Zenwalk 6.2 BETA has been released !

Official news from the Zenwalk crew
User avatar
hyperion
Administrator
Administrator
Posts: 2696
Joined: Fri Mar 10, 2006 8:52 am
Location: Nantes - France
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by hyperion » Sun Jul 26, 2009 9:17 pm

It probably should be noted that I am unable to make XFCE to keep my selected desktop screen resolution between logouts/restarts. Not exactly a showstopper, but annoying for sure.


confirmed, and submitted to xfce bugzilla.

Workaround :

Create an executable file called :

Code: Select all

/usr/libexec/xfce-setting-display-helper.sh


Containing :

Code: Select all

#!/bin/bash
xrandr --screen 0 \
-s $(xfconf-query -c displays -p /Default/Screen_0/Resolution) \
-r $(xfconf-query -c displays -p /Default/Screen_0/RefreshRate)


And launch it at session startup (xfce application autostart)

It's the best I can provide atm

JP

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

GRUB2 destroyed my system

Post by rene » Tue Jul 28, 2009 3:30 am

Sorry for doubly posting this. This was also posted as a new topic just now, but I meant for it to be a reply in this thread (I'm testing the beta and have a few more comments, which I'll keep in this thread)

Good day.

Installing Zenwalk 6.2 BETA, or more specifically GRUB2, basically destroyed my system by wiping the partitioning.

Fortunately, I have a fairly long history with partitioning and managed to manually -- and fairly painfully -- restore the tables but for most users this would've not been possible: the logical partition chain got broken, right at the start, and the only reason I could place it back was due to me knowing both the formats and the fact that my first logical was exactly 8G in size (exactly as in to the sector).

Given this experience I can't say I feel it would be grand idea to ship 6.2 with GRUB2 as the only choice. GRUB used to a piece of shat, and GRUB2 is now.

Admittedly, the most direct aspect of GRUB2 leaving me with a dead system was probably due to me using a somewhat nonstandard layout, where I basically use the "Linux Extended" type (0x85) as the only true Linux partition, with type 0x82 and 0x83 logicals as "subpartitions" inside. Linux is my main system, so I place this "one primary Linux partition" at the start of the disk meaning my chain of logicals also starts at the beginning of the disk. Now what seems to have happened is GRUB2 falsely assuming that "an MBR" consists of more than the first sector of the disc and overwriting not just sector 0, but (at least) also sector 1 where the head node for the chain of logicals lived. If anyone cares at a technical level -- note that while this implies non-cylinder alignment nothing else in Linux ever insists on such alignment and that this in fact how sfdisk creates the chain.

And in any case, please don't let the above explanation obscure the main point. I have no doubt that GRUB2 works beautifully if presented with average, nothing special, no surprises, run of the mill systems. It's the fact that it breaks down horribly when something else comes along that shows it to be bad software. Or better, taking into account that the result is a hosed system, junk software.

I'll admit that by now LILO is getting a bit long in the tooth as well but I'll of course never again use GRUB2. LILO as a choice would be a good thing.

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Corruption and hangs on older Intel HW (865G at least)

Post by rene » Tue Jul 28, 2009 4:07 am

Good day.

After I got past the disaster called GRUB2 (see above) I managed to get 6.2 BETA installed and booting by booting it with the LILO on my Zenwalk 6.0 system so that I could in fact test it further.

My big problem with 6.0 was the xorg intel driver and the good news is that the 6.2 "intel" driver works better than the 6.0 xorg "intel" driver for me, on Intel 865G. The bad news is that the former basically didn't work at all (and I had to replace it by an older slackware i810 driver) so that this isn't in fact saying all that much.

On this hardware, and given that it's nothing special I suppose more older Intel HW, the new "intel" driver causes screen corruption and worse still, massive hangs. The corruption seems easiest to trigger by scrolling around in a terminal, but happens also in the menus and in firefox. Letters that overlay one another, sometimes strangely magnified and cursive... not good anyways.

