Tiny Core Linux

Tiny Core Base => TCB News => Release Candidate Testing => Topic started by: roberts on August 02, 2012, 09:56:45 PM

Title: Core v4.6rc1
Post by: roberts on August 02, 2012, 09:56:45 PM
The first release candidate of Core 4.6 is ready for public testing:
http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/4.x/x86/release_candidates

Change Log:
* New mirrorpicker GUI - Select fastest mirror.
* Updated Apps GUI to add first run and menu option for mirrorpicker GUI.
* Updated ScmApps GUI to add first run and menu option for mirrorpicker GUI.
* Updated tce-status added -o option for orphan file checking.
* Updated Apps GUI to add Orphan file check.
* Updated tc-config improved PRETCE raid disk support.
* Renamed ab to tcb to avoid Apache name conflict and allow to start without specifying a search target.

Note: Requires new Xprogs.tcz also in the rc testing area until final release.
Title: Re: Core v4.6rc1
Post by: coreplayer2 on August 02, 2012, 10:03:27 PM
Thanks Roberts  am really digging the Orphan checking option thanks
Title: Re: Core v4.6rc1
Post by: AbNoRMiS on August 02, 2012, 10:07:46 PM
sorry, maybe it stupid question
but why renamed ab to tcb but not to tce?
Title: Re: Core v4.6rc1
Post by: Rich on August 02, 2012, 10:11:14 PM
Hi AbNoRMiS
I'm guessing it stands for  Tiny Core Browser.
Title: Re: Core v4.6rc1
Post by: AbNoRMiS on August 02, 2012, 10:22:51 PM
thanks Rich :)
lastly this came to me

p.s.
but agree that tce is also would be nice ;)
Title: Re: Core v4.6rc1
Post by: gerald_clark on August 03, 2012, 12:32:54 AM
Orphan file check does not work with ftp mirrors.
Mirror picker hangs on non responding mirrors like ftp.sh.cvut.cz.
Title: Re: Core v4.6rc1
Post by: hiro on August 03, 2012, 03:12:19 AM
what is this renaming craze? why not rename apache?!
Title: Re: Core v4.6rc1
Post by: curaga on August 03, 2012, 07:39:28 AM
Mirror picker hangs on non responding mirrors like ftp.sh.cvut.cz.

wget's default timeout is used currently, maybe it should be shortened a bit.

But the main issue was unresponsive mirrors, so mirrors.tcz is updated with the Czech and Poland mirrors removed (Czech unresponsive, Poland didn't carry Core any more).
Title: Re: Core v4.6rc1
Post by: roberts on August 03, 2012, 10:17:01 AM
Quote
Orphan file check does not work with ftp mirrors.

Code: [Select]
BusyBox v1.20.1 (2012-05-31 03:26:11 UTC) multi-call binary.

Usage: wget [-c|--continue] [-s|--spider] [-q|--quiet] [-O|--output-document FILE]
        [--header 'header: value'] [-Y|--proxy on/off] [-P DIR]
        [--no-check-certificate] [-U|--user-agent AGENT] [-T SEC] URL...

Retrieve files via HTTP or FTP

        -s      Spider mode - only check file existence
        -c      Continue retrieval of aborted transfer
        -q      Quiet
        -P DIR  Save to DIR (default .)
        -T SEC  Network read timeout is SEC seconds
        -O FILE Save to FILE ('-' for stdout)
        -U STR  Use STR for User-Agent header
        -Y      Use proxy ('on' or 'off')

I am using the -s option.  From above it would seem a busybox issue. I will disable orphan file checking if current mirror is an ftp site.
Title: Re: Core v4.6rc1
Post by: roberts on August 03, 2012, 10:19:20 AM
what is this renaming craze? why not rename apache?!

Craze, really?
How about you asking the Apache group not to name Apache Benchmark ab?
Such change was prompted because I am trying to improve Core's CLI usage.
Loading the apache extension supplants ab in search path.
Title: Re: Core v4.6rc1
Post by: roberts on August 03, 2012, 10:22:01 AM
thanks Rich :)
lastly this came to me

p.s.
but agree that tce is also would be nice ;)
tce is fine with me. I will accommodate.
Title: Re: Core v4.6rc1
Post by: Rich on August 03, 2012, 10:32:43 AM
Hi roberts
One small suggestion. Maybe  tce-ab , this way the naming convention will be in line with other scripts of this
nature. Leaving the  ab  in the name might serve as a reminder of what the command used to be called.
Title: Re: Core v4.6rc1
Post by: roberts on August 03, 2012, 10:39:08 AM
OK.
Title: Re: Core v4.6rc1
Post by: tinypoodle on August 03, 2012, 10:58:39 AM
While so far typing "ab" was extremely convenient when frequently used, IMHO "tce-ab" is not - as "tce-" could not be uniquely tab-completed (and after having to type "tce-a" tab-completion is not of any benefit anymore).
Plus the "-" key is somehow remote on kb in relation to the others.

