Tiny Core Linux
Tiny Core Base => TCB News => Release Candidate Testing => Topic started by: roberts on November 17, 2009, 01:25:14 PM
-
The Second Release Candidate of v2.6 (tinycore_2.6rc2.iso), is now posted and ready for testing only.
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/2.x/release_candidates
tinycore_2.6rc2.iso
tinycore_2.6rc2.iso.md5.txt
Cumulative change log for v2.6rc2:
* Updated bzImage for smaller size (upx'ed)
* Updated rebuildfstab to drop extra mount points.
* New select utility replaces dialog
* Dropped dialog, use dialog.tcz extension (posted)
* Updated xsetup.sh
* Updated mktclocal
* Updated tc-terminal-server
* Updated cpanel to add Apps Audit
* Updated missing link libpng.so
* Updated blkid, libblkid, and libuuid
* Dropped cryptohome support.
* Updated tc-config
* Updated rc.shutdown
* Busybox mount and losetup replaces GNU versions
* Updated tce-load to support busybox mount.
* Dropped busybox ed, eject, fsck, expand, and unexpand applets.
* Busybox lsmod, insmod, rmmod, modprobe replace GNU versions.
* Fixed bug in flwm_topside_restart
* Updated flwm menu .wmx to reflect recent changes.
Notes:
You must have your tcz collection updated, i.e., all squashfs and all with 4096 block size.
The repository has been updated.
Be sure to update your kernel, with new smaller, boots faster, bzImage.
The distribution files are also availble:
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/2.x/release_candidates/distribution_files
Please check your md5sum.
EDIT: Added emphasis to the tcz collection requirement.
-
Isn't it a contradiction?
* Busybox mount and losetup replaces GNU versions
* Updated tce-load to support busybox mount.
-
We are using busybox mount applet. tce-load was adjusted to use full specificatons, thus clearing up the dmesg items that appeared without such.
-
Everything working, but no mercy on the extensions needing to be correct. I had to rebuild a few of my own old ( probably cramfs ?) extensions. If an extension fails, the rest stop loading at that point.
-
Works well for me. Thank you!
I appreciate the distribution files option.
The cautionary remarks from jpeters were useful to me: before leaving version 2.5 behind, I used his update script to get the latest copies of all extensions, then manually replaced all the X*core*tcz* files just in case. The result was a clean and fast first boot and happy computing since then. All this on the Acer Aspire One netbook (first edition, model 110-1955 model ZG5).
Here's one cosmetic suggestion for the download page at
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/ :
Shrink the top link text down to simply "Current Releases". Reasons:
1. The Distribution Files are in a subdirectory, where the name is totally appropriate.
2. The proposed short sweet title is (IMHO) more welcoming to newcomers.
3. It makes a nice parallel counterpart to the next item, "Release Candidates".
Of course it's Robert's decision.
-
...
Here's one cosmetic suggestion for the download page at
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/ :
Shrink the top link text down to simply "Current Releases". Reasons:
1. The Distribution Files are in a subdirectory, where the name is totally appropriate.
2. The proposed short sweet title is (IMHO) more welcoming to newcomers.
3. It makes a nice parallel counterpart to the next item, "Release Candidates".
Of course it's Robert's decision.
Website updated.
-
2.6rc2 is randomly freezing as 2.6rc1 did. It happens during startup, during normal use, installing applications, etc.
-
It is freezing booting from a different USB and from CD too on the Acer Aspire 5720Z notebook which works fine with 2.5 and prior releases.
Sticks work fine on a different desktop machine.
-
Are you saying that it only occurs with that one notebook?
What happens if you use the kernel from v2.5?
-
Are you saying that it only occurs with that one notebook?
Yes
What happens if you use the kernel from v2.5?
Will try.
-
What happens if you use the kernel from v2.5?
2.6rc2 works fine with 2.5 kernel. What is the difference between kernels? uname reports the same complation date but their size are different and both named 2.6.29.1-tinycore ?
EDIT:
Hm... With 2.5 kernel it starts always as expected. Just now it got frozen just leaving it running after boot with LXDE but no other applications running. Except this single case it is usable and works with no problem.
-
To me it sounds like you may have some bad RAM, that is often cited as a cause of system freezing. If you have not already, it might pay to run memtest on that machine just to be safe.
The kernel image has been upx'd in this RC.
-
To me it sounds like you may have some bad RAM, that is often cited as a cause of system freezing. If you have not already, it might pay to run memtest on that machine just to be safe.
It is possible. Till now there were no any indication. Will check.