Worse is that browsing around in firefox quickly triggers complete hangs. Everything freezes, no VT switching, no Ctrl-Alt-Del. Hitting the button on the case does work to (slowly) cleanly shut down the system, but also seems the only thing that still works. When it hangs, the HD light is periodically flashing which seems to indicate it's logging something. I didn't find any repeated messages, but did catch X segfaulting once:

Backtrace:

0: /usr/bin/X(xorg_backtrace+0x3b) [0x812c74b]
1: /usr/bin/X(xf86SigHandler+0x51) [0x80c04f1]
2: [0xb7fb3400]
3: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a3bb73]
4: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a3cf3f]
5: /usr/bin/X [0x81777b4]
6: /usr/bin/X(miPaintWindow+0x231) [0x810c8c1]
7: /usr/bin/X(miWindowExposures+0x142) [0x810cc32]
8: /usr/bin/X [0x80d7e22]
9: /usr/bin/X(MapWindow+0x2db) [0x80719db]
10: /usr/bin/X(InitRootWindow+0xfc) [0x8071afc]
11: /usr/bin/X(main+0x33e) [0x806c4ae]
12: /lib/libc.so.6(__libc_start_main+0xe5) [0xb7b916a5]
13: /usr/bin/X [0x806b9f1]

Fatal server error:
Caught signal 11. Server aborting

In short: 6.2 doesn't work on older Intel out of the box. Option "NoAccel" does work though, so maybe it's something fairly easily avoidable by some specific and minor impact Option XaaNoFoo somewhere - but, no, it's not "Option XaaNoOffscreenPixmaps" which the intel driver seems to not even be recognizing.

I noticed from the startup messages that this is a pre-release X server. Upgrading to a final release as soon as possible would be good. This is very common hardware so might be fixed upstream already.

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Miscelaneous minor comments

Post by rene » Tue Jul 28, 2009 4:26 am

Good day.

A few minor remaining remarks:

=1=

When GRUB screws you over, it screws you over good. After wiping my partition tables, it (of course) failed to boot into the just installed system and dumped me into rescue mode:

--
grub rescue> help
Unknown command `help'
Try `help' for usage
grub rescue>
--

Oh, how helpful...

=2=

On text terminal logging in (on plain VGA text terminal at least) I am greeted with

"Cannot open font file latarcyrheb-sun16"

If it matters for this, I use en_US.utf8

=3=

In Iceweasel, Help | About craps out with a box popping up saying:

"XML Parsing Error: undefined entity
Location: chrome://browser/content/aboutDialog.xul
Line Number 40, Column 9:

&logoCopyright
"

=4=

Due to me needing to use LILO from my Zenwalk 6.0 system to boot 6.2 (see above) I needed to mount the 6.2 partition from 6.0. While not a big problem this only works with the standard 6.0 kernel when you specify "ro" Otherwise, the mount fails with

"EXT4-fs: sda7: Filesystem with huge files cannot be mounted read-write without CONFIG_LFS"

No idea if 6.0 is still going to see updates, but if it's still going to get a kernel update, this would probably be good to have supported.

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Re Intel driver problems: "NoDRI" hangs everything

Post by rene » Tue Jul 28, 2009 4:54 am

Good day.

Last remark for now: with respect to the "intel" driver problems: adding "Option" "NoDRI" to the device section to try if that helps any makes X fail to even start. It hangs at a black screen.

"AIGLX" "Off" or disabling the compositor doesn't help any

User avatar
franco
Master Zenwalker
Master Zenwalker
Posts: 616
Joined: Fri Dec 28, 2007 2:36 pm
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by franco » Tue Jul 28, 2009 7:45 am

Strange thing.
I was able to install grub but I was not able to create a configuration file.
So I was booting from the install cd.(ata root=/dev/sda(?) rdinit= ro.
Then I tried SBM (smart boot manager), I install whit the iso but I find myself again in grub.
So I uninstall SBM and when I try to boot whit the cd I found a kernel panic. Cheking the files whit Gparted I found ext4 completely corrupted.
I reinstall the system and grub automatically work.
For curiosity I did the same whit my old computer and everything happen again and whit the same result.
This was whit Zen Alpha.
I install the beta in both computers and evething went smooth.
Question of space? Problem whit ext4? The mbr was not clean?

User avatar
BandiPat
Community Packager
Community Packager
Posts: 1073
Joined: Thu Feb 01, 2007 7:59 pm
Location: East Coast, USA
Contact:

Re: Miscelaneous minor comments

Post by BandiPat » Tue Jul 28, 2009 2:55 pm

rene wrote:Good day.