Just my $0.02  ;)
Title: Re: Core v4.6rc1
Post by: Rich on August 03, 2012, 11:35:39 AM
Regardless of what it is called, one could always create a link to it with a name of their choice if they wish.
Title: Re: Core v4.6rc1
Post by: AbNoRMiS on August 03, 2012, 12:13:52 PM
perhaps there is no contradiction?
if script will be named tce-ab to satisfy naming convention
but also add link tce to tce-ab and include it in core
to satisfy those who often uses cli ab and accustomed to short name
Title: Re: Core v4.6rc1
Post by: Rich on August 03, 2012, 01:28:47 PM
Hi AbNoRMiS
Adding a third name will not help matters. There are already posts that reference the  ab  command and now there
will also be posts that reference the  tce-ab  command. Hopefully, newcomers to TC read these posts looking for
guidance, and in the interest of minimizing confusion, consistency of commands is a good thing.
Title: Re: Core v4.6rc1
Post by: AbNoRMiS on August 03, 2012, 02:22:48 PM
honestly i'm not sure that beginner when reading outdated guidance will guess that must use tce-ab instead of ab :)
for newcomers to avoid any confusion it would be nice to fix tinycore guidances according to recent changes
Title: Re: Core v4.6rc1
Post by: Rich on August 03, 2012, 02:33:42 PM
Regardless, I feel this point has been pretty well covered. If anyone feels a need to discuss this any further, I would
like to suggest they start a separate thread.
Title: Re: Core v4.6rc1
Post by: nim108 on August 03, 2012, 06:40:36 PM
Bah, was hoping for another kernel update here as 3.0.21 does not support certain processors, which is a pretty big issue (i.e. the Atom E680 in an embedded board I am using is not supposed by 3.0.21; it kernel panics on startup). Is it safe to say that the next kernel update will only happen with a major release? It is not a major issue as I have recompiled the latest 3.0.xx kernel for my needs but to make this distro more "universal" and bug-free (let's face it, 3.0.21 is quite in the early stages of a very new 3.0 kernel), it may be worth exploring the option to upgrade sooner rather than later.

Anyway, sorry to go off track. Back on topic with 4.6, it did not break anything for me when I swapped out the core.gz, so it seems to work fine here. I will report back with any issues. Thanks as always for the updates! TC is great!
Title: Re: Core v4.6rc1
Post by: curaga on August 03, 2012, 06:55:26 PM
Yes, not counting fairly serious bugs, we stick to one kernel per major release.

Do you have a link to the exact change that fixes your issue?

Quote
(let's face it, 3.0.21 is quite in the early stages of a very new 3.0 kernel)

How do you come to that? The stable updates usually only go to 7-8, the longterm/enterprise lines like 3.0 are exceptions.
Title: Re: Core v4.6rc1
Post by: nim108 on August 03, 2012, 07:40:18 PM
Yes, not counting fairly serious bugs, we stick to one kernel per major release.

Do you have a link to the exact change that fixes your issue?

Quote
(let's face it, 3.0.21 is quite in the early stages of a very new 3.0 kernel)

How do you come to that? The stable updates usually only go to 7-8, the longterm/enterprise lines like 3.0 are exceptions.
Yes, seems support for the e6xx Atom series was added in kernel 3.0.32: http://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.0.32

pch_gbe: support ML7831 IOH
commit 7756332f5b64c9c1535712b9679792e8bd4f0019 upstream.
Support new device OKI SEMICONDUCTOR ML7831 IOH(Input/Output Hub)
ML7831 is for general purpose use.
ML7831 is companion chip for Intel Atom E6xx series.
ML7831 is completely compatible for Intel EG20T PCH.

This is just an example, I've seen newer kernels in the 3.0.xx series add support for more architectures as well. Just wanted to make you guys aware of it. And about 3.0.21, I just assumed that since 3.0 was a total revamp of the Linux kernel that 3.0.21 may still be a bit buggy. I keep track of the changelogs at kernel.org with every kernel update and compared to the 2.6.xx kernels of the past, there are seemingly many more fixes with every release of a 3.0.xx kernel. I'm sure 3.0.21 will hold up just fine, it works great on every other unit I have tested it on :)

Again, sorry for going off topic here ...
Title: Re: Core v4.6rc1
Post by: tinypoodle on August 04, 2012, 03:59:19 AM
I just assumed that since 3.0 was a total revamp of the Linux kernel that 3.0.21 may still be a bit buggy.

Your assumption is based on an absolute misperception ;)
www.pcmag.com/article2/0,2817,2388926,00.asp
Title: Re: Core v4.6rc1
Post by: SvOlli on August 04, 2012, 05:41:54 AM
busybox now features an (output incompatible, but very small) implementation of lsof, which might be also worth adding.
Title: Re: Core v4.6rc1
Post by: roberts on August 08, 2012, 12:20:36 PM
Look busybox lsof applet in rc2.
Title: Re: Core v4.6rc1
Post by: tinypoodle on August 08, 2012, 12:35:23 PM
Look busybox lsof applet in rc2.

Nice to have :)