message as 'Just successfully/unsucsessfully updated X extensions'.

I think such solution wouldnt be too difficult to implement and would be a great improvement in usability for all these users who are uncomfortable with command line.


----
remark to user experience with interface Tool->Upgrade

Clicking the interface Tool->Upgrade never resulted in update and stopped working totally in version 2.5rc.

[roberts]

Appears that the jwm.tcz extension needs adjusting. You should post extension issues in the extension area under the jwm.tcz thread.

When testing a release candidate before you post, you should always verify your claim by booting base norestore. Doing such you will see the system menu is correct for the default window manager, thereby showing that it is an extension issue.

As to your suggestion, I have already answered it but will do so again:

I did consider an option as suggested, but I have concerns on space requirements, up to double the size of current tce directory and on a smallish device, pendrive, or CF, etc, could be problematic.

BTW, You can run 'query' within X. And it is really not too difficult to "Exit to prompt" and type tce-update.

[JoXo009]

When testing a release candidate before you post, you should always verify your claim by booting base norestore

Good to know, thanks for the hint.

concerns on space requirements, up to double the size of current tce directory

I think the update suggestion you already answered suggested to download updates with TC running into a second directory and then move the download during next reboot, what obviously means you need (double) space for the old and the new version of the extension.

That's not what I meant. My suggestion is exactly the same what a user would do. Rebooting with boot codes base norestore text and calling tce-update - My suggestion is to do it automatically by a script and to trigger the script by a flag (maybe the existence of the file 'update-requested) which can be set from running TC.

And it is really not too difficult to "Exit to prompt" and type tce-update.

True, for an expert, who knows how - after a typo - to get through without plugging off electricity and ruin all his code, it's really not difficult.

But for someone who, following such suggestions crashed his system several times resulting in countless hours of rebuilding work, it can be a nightmare.

[roberts]

base norestore text does not easily allow for wireless/dialup or other extension required net access.
That is why I had posted:

In 2.5rc2 tce-update takes into consideration wireless and/or other net access issues, by allowing updates to occur with user selectable options to minimize the number of running extensions.