A few minor remaining remarks:

=1=

When GRUB screws you over, it screws you over good. After wiping my partition tables, it (of course) failed to boot into the just installed system and dumped me into rescue mode:

--
grub rescue> help
Unknown command `help'
Try `help' for usage
grub rescue>
--

Oh, how helpful...

Grub2 won't wipe your partition table info in the MBR, as it doesn't write that much data. Your MBR was already corrupted before Grub2 tried to write it's data. Why would I say this? Because I too have experienced this to the point I thought my hard drive was bad! You should clean your MBR with the "dd" command writing to the first 440 bytes with zero's. Again, not a Grub2 problem, but user error.
rene wrote:=2=

On text terminal logging in (on plain VGA text terminal at least) I am greeted with

"Cannot open font file latarcyrheb-sun16"

If it matters for this, I use en_US.utf8

This has been showing up for a long time for some reason, so apparently someone needs to be made aware of this. Doesn't really cause a problem, but it should not be there.
rene wrote:=3=

In Iceweasel, Help | About craps out with a box popping up saying:

"XML Parsing Error: undefined entity
Location: chrome://browser/content/aboutDialog.xul
Line Number 40, Column 9:

&logoCopyright"

I suspect part of this may be coming from your bad video setup. Although, many still complain about problems with Iceweasel, it should work well. I've also seen when using old settings files (~/.mozilla) problems can pop up, so be sure to remove/rename your old when starting the new 3.5 Iceweasel to check.
rene wrote:=4=

Due to me needing to use LILO from my Zenwalk 6.0 system to boot 6.2 (see above) I needed to mount the 6.2 partition from 6.0. While not a big problem this only works with the standard 6.0 kernel when you specify "ro" Otherwise, the mount fails with

"EXT4-fs: sda7: Filesystem with huge files cannot be mounted read-write without CONFIG_LFS"

No idea if 6.0 is still going to see updates, but if it's still going to get a kernel update, this would probably be good to have supported.

I'm going to guess that since Lilo was never designed to use ext4, that may be part of your problem. Goes back to using old/discontinued software can be more problem than learning how to correctly use the newer, safer software. It's all pretty much up to you how well you want your system to operate, but don't blame good software for your failure to properly research things. :-\

Shador
Master of the known universe
Master of the known universe
Posts: 755
Joined: Fri May 16, 2008 2:44 pm
Location: Munich, Germany
Contact:

Re: GRUB2 destroyed my system

Post by Shador » Tue Jul 28, 2009 6:53 pm

rene wrote:Sorry for doubly posting this. This was also posted as a new topic just now, but I meant for it to be a reply in this thread (I'm testing the beta and have a few more comments, which I'll keep in this thread)

Good day.

Installing Zenwalk 6.2 BETA, or more specifically GRUB2, basically destroyed my system by wiping the partitioning.

Fortunately, I have a fairly long history with partitioning and managed to manually -- and fairly painfully -- restore the tables but for most users this would've not been possible:

Right! But for most users such a disk layout would also have not been possible. If you are able to create such an uncommon scenario, you're probably also able to restore it.
rene wrote:the logical partition chain got broken, right at the start, and the only reason I could place it back was due to me knowing both the formats and the fact that my first logical was exactly 8G in size (exactly as in to the sector).

Given this experience I can't say I feel it would be grand idea to ship 6.2 with GRUB2 as the only choice. GRUB used to a piece of shat, and GRUB2 is now.

Admittedly, the most direct aspect of GRUB2 leaving me with a dead system was probably due to me using a somewhat nonstandard layout, where I basically use the "Linux Extended" type (0x85) as the only true Linux partition, with type 0x82 and 0x83 logicals as "subpartitions" inside. Linux is my main system, so I place this "one primary Linux partition" at the start of the disk meaning my chain of logicals also starts at the beginning of the disk. Now what seems to have happened is GRUB2 falsely assuming that "an MBR" consists of more than the first sector of the disc and overwriting not just sector 0, but (at least) also sector 1 where the head node for the chain of logicals lived. If anyone cares at a technical level -- note that while this implies non-cylinder alignment nothing else in Linux ever insists on such alignment and that this in fact how sfdisk creates the chain.

