Tiny Core 3.0 Alpha 6 Testing

[curaga]

To continue with my nit-picking today: I just stumbled over a potential contradiction to "... and libfontenc has not received an update since the 7.4 days. ..." (see post) insofar as according to a certain Xorg 7.5 page libfontenc-1.0.5 is the relevant version for that release.

Looking at some release dates on a log page I assume that what was build a year ago would most likely be libfontenc-1.0.4. But hey, I could be totally wrong with all my conclusions ...

That's correct, I based my assumption on the release date of 1.0.5. Though, looking at the changelog between 1.0.4 and .5 I don't see a need for update :

2009-08-28   libfontenc 1.0.5libfontenc-1.0.5   Peter Hutterer   1   -1/+1
2009-02-03   Add README with pointers to mailing list, bugzilla & git repos   Alan Coopersmith   1   -0/+25
2009-01-28   Janitor: Correct make distcheck and compiler warnings.   Paulo Cesar Pereira de Andrade   8   -29/+25
2007-12-06   Add missing PHONY line for automatic ChangeLog generation   James Cloos   1   -0/+2
2006-12-15   bump to 1.0.4libfontenc-1.0.4

[Juanito]

I read that some more system libraries were updated (e.g. 'libjepg'). I assume that base-dev.tcz will soon reflect those changes (if not already done).
...
Likewise will 'libss.a' remain in base-dev.tcz after the shared library was removed from Core

base-dev is current as of alpha5 - I need to remove libss.a, etc

[gutmensch]

sorry if I'm repeating an error (just in case), but searching for that one is quite hard and revealed nothing in the forum...

testing with 3.0alpha6:

tce-load -w -i mysql.tcz

/tmp/tcloop/mysql.tcz/... instead of /tmp/tcloop/mysql/... will be setup, then additionally the startup script from usr/local/tce.installed/mysql will not start (maybe because mysql.tcz is not mysql?) I'm experiencing the same problem while trying to make a new extension and I'm especially regarding the rights for tce.installed and its files... can anybody verify this or is my tc messed up?

best regards,
robert

[maro]

@gutmensch: I suspect it's your system.

I did a quick test with your command (i.e. tce-load -wi mysql.tcz) and the extension got mounted to the "usual place" with the "usual" directory name:

Code: [Select]

tc@box:~$ ls -l /tmp/tcloop/
total 0
drwxr-xr-x    3 root     root            26 Oct 27  2009 mysql/
drwxr-xr-x    3 root     root            26 Oct 16  2009 perl5/

tc@box:~$ mysql -V
mysql  Ver 14.14 Distrib 5.1.40, for pc-linux-gnu (i686) using  EditLine wrapper

tc@box:~$ ps | grep mysql | grep -v grep
  558 root     /bin/sh /usr/local/bin/mysqld_safe --ledir=/usr/local/libexec/ --socket=/tmp/mysql.sock --port=3306
                     --pid-file=/tmp/hostname.pid --datadir=/usr/local/var/mysql --language=/usr/local/share/mysql/english --user=tc
  657 tc       /usr/local/libexec//mysqld --basedir=/usr/local --datadir=/usr/local/var/mysql --user=tc
                    --language=/usr/local/share/mysql/english --log-error=/usr/local/var/mysql/box.err --pid-file=/tmp/hostname.pid
                    --socket=/tmp/mysql.sock --port=3306
It seem to be running as it probably should.

The one thing which I consider not so "pretty" is the fact the the startup script uses tar xvxf .... That produces a listing of all extracted files, which I personally could do without. I guess tar xzf ... would be "nicer" from an "aesthetic" POV.

[bmarkus]

While mysql.tcz installs fine here, it is a rather old extension and deserves and update both in version and TC compatibility. Will I will do it this week.

BTW, installing mysql.tcz Appbrowser displays:



See bottom status line. Is it a bug or feature?

[roberts]

Will likely be corrected when the v flag is removed from the mydsl.tcz mysql.tcz startup script tar extraction command. Startup scripts should not be sending output to the display.

EDIT: Fixed to mysql. Too many years there. Too many times typing that other system. 

[bmarkus]

Will likely be corrected when the v flag is removed from the mydsl.tcz startup script tar extraction command. Startup scripts should not be sending output to the display.

Thanks. Anyhow I will remove tar from the start-up script.

[maro]

Gee, I wonder why Robert has typed "... the mydsl.tcz startup  ..." 

[Arslan S.]

i can confirm what gutmensch says

some of the extensions are mounted with .tcz suffix after reboot

add compiletc to onboot list and reboot

tc 3.0a6

[roberts]

I just did and did not get a failure. compiletc mounted normal.
Please post the output of showbootcodes and any other info regarding your setup particulars.

[Arslan S.]

boot codes: quiet tce=sda9

i have setup tc on hard drive with backup/restore

http://farm5.static.flickr.com/4017/4681558220_e42a26d771_b.jpg

[^thehatsrule^]

From a quick glance, it seems that the problem starts after coreutils is installed

[gutmensch]

I think I could track down one thing...

coreutils.tcz is causing the problem on my system. every extension loaded before coreutils is ok, every extension loaded after coreutils in onboot.lst gets mounted with suffix. Strangely on tc3.0a3 the extension after coreutils (in my case bash.tcz) doesn't load at all although it's in onboot.lst and in the repo. On tc3.0a6 bash.tcz loads but with suffix.

bootcodes: kmap=de-latin1 tce=hda1 home=hda1 restore=hda1 opt=hda1 lang=de_DE quiet

@hats: wow that quick shot came unexpectedly from out of nowhere ;-)

[roberts]

Found it. Fixed it.
Nothing wrong with coreutils, except that its utils are so much more bloated than what I need during boot.

[SvOlli]

I bit this isn't a new one, but I wanted to note it anyway:

If you boot with the "user=xxx" option, the directories "/home/xxx/.OnDemand" and "/home/xxx/.wmx" will still be owned by "tc" instead of "xxx".