-
I just ran a test with a full set of 4096-ized extensions and it looks good.
If an extension fails, the rest stop loading at that point.
Yes. This is annoying because the tools to fix the problem may not be loaded yet.
-
You should update your extensions before you update the system files.
Perhaps you missed:
Notes:
You must have your tcz collection updated, i.e., all squashfs and all with 4096 block size.
The repository has been updated.
-
Booting straight from the .iso, in VirtualBox, same configuration as with 2.5, and I still get the "ACPI: Unable to load the System Description Tables" message. System comes right up, though, and seems to be fully functional, unlike RC1, that took nearly 5 minutes just to get past the ACPI message.
Is there any additional information you would like about the configuration, dmesg, etc that would be of assistance? 2.6 is the first release (RC or otherwise) to produce an error on this box, I'd like to help fix that if I can.
-
Not sure if this is from fltk or flwm, but whenever a new window opens with appbrowser my mouse cursor and window focus is moved to that new window. I've noticed this before in previous versions. This also happens with the control panel, but not wbar, so probably with fltk?
This is more of an annoyance then a bug, especially when loading extensions with many deps such as compiletc from appbrowser. Would it be possible for the new windows to not take the user's focus so the user can do other work while waiting?
-
Here is a bug with appbrowser. Instead of saying an extension has already been downloaded it displays a wget error:
Error: wget: server returned error: HTTP/1.1 416 Requested Range Not Satisfiable on sed.tcz
How to reproduce:
1. "Download Only" sed.tcz through appbrowser
2. Try to "Download Only" sed.tcz again
-
To me it sounds like you may have some bad RAM, that is often cited as a cause of system freezing. If you have not already, it might pay to run memtest on that machine just to be safe.
The kernel image has been upx'd in this RC.
After 8 hours running Memtest-86 v3.5 (Pass: 10) no error found. It will run for longer time of course....
-
CPU overheating is also a common reason for computer lockup.
This is especially a problem with laptops that have small fans that
may not be properly managed by the kernel.
Check your BIOS to see if you can control the fan, or do a overtemp shutdown.
-
CPU overheating is also a common reason for computer lockup.
This is especially a problem with laptops that have small fans that
may not be properly managed by the kernel.
Check your BIOS to see if you can control the fan, or do a overtemp shutdown.
What? Why all other OS's including TC 2.5 and 2.4, 2.3, .... are running fine for a week or more time continously on max CPU speed and below 55 C except MANDRIVA 2010 and TC 2.6rcx? Definitely it is not a heat problem, as freeze happens immediately where CPU and all the system is cold. If you run 2.5, 2.6, 2.5, 2.6, ... one after the other 2.5 runs while 2.6 freeze. Also, when MANDRIVA 2010 heated the system I recognized it by fingers touching the housing and not by the errors running it.
-
CPU overheating is also a common reason for computer lockup.
This is especially a problem with laptops that have small fans that
may not be properly managed by the kernel.
Check your BIOS to see if you can control the fan, or do a overtemp shutdown.
What? Why all other OS's including TC 2.5 and 2.4, 2.3, .... are running fine for a week or more time continously on max CPU speed and below 55 C except MANDRIVA 2010 and TC 2.6rcx? Definitely it is not a heat problem, as freeze happens immediately where CPU and all the system is cold. If you run 2.5, 2.6, 2.5, 2.6, ... one after the other 2.5 runs while 2.6 freeze. Also, when MANDRIVA 2010 heated the system I recognized it by fingers touching the housing and not by the errors running it.
Most freezes as you well know better than I occur due to code being corrupted as a result of self modification (stack overflows) or improper execution (memory leaks). I doubt you will find a memory hardware failure. More likely will be a driver that is acting up. Perhaps you could disable hardware peripherals in the bios as a test.
-
Tinycore 2.6rc2 Base norestore boot code gives error ACPI: Unable to load the System Description Tables". USB doesn't seem to be recognized at any time under 2.6rc2. Touch pad works after xsetup. USB mouse never got working. My laptop seems choppy and sluggish. With my normal boot code the extensions don't load. They should be the 4096 type. I updated them last night.
Running tinycore2.5 presents no problem. I have a MSI U100 with compiled wireless driver.
-
Has anyone that is having issues tried using 2.5 with the 2.6 kernel? That would help narrow down if it is the kernel image itself as we have tried the opposite.
-
Has anyone that is having issues tried using 2.5 with the 2.6 kernel? That would help narrow down if it is the kernel image itself as we have tried the opposite.