And in any case, please don't let the above explanation obscure the main point. I have no doubt that GRUB2 works beautifully if presented with average, nothing special, no surprises, run of the mill systems. It's the fact that it breaks down horribly when something else comes along that shows it to be bad software. Or better, taking into account that the result is a hosed system, junk software.

I'll admit that by now LILO is getting a bit long in the tooth as well but I'll of course never again use GRUB2. LILO as a choice would be a good thing.

Can you please report this problem to the grub-devs. It would really help a lot more if you told them were problems exists than if you tell them that their software is just plain shit. Bugs are common and will occur always and especially due to that such critical software like grub2 needs lots of lots of testing.
This sounds to me as if they just missed a check for a somewhat special scenario and accidentally overwrote the table. Usually in most scenarios this won't happen.
Look here: http://www.gnu.org/software/grub/grub-2-bugs.en.html
You can also talk directly to the devs on their maillist (grub-devel) or on their irc channel grub on freenode.net. They really care about standards compliance and not overwriting possible needed data is definitely a standard. :) Though if the post-MBR gap exits it can be used in any way, so there's nothing forbidding grub2 there but also nothing warning it about your head node and that gap surely existed if grub2 wrote there it does never do any assumptions about the MBR length.

Why are Debian, Ubuntu, suse and all those giants shipping grub legacy and (partially) soon grub2? Right, it has or will have support for many features like network boot, nice gui, disaster recovery, authentication/password protection, ...


And, PLEASE, change you attitude from: "New things are bad, old should be patched until they are crap" to: "Some new things are bad and there's point where old things should be replaced by new redesigned replacements, even though there may exist still some faults." Lilo surely also had bugs in the past and got rid of them like most old software does. ;)

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by rene » Tue Jul 28, 2009 11:32 pm

BandiPat:

Please note that I am very definitely not a Linux newbie and do in fact know what I am talking about. I'm reporting problems and just about the least helpful thing you can do for any single project is focus on a report instead of a problem reported.

And especially when you then seem to have little idea what you're in fact saying. With respect to your answers:

1. Here I'm reporting that the build of GRUB as shipped replies to the command 'help' with "unknown command, please try 'help'.", not anything else.

And otherwise, no, my MBR / partition tables are definitely not corrupt and yes, installing GRUB overwrote more than the first sector on disk: at least 2. I actually stared at the damage in a disk editor. It did, that's the bug I'm reporting.

3. No, this doesn't have anything to do with video. It's an XML parsing error for heaven's sake. Old .mozilla could be.

4. No, this hasn't anything to do with LILO. This is just reporting that the ZW 6.0 kernel due to being a !CONFIG_LFS kernel won't mount a ZW 6.2 root filesystem read-write (and other than that, the entire point of LILO versus GRUB is that it doesn't know about any filesystem but simply asks the kernel for a list of sectors).

It's just meant as a heads up to whomever does your kernel packaging that if ZW 6.0 is still going to see a new kernel it might want to have CONFIG_LFS enabled. Just the last of the minor "let's try to be helpful and report the things I encounter while testing this beta" points, not a bug report.

In short: thank you ever so much for confirming that I'm not chronically delusional when I see a message about a font. Otherwise, I seem to have gotten fairly little from the reply so please feel free to not post any followups.

User avatar
Duncan_Idaho
Regular Zenwalker
Regular Zenwalker
Posts: 96
Joined: Tue Jan 08, 2008 8:35 pm
Location: Sietch Tabr, Arrakis
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by Duncan_Idaho » Wed Jul 29, 2009 12:00 am

this is kinda offtopic but I can confirm the mozilla bug in a clean install :-\
Trust decentralization over centralization, voluntarism over coercion, bottom-up over top-down, adaptation over planning, openness over secrecy, practice over ideology, and markets over politics. Freedom works. Now go do it.
--Eric S. Raymond

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by rene » Wed Jul 29, 2009 12:19 am

Shador:

Thanks for the reply.

Well, actually, "sfdisk" simply creates the tables this way. Admittedly, most people don't use sfdisk either but normal fdisk or cfdisk does the same with respect to the layout. Those do insist on cylinder alignment more (ie, the head of the chain of logicals wouldn't generally live in the second sector) but the point here really is that installing GRUB should never ever overwrite more than the first sector since it's never guaranteed to be available for use.