I tested the 2.6rc2 bzImage with 2.5. I still get AE_NO_ACPI_ unable to load the System Description tables error. Also a PNP bios dev node error while booting.
Extensions in tce seem to load. I can.t get wireless to work.( rtl8187se works with 2.5)
USB mouse moves very erratically. Touch pad works.
cpufreq gives cpu unknown or wrong cpu. Conky shows 0 degrees for cpu.
-
Thanks. If 2.5 and it's kernel pose no problem, then it seems like the kernel image.
If you haven't already, testing the 2.5 kernel on the 2.6 RC like bmarkus did would help nail down the culprit.
-
Just curious, what boot loader and which version are being used for those experiencing issues with 2.6?
Does updating your boot loader help?
-
Just curious, what boot loader and which version are being used for those experiencing issues with 2.6?
Does updating your boot loader help?
I'm just booting straight off the plain .iso. No other bootloaders, no hard drive, and 512MB ram.
-
To me it sounds like you may have some bad RAM, that is often cited as a cause of system freezing. If you have not already, it might pay to run memtest on that machine just to be safe.
The kernel image has been upx'd in this RC.
After 8 hours running Memtest-86 v3.5 (Pass: 10) no error found. It will run for longer time of course....
After running it for 18 hours I stopped memtest. No errors found.
-
Has anyone that is having issues tried using 2.5 with the 2.6 kernel? That would help narrow down if it is the kernel image itself as we have tried the opposite.
2.5 with 2.6 kernel is freezing too. So it is up to the new kernel.
Updated BIOS, no change.
-
Just curious, what boot loader and which version are being used for those experiencing issues with 2.6?
Does updating your boot loader help?
USB install is done with UNETBOOTIN but behaviour is the same runnin the system from CD.
-
OK. Now we are getting somewhere.
I will try updating isolinux to see if that make the CD boot.
Can you try an update to unetbootin?
-
OK. Now we are getting somewhere.
I will try updating isolinux to see if that make the CD boot.
Can you try an update to unetbootin?
Using latest unetbootin-windows-377.exe result is the same.
-
if issue is isolinux, was it also necessary to revert squashfs ?
-
OK. Now we are getting somewhere.
I will try updating isolinux to see if that make the CD boot.
Can you try an update to unetbootin?
Using latest unetbootin-windows-377.exe result is the same.
So unetbootin does not support upx'ed kernels. However, you should be able to use Tiny Core's usbinstall to get a working pendrive that does,
-
if issue is isolinux, was it also necessary to revert squashfs ?
I don't understand the question, unless it is on why we decided to standardize on squashfs/4096? The answer to that is, to try to simplify.
-
OK. Now we are getting somewhere.
I will try updating isolinux to see if that make the CD boot.
Can you try an update to unetbootin?
Using latest unetbootin-windows-377.exe result is the same.
So unetbootin does not support upx'ed kernels. However, you should be able to use Tiny Core's usbinstall to get a working pendrive that does,
Yes, this is what I wanted to try also this night.
-
Now that dialog has been removed, XVesa Setup is awkward to use. Perhaps the list of resolutions could be laid out in a table that would be easier to read. (And white on black is fine with me.)
-
So unetbootin does not support upx'ed kernels. However, you should be able to use Tiny Core's usbinstall to get a working pendrive that does,
Result with stick created with usbinstall on a desktop machines:
Type h - works fine on desktop PC, 'Boot error' on Aspire notebook
Type e - works fine on desktop PC, boots on Aspire but freeze randomly
Type z - 'Boot error' on Aspire known from past experience, not tested now
This indicates that issue is not UNETBOOTIN specific.
-
I am booting with grub4dos. I also booted directly from a USB using the netbook bios. Same problems with Unable to load the System Description Tables error.
-
OK. Upx'ed kernel will be pulled in next RC. Thanks to all who helped test.
-
Now that dialog has been removed, XVesa Setup is awkward to use. Perhaps the list of resolutions could be laid out in a table that would be easier to read. (And white on black is fine with me.)
I have added paging capability to select.
-
Here is a bug with appbrowser. Instead of saying an extension has already been downloaded it displays a wget error:Error: wget: server returned error: HTTP/1.1 416 Requested Range Not Satisfiable on sed.tcz
How to reproduce:
1. "Download Only" sed.tcz through appbrowser
2. Try to "Download Only" sed.tcz again
I had generic messages but got complaints. Now I only pass on the error message from the utilty, in this case it could be busybox wget or GNU wget depending on extensions loaded.