Second, don't get me wrong, I agree that directly reporting to the GRUB folks would be helpful but here it's definitely relevant as well. I mean, really, no matter what else, installing Zenwalk 6.2 BETA on a dedicated testing partition had the effect of destroying everything on the disk and that's a pretty serious bug to report. The point of beta's is getting bugs reported...

I'm here from several years of basically using my own distribution, also on a slackware base, and mostly because I started lacking the time and inclination to deal with all the problems directly myself. Generally, users of a distribution are expected to report problems to the distribution after which the distribution filters the noise and relays that which is applicable upstream. Yeah, I understand that Zenwalk might be a bit small and non-professional (in a financial sense) to have that be as true as it would be for the bigger distributions, but hey, it's the idea of things...

You do seem to know what you're talking about with respect to "the gap" so let me assure you that yes, there really wasn't one. Here's my partition layout:

Code: Select all

   Device Boot    Start       End   #sectors  Id  System
/dev/sda1             1 692272486  692272486  85  Linux extended
/dev/sda2     692272487 976773167  284500681   7  HPFS/NTFS
/dev/sda3             0         -          0   0  Empty
/dev/sda4             0         -          0   0  Empty
/dev/sda5             2  16777217   16777216  83  Linux
/dev/sda6      16777219  33554434   16777216  83  Linux
/dev/sda7      33554436  50331651   16777216  83  Linux
/dev/sda8      50331653  67108868   16777216  83  Linux
/dev/sda9      67108870 379690677  312581808  83  Linux
/dev/sda10    379690679 692272486  312581808  83  Linux


The tables therefore live in sector 0 (MBR), 1 (first sector of the extended), 16777218 and so on. What happened is that not just sector 0 but also sector 1 got overwritten while installing 6.2 BETA. After figuring out what the problem was (I of course had a dead system at that point) I could manually restore sector 1 but really only because I also knew that my /dev/sda5 was exactly 16777216 sectors in size. Had that not been the case, I wouldn't have known to pick up the chain at sector 16777218. That is, definitely more luck than anything else that I was able to patch things back again.

Finally, do please note that my attitude is definitely not that new things are bad. My attitude is that new things that destroy my computer are bad. I love most other new things...

I will contact the GRUB developers. I expect they'll just boot me back here though since they'll assume (as you do) that they never make assumptions about MBR length. They seem to do though, and do so as a bug...

User avatar
BandiPat
Community Packager
Community Packager
Posts: 1073
Joined: Thu Feb 01, 2007 7:59 pm
Location: East Coast, USA
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by BandiPat » Wed Jul 29, 2009 7:34 pm

Rene,
Don't know if you posted on the grub-dev list or someone passed it along for you, but wanted to make you aware one of the devs did answer. Since I'm not sure if you will see it there, I'll post it here for everyone.
> But, well. Both LILO and old GRUB support this configuration and it's
> a valid configuration as far as the tables themselves are concerned.
> If GRUB2 cannot support this, I consider that a regression. I know
> LILO a bit, but not old GRUB. Cannot GRUB2 handle this in whatever way
> GRUB does/did?
The way you refer to is called blocklists. Basically LILO saves a list
of sectors where its files are. This approach was an error from the
begining since filesystem has a complete right to move files to
another sectors. At first it was manageable because filesystems tended
to let files in place but there were a lot of situations where this
would lead to weird effect. E.g.
1) install grub-legacy
2) rm -rf /boot/grub/stage2
3) Reboot
It will boot even though stage2 was deleted because blocks still
contain stage2.
4) put stage2 to /boot/grub/stage2
Now your system looks completely ok but new stage2 is probably
residing in another blocks and after random amount of time it will
stop booting.
Now with advent of advanced filesystems blocklists may fail even if
you don't do anything because new filesystem use their right to move
files to other blocks
This is actually situation of "pay now or pay later". We try not to
break configurations used in major distributions if possible and not
too much work to support them reliably. But you have used tools in a
way to force them to create a bad configuration. Don't expect us to
support this. Additionally your configuration has no benefit over
standard one. You say that it's easy to remember to recover after
failure. If you use sane configuration you'll have no failures. And
first partition starting at sector number e.g. 1000 or 2000 isn't
more difficult to remember than it starting at sector 1. 1000 sectors
is only 500 KiB you probably won't miss them much but it gives grub2
enough space for a lot of things.
-
Regards
Vladimir 'phcoder' Serbinenko

Personal git repository: http://repo.or.cz/w/grub2/phcoder.git

Oh wait, you ask me not to post a followup, didn't you? See how stupid I am! I can't even follow simple directions. :D

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by rene » Wed Jul 29, 2009 7:47 pm

Following up on this, the GRUB2 bug was confirmed by the GRUB2 developers.

It was also located, but is going to be fixed only in the sense that it will fail to install instead of overwriting extra sectors. GRUB2 simply cannot work without a gap after the MBR by design which on a purely technical level makes sense as a decision for GRUB2 but which, given that both LILO and old GRUB can, does still count as somewhat of a regression as fas as I'm concerned,

In any case: heads up, a version that'll barf but at least not destroy is coming.

Bandipat: was writing the above while you posted. Yes, I got that in private mail.

But note, my configuration isn't in any sense "bad", it's just uncommon and GRUB2 has decided that it's not going to be able to be installed into the MBR when there's no gap. Somewhat fair enough. I might be persuaded to view this as saying that I'm going to again keep a shared /boot around onto which I'll install it. Haven't tried it but the 6.2 install did offer the option of installing into a bootblock as well. Don't yet know how the size requirements are going to pan out there, but ...

For now, I'll just take it to mean I'll avoid GRUB2. And I'd definitely advise to ship 6.2 with a GRUB2 that'll at least fail instead of destroy.

With respect to the Intel driver issue: the hangs have been widely reported, but no solutions yet. Arch Linux is shipping an additional xf86-video-intel-legacy package with an older "intel" driver which works better -- although that one won't allow me to enable DRI due to memory allocation failures linked to a memory size misdetection in that one. Sigh.

The joys of open source. Basically, nothing just works, ever. Trust me, in 10 years you'll have grown very tired of it too.
Last edited by rene on Wed Jul 29, 2009 8:05 pm, edited 1 time in total.

Shador
Master of the known universe
Master of the known universe
Posts: 755
Joined: Fri May 16, 2008 2:44 pm
Location: Munich, Germany
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by Shador » Wed Jul 29, 2009 7:59 pm

rene wrote:1. Here I'm reporting that the build of GRUB as shipped replies to the command 'help' with "unknown command, please try 'help'.", not anything else.

This is a grub rescue mode problem, the help command isn't loaded. This should probably be reported too. But it's defnitely no zenwalk specific packaging prolem.

rene wrote:Second, don't get me wrong, I agree that directly reporting to the GRUB folks would be helpful but here it's definitely relevant as well. I mean, really, no matter what else, installing Zenwalk 6.2 BETA on a dedicated testing partition had the effect of destroying everything on the disk and that's a pretty serious bug to report. The point of beta's is getting bugs reported...

I'm here from several years of basically using my own distribution, also on a slackware base, and mostly because I started lacking the time and inclination to deal with all the problems directly myself. Generally, users of a distribution are expected to report problems to the distribution after which the distribution filters the noise and relays that which is applicable upstream. Yeah, I understand that Zenwalk might be a bit small and non-professional (in a financial sense) to have that be as true as it would be for the bigger distributions, but hey, it's the idea of things...

Of course, you're right here, but it also isn't wrong to ask you to do it yourself, especially if it's something more complicated which you do know best.
Getting bugs reported with a beta is exactly what we want, but we also want reporters to know that a beta isn't perfect and to write their bug reports accordingly. Calling something "shat" isn't very motivating, diplomatic and won't animate people to take your opinion serious. :)

rene
Regular Zenwalker
Regular Zenwalker
Posts: 40
Joined: Tue Mar 10, 2009 8:36 pm
Contact:

Re: Zenwalk 6.2 BETA has been released !

Post by rene » Wed Jul 29, 2009 8:24 pm

Well, that last problem neatly solves itself, since I don't take developers who think diplomacy is always a good thing serious. The open source crowd and diplomacy? Whaha! When they've just spent an hour manually patching back a partition table? Whahoeha!

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests