Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
Desktop:
https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test
Ideally, all Alpha priority test cases for installation [2] and desktop [3] should pass in order to meet the Alpha Release Criteria [4]. Help is available on #fedora-qa on irc.freenode.net [5], or on the test list [6].
[1] http://poelstra.fedorapeople.org/schedules/f-15/f-15-quality-tasks.html [2] https://fedoraproject.org/wiki/QA:Installation_validation_testing [3] https://fedoraproject.org/wiki/QA:Desktop_validation_testing [4] https://fedoraproject.org/wiki/Fedora_15_Alpha_Release_Criteria [5] irc://irc.freenode.net/fedora-qa [6] https://admin.fedoraproject.org/mailman/listinfo/test
_______________________________________________ test-announce mailing list test-announce@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/test-announce
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
Desktop:
https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test
Ideally, all Alpha priority test cases for installation [2] and desktop [3] should pass in order to meet the Alpha Release Criteria [4]. Help is available on #fedora-qa on irc.freenode.net [5], or on the test list [6].
[1] http://poelstra.fedorapeople.org/schedules/f-15/f-15-quality-tasks.html [2] https://fedoraproject.org/wiki/QA:Installation_validation_testing [3] https://fedoraproject.org/wiki/QA:Desktop_validation_testing [4] https://fedoraproject.org/wiki/Fedora_15_Alpha_Release_Criteria [5] irc://irc.freenode.net/fedora-qa [6] https://admin.fedoraproject.org/mailman/listinfo/test
Please note that due to some broken deps there is no KDE live images. when the deps are resolved we will create and push them
Dennis
Quick question: When installing from the i386 DVD, I noticed that the installed skipped the first part of anaconda where the blue/red screen comes up and offers to check the install media. The installer just skipped to the graphical install. Is this a bug or something new? I've got a problem TC2 and mutter constantly crashing on login so I'm check the media manually on my F14 box but the on-install media check was nice.
-- John Watzke
On Tue, 2011-02-15 at 00:07 -0600, John Watzke wrote:
Quick question: When installing from the i386 DVD, I noticed that the installed skipped the first part of anaconda where the blue/red screen comes up and offers to check the install media. The installer just skipped to the graphical install. Is this a bug or something new?
https://bugzilla.redhat.com/show_bug.cgi?id=676551
On 02/15/2011 06:07 AM, John Watzke wrote:
Quick question: When installing from the i386 DVD, I noticed that the installed skipped the first part of anaconda where the blue/red screen comes up and offers to check the install media. The installer just skipped to the graphical install. Is this a bug or something new? I've got a problem TC2 and mutter constantly crashing on login so I'm check the media manually on my F14 box but the on-install media check was nice.
Do you have a ATI graphics card by any chance?
I'm wondering if we are seeing a trend happening here with ATI graphics cards...
Btw Install abrt-cli package and switch to terminal and file a bug. ( abrt-cli -l to list crashes and abrt-cli -r @$number to report it )
JBG
On Tue, 2011-02-15 at 07:00 +0000, "Jóhann B. Guðmundsson" wrote:
On 02/15/2011 06:07 AM, John Watzke wrote:
Quick question: When installing from the i386 DVD, I noticed that the installed skipped the first part of anaconda where the blue/red screen comes up and offers to check the install media. The installer just skipped to the graphical install. Is this a bug or something new? I've got a problem TC2 and mutter constantly crashing on login so I'm check the media manually on my F14 box but the on-install media check was nice.
Do you have a ATI graphics card by any chance?
I'm wondering if we are seeing a trend happening here with ATI graphics cards...
Not really. There are cards from all three major vendors which are broken, and cards which work. It's not really unexpected, we just need to make sure bugs are filed for all the broken ones.
On 02/15/2011 07:10 AM, Adam Williamson wrote:
Not really. There are cards from all three major vendors which are broken, and cards which work. It's not really unexpected, we just need to make sure bugs are filed for all the broken ones.
If you are aware of several cards that are broken from three major vendors than perhaps *hint* *hint* you should setup a wiki page which contains a list of all known broken cards with corresponding bug reports so..
A) reporters can be point out to that wiki page to check if they have any of the card on that list . B) keep track of process fixing bugs corresponding to those cards C) reporters could add their cards and bug report to that list D) <insert something more here >
Perhaps we might also want to create a tracker bug for each of those major vendors...
=)
JBG
On Tue, 2011-02-15 at 07:47 +0000, "Jóhann B. Guðmundsson" wrote:
On 02/15/2011 07:10 AM, Adam Williamson wrote:
Not really. There are cards from all three major vendors which are broken, and cards which work. It's not really unexpected, we just need to make sure bugs are filed for all the broken ones.
If you are aware of several cards that are broken from three major vendors than perhaps *hint* *hint* you should setup a wiki page which contains a list of all known broken cards with corresponding bug reports so..
No. I don't want to do that. It would be an ugly, manual (and therefore broken) way to do it. What I'd want to do is just set up a tracker bug for cases of GNOME Shell being borked on cards where we'd expect it to work, and probably another for cases of fallback failing on cards where we don't expect GNOME Shell to work.
Perhaps we might also want to create a tracker bug for each of those major vendors...
Yes, but I don't think it's really necessary to split by vendor.
On 02/15/2011 08:24 AM, Adam Williamson wrote:
probably another for cases of fallback failing on cards where we don't expect GNOME Shell to work.
I'm pretty sure this one needs to be wikified and stuffed into release notes and what not.
Do you have this list handy somewhere?
If so you can mail it to me and I can mockup a wikipage for it..
JBG
On Tue, 2011-02-15 at 08:44 +0000, "Jóhann B. Guðmundsson" wrote:
On 02/15/2011 08:24 AM, Adam Williamson wrote:
probably another for cases of fallback failing on cards where we don't expect GNOME Shell to work.
I'm pretty sure this one needs to be wikified and stuffed into release notes and what not.
Do you have this list handy somewhere?
If so you can mail it to me and I can mockup a wikipage for it..
GNOME's fallback mechanism is too specific a thing to put in the criteria. The criterion that covers it would be:
"In most cases, the installed system must boot to a functional graphical environment without user intervention (see Blocker_Bug_FAQ) "
Obviously, when GNOME's fallback mechanism fails, that criterion is not satisfied. If we have enough cases of that being broken, then we trigger the 'in most cases' clause, and say the criterion is generally not met by the release.
On Tue, 2011-02-15 at 08:44 +0000, "Jóhann B. Guðmundsson" wrote:
On 02/15/2011 08:24 AM, Adam Williamson wrote:
probably another for cases of fallback failing on cards where we don't expect GNOME Shell to work.
I'm pretty sure this one needs to be wikified and stuffed into release notes and what not.
It'd just go into commonbugs.
Do you have this list handy somewhere?
I'm working on it now.
If so you can mail it to me and I can mockup a wikipage for it..
No need, thanks.
Do you have a ATI graphics card by any chance?
I'm wondering if we are seeing a trend happening here with ATI graphics cards...
Btw Install abrt-cli package and switch to terminal and file a bug. ( abrt-cli -l to list crashes and abrt-cli -r @$number to report it )
Thanks for the tip on abtr-cli. I figured there was some sort of cli to report this but I didn't have a chance to look it up last night as I was getting pretty sleepy. Yes, this is an ATI card and it worked just fine in both the gnome3 test day and TC1. I'll unleash the bugs tonight when I get home.
-- John Watzke
On Tue, 2011-02-15 at 08:34 -0600, John Watzke wrote:
Do you have a ATI graphics card by any chance? I'm wondering if we are seeing a trend happening here with ATI graphics cards... Btw Install abrt-cli package and switch to terminal and file a bug. ( abrt-cli -l to list crashes and abrt-cli -r @$number to report it )
Thanks for the tip on abtr-cli. I figured there was some sort of cli to report this but I didn't have a chance to look it up last night as I was getting pretty sleepy. Yes, this is an ATI card and it worked just fine in both the gnome3 test day and TC1. I'll unleash the bugs tonight when I get home.
Have you reported your issues yet? Thanks.
On Tue, 2011-02-15 at 00:07 -0600, John Watzke wrote:
Quick question: When installing from the i386 DVD, I noticed that the installed skipped the first part of anaconda where the blue/red screen comes up and offers to check the install media. The installer just skipped to the graphical install.
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img. There is some discussion (and I believe a bug report) on whether or not to prompt for mediacheck. The only prompt for external resources will be the yum repository screen. Here you can supply additional package repositories (local + remote).
Thanks, James
On 2011/02/15 08:36 (GMT-0500) James Laska composed:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img.
And goodbye to my to my customary installation method of loading installation kernel and initrd from my many little 200M /boot partitions that don't have room for being made larger. Good thing the two files needn't be located the same place, but it makes keeping track of where they are more complex. I guess I can use home's root or give home's root an install dir, and put even kernel there too. Or, maybe on my user/local partition. Or, on a FAT or NTFS partition?!?!?
On 02/15/2011 10:40 AM, Felix Miata wrote:
On 2011/02/15 08:36 (GMT-0500) James Laska composed:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img.
And goodbye to my to my customary installation method of loading installation kernel and initrd from my many little 200M /boot partitions that don't have room for being made larger. Good thing the two files needn't be located the same place, but it makes keeping track of where they are more complex. I guess I can use home's root or give home's root an install dir, and put even kernel there too. Or, maybe on my user/local partition. Or, on a FAT or NTFS partition?!?!?
Hm, yeah, this is going to break koan as well, isn't it for systems with small /boot filesystems.
Once upon a time, Orion Poplawski orion@cora.nwra.com said:
On 02/15/2011 10:40 AM, Felix Miata wrote:
On 2011/02/15 08:36 (GMT-0500) James Laska composed:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img.
And goodbye to my to my customary installation method of loading installation kernel and initrd from my many little 200M /boot partitions that don't have room for being made larger. Good thing the two files needn't be located the same place, but it makes keeping track of where they are more complex. I guess I can use home's root or give home's root an install dir, and put even kernel there too. Or, maybe on my user/local partition. Or, on a FAT or NTFS partition?!?!?
Hm, yeah, this is going to break koan as well, isn't it for systems with small /boot filesystems.
Hmm, also what does this do to PXE booting. IIRC there is a (relatively low) limit on the size of the initrd loaded by pxelinux.
On 02/15/2011 11:59 AM, Chris Lumens wrote:
Hmm, also what does this do to PXE booting. IIRC there is a (relatively low) limit on the size of the initrd loaded by pxelinux.
It's worked fine for me in all systems tested.
- Chris
Me too. Though if you don't have enough memory I got a traceback that was too big to fit on the screen so was hard to debug.
On Tue, 15 Feb 2011, Chris Adams wrote:
Hmm, also what does this do to PXE booting. IIRC there is a (relatively low) limit on the size of the initrd loaded by pxelinux.
Most tftp servers can cope with bigger sizes now, but not necessarily in consistent ways. pxelinux talking to linux tftp should be fine, but expect problems if you use a Solaris tftp server.
Michael Young
On Tue, 2011-02-15 at 11:28 -0700, Orion Poplawski wrote:
On 02/15/2011 10:40 AM, Felix Miata wrote:
On 2011/02/15 08:36 (GMT-0500) James Laska composed:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img.
And goodbye to my to my customary installation method of loading installation kernel and initrd from my many little 200M /boot partitions that don't have room for being made larger. Good thing the two files needn't be located the same place, but it makes keeping track of where they are more complex. I guess I can use home's root or give home's root an install dir, and put even kernel there too. Or, maybe on my user/local partition. Or, on a FAT or NTFS partition?!?!?
Hm, yeah, this is going to break koan as well, isn't it for systems with small /boot filesystems.
I could be wrong, but I don't think koan was initially designed with a focus on small /boot scenarios. Given it just downloads the vmlinuz +initrd.img, it certainly lends well to that scenario. However, any tool that's downloading vmlinuz+initrd.img to prep a system for install will be impacted by any file system layout for '/' or '/boot'.
I suspect we'll have several additional tools that need adjustment to handle any /boot (or /) size restrictions.
Thanks, James
On Tue, 2011-02-15 at 08:36 -0500, James Laska wrote:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img. There is some discussion (and I believe a bug report) on whether or not to prompt for mediacheck. The only prompt for external resources will be the yum repository screen. Here you can supply additional package repositories (local + remote).
Soo, does there still need to be a install/install.img file for installing from iso via network, or a initrd.img or anything at all or just the iso is now needed and nothing else?
On Tue, 2011-02-15 at 17:11 -0600, Mike Chambers wrote:
On Tue, 2011-02-15 at 08:36 -0500, James Laska wrote:
Good-bye install.img, hello initrd.img!
https://fedoraproject.org/wiki/Anaconda/Features/UnifiedInitrd
The installer no longer needs to locate where install.img is (network, DVD, HD). The content previously included in the stage#2 install.img file, is now included in the initrd.img. There is some discussion (and I believe a bug report) on whether or not to prompt for mediacheck. The only prompt for external resources will be the yum repository screen. Here you can supply additional package repositories (local + remote).
Soo, does there still need to be a install/install.img file for installing from iso via network, or a initrd.img or anything at all or just the iso is now needed and nothing else?
There shouldn't be a need for this, since it no longer exists. Just boot and go (or that's the theory). The only question should be where do you want to install packages from. That step will happen during the graphical reposetup portion of the installer. Of course, this is why we test, so we can shake out any issues that slipped through the cracks.
Hope that helps answer your question! James
Hi,
2011/2/14 Andre Robatino robatino@fedoraproject.org:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
Desktop:
https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test
Ideally, all Alpha priority test cases for installation [2] and desktop [3] should pass in order to meet the Alpha Release Criteria [4]. Help is available on #fedora-qa on irc.freenode.net [5], or on the test list [6].
Two issues here. On firstboot, when I tried to type my name I've got an exception. I think, that lettel "ł" caused it. Second issue - I don't have a "restart" option in Gnome menu.
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
Hi,
2011/2/14 Andre Robatino robatino@fedoraproject.org:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
Desktop:
https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test
Ideally, all Alpha priority test cases for installation [2] and desktop [3] should pass in order to meet the Alpha Release Criteria [4]. Help is available on #fedora-qa on irc.freenode.net [5], or on the test list [6].
Two issues here. On firstboot, when I tried to type my name I've got an exception. I think, that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
ti., 15.02.2011 kl. 21.25 -0800, skrev Adam Williamson:
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
We still have to reboot to use a newly installed kernel. Isn't that a valid enough use case for this? Aren't there other kinds of updates that mandate a reboot as well?
gnome-packagekit has several strings referring to restart being needed after updating system packages...
Cheers Kjartan
On 02/16/2011 08:01 AM, Kjartan Maraas wrote:
We still have to reboot to use a newly installed kernel. Isn't that a valid enough use case for this? Aren't there other kinds of updates that mandate a reboot as well?
gnome-packagekit has several strings referring to restart being needed after updating system packages...
Well I dont think the UI Designers are going to remove the ability to reboot from update application itself but then again you never know..
JBG.
On Wed, 2011-02-16 at 09:01 +0100, Kjartan Maraas wrote:
We still have to reboot to use a newly installed kernel. Isn't that a valid enough use case for this? Aren't there other kinds of updates that mandate a reboot as well?
Updates that require a reboot are supposed to be handled by packagekit...
gnome-packagekit has several strings referring to restart being needed after updating system packages...
...indeed, so if you do it via the update manager, you don't need a desktop-provided option to do so.
seems I was a bit out of date, though, as Bastien replied on the desktop list: reboot may come back as a de-emphasized option in the shutdown menu somewhere.
On 02/16/2011 12:25 AM, Adam Williamson wrote:
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
Well, I reboot after every kernel update, every time there are numerous updates from the repos, when debugging and testing changes to gdm, dracut, systemd, networking, and on-and-on.
I am somehow now reminded of the early days of software design (I go back to second gen mainframes) when we sat in a room and designed software the way we thought it should be and then cursed the users who complained it wasn't the way they wanted it. Heaven forbid that we even considered an upfront requirements definition phase that included the user community. Of course, we blamed users for the additional costs involved in "correcting" errors late in the projects and after they were implemented.
I am optimistic, however, since the Fedora project seems to always, somehow, make the right, even unpopular, decisions concerning features. I.e., deferring systemd to F15 late in the F14 cycle.
I suppose gnome is too far down the path now to consider deferring it to F16? Is it even possible?
On Wed, 2011-02-16 at 08:52 -0500, Clyde E. Kunkel wrote:
On 02/16/2011 12:25 AM, Adam Williamson wrote:
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
Well, I reboot after every kernel update, every time there are numerous updates from the repos, when debugging and testing changes to gdm, dracut, systemd, networking, and on-and-on.
I am somehow now reminded of the early days of software design (I go back to second gen mainframes) when we sat in a room and designed software the way we thought it should be and then cursed the users who complained it wasn't the way they wanted it. Heaven forbid that we even considered an upfront requirements definition phase that included the user community. Of course, we blamed users for the additional costs involved in "correcting" errors late in the projects and after they were implemented.
I am optimistic, however, since the Fedora project seems to always, somehow, make the right, even unpopular, decisions concerning features. I.e., deferring systemd to F15 late in the F14 cycle.
I suppose gnome is too far down the path now to consider deferring it to F16? Is it even possible?
Development has been going on for some time now, so it would be a shame for GNOME3 to miss Fedora 15. The desktop team has set the bar in terms of expected functionality and criteria for a successful GNOME3.0 + Fedora 15 release.
https://fedoraproject.org/wiki/Features/Gnome3#Scope
QA is positioned well to provide feedback against the goals listed above, by way of bugs and flame-free discussion, around those goals.
Thanks, James
On 02/16/2011 10:25 AM, James Laska wrote:
On Wed, 2011-02-16 at 08:52 -0500, Clyde E. Kunkel wrote:
On 02/16/2011 12:25 AM, Adam Williamson wrote:
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
<snip> that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
Well, I reboot after every kernel update, every time there are numerous updates from the repos, when debugging and testing changes to gdm, dracut, systemd, networking, and on-and-on.
I am somehow now reminded of the early days of software design (I go back to second gen mainframes) when we sat in a room and designed software the way we thought it should be and then cursed the users who complained it wasn't the way they wanted it. Heaven forbid that we even considered an upfront requirements definition phase that included the user community. Of course, we blamed users for the additional costs involved in "correcting" errors late in the projects and after they were implemented.
I am optimistic, however, since the Fedora project seems to always, somehow, make the right, even unpopular, decisions concerning features. I.e., deferring systemd to F15 late in the F14 cycle.
I suppose gnome is too far down the path now to consider deferring it to F16? Is it even possible?
Development has been going on for some time now, so it would be a shame for GNOME3 to miss Fedora 15. The desktop team has set the bar in terms of expected functionality and criteria for a successful GNOME3.0 + Fedora 15 release.
https://fedoraproject.org/wiki/Features/Gnome3#Scope
QA is positioned well to provide feedback against the goals listed above, by way of bugs and flame-free discussion, around those goals.
Thanks, James
Not a flame. Just frustration. At least some discussion is now underway and folks are expressing opinions.
I am now seeing links to pages that I missed so far during Gnome 3 development. I am sure they will be useful and I am sure things will work out, they always do.
Reminds me of the marketing folks during my business life: "Tell them three times what you want them to know. Then keep telling them until they buy." Of course, your product has to work. :-)
On Wed, 2011-02-16 at 11:43 -0400, Clyde E. Kunkel wrote:
On 02/16/2011 10:25 AM, James Laska wrote:
On Wed, 2011-02-16 at 08:52 -0500, Clyde E. Kunkel wrote:
On 02/16/2011 12:25 AM, Adam Williamson wrote:
On Tue, 2011-02-15 at 21:45 -0500, Clyde E. Kunkel wrote:
On 02/15/2011 09:06 PM, Adam Williamson wrote:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote: > <snip> > that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
> option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
It was discussed during the Test Day. I don't have a web reference for this exact issue, but the position of the design team is that they think the only common use case for rebooting is to boot into a different operating system in a multi-boot configuration, and they want to handle that as a special case somehow (a direct 'reboot to Windows' option has been suggested). They don't believe there are any sufficiently common use cases for rebooting other than that one to justify the added complexity of providing it as an option. (Desktop team, please correct me if I'm representing this wrong).
Well, I reboot after every kernel update, every time there are numerous updates from the repos, when debugging and testing changes to gdm, dracut, systemd, networking, and on-and-on.
I am somehow now reminded of the early days of software design (I go back to second gen mainframes) when we sat in a room and designed software the way we thought it should be and then cursed the users who complained it wasn't the way they wanted it. Heaven forbid that we even considered an upfront requirements definition phase that included the user community. Of course, we blamed users for the additional costs involved in "correcting" errors late in the projects and after they were implemented.
I am optimistic, however, since the Fedora project seems to always, somehow, make the right, even unpopular, decisions concerning features. I.e., deferring systemd to F15 late in the F14 cycle.
I suppose gnome is too far down the path now to consider deferring it to F16? Is it even possible?
Development has been going on for some time now, so it would be a shame for GNOME3 to miss Fedora 15. The desktop team has set the bar in terms of expected functionality and criteria for a successful GNOME3.0 + Fedora 15 release.
https://fedoraproject.org/wiki/Features/Gnome3#Scope
QA is positioned well to provide feedback against the goals listed above, by way of bugs and flame-free discussion, around those goals.
Thanks, James
Not a flame. Just frustration. At least some discussion is now underway and folks are expressing opinions.
Doh, sorry Clyde. I didn't take your mail that way, or mean to imply it was. Just doing my best to encourage positive discussion on the topic! :)
Reminds me of the marketing folks during my business life: "Tell them three times what you want them to know. Then keep telling them until they buy." Of course, your product has to work. :-)
I recall those same words in the context of public speaking. So true :)
Thanks, James
On 02/16/2011 02:45 AM, Clyde E. Kunkel wrote:
option in Gnome menu.
That is a, ahem, policy decision.
Hi, Adam,
Could you say some more about the policy decision or point to a msg or whatever. Need to understand what is going on.
Bring up this questionable UI Design up on the desktop list.
It beginning to become more and more apparent that the Gnome developers forgot to sprinkle some experience wisdom and common sense into the Gnome UI Designers when they went off the charts as they have done with this decisions.
I think one triple U distribution that exist out there more or less succession is based on fixing those UI Design Bugs that upstream Gnome UI Designers get wrong which seems to be Gnomes upstream Achilles' heel they get soo many things right but then screw it over minor nuances that irritates the end user so much he stops using it and or chooses a distribution that ships workaround hacks for those nuances..
If the Gnome Desktop was not the distributions "Default" and the Gnome Desktop Team was treated and acted equally to any other SIG I would point you to their relevant meeting or discussion on the mailing list but since that does not exist I cant.
One of the pitfalls of having a "Default" is that community surrounding the "Default" has something to say in what's being shipped and in this case it just so happens to be majority of our existing user base and unfortunately the Gnome Desktop Team has had a bit history ignoring existing user base and jump to the other side where the grass is greener chasing girl scouts and the end of the rainbow unfortunately it looks like the board has taken that attitude as well.
We could I suppose setup a wiki page that gathers all those questionable UI design decisions along with patches for those designers bugs and start patching it here downstream with or without the Gnome Desktop Team consent ( which most likely they would object since they are more or less upstream ) however that would go against our whole upstream mantra and ofcourse is not the smartest thing to do but might be necessary to keep our users happy.
But before taking any workaround steps talk to the maintainers on the desktop list they are present and responsive and with luck they can *convince* the UI designers that perhaps it's better to keep something still around like for example the ability to reboot as people have been doing since the dawn of personal computer.
Note I dont think any upstream UI designers are present/subscribed on that list..
I'm expecting FESCO to hold the same standards as they set and did with systemd since this affects majority of Fedora userbase and is the "Default" thus affects majority of our users and significantly proportion of the reputation of the project and if Gnome3 does not satisfy that standard they deem it not ready and postpone it to F16 and we ship Gnome2 with F15.
The work around is obvious open up a terminal and reboot as is the other one for many of those design bug open up terminal and run gconf/dconf editor..
The ability to reboot is still present in GDM for now at least.
JBG
2011/2/16 Adam Williamson awilliam@redhat.com:
On Tue, 2011-02-15 at 23:30 +0100, Michał Piotrowski wrote:
Hi,
2011/2/14 Andre Robatino robatino@fedoraproject.org:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
Desktop:
https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test
Ideally, all Alpha priority test cases for installation [2] and desktop [3] should pass in order to meet the Alpha Release Criteria [4]. Help is available on #fedora-qa on irc.freenode.net [5], or on the test list [6].
Two issues here. On firstboot, when I tried to type my name I've got an exception. I think, that lettel "ł" caused it. Second issue - I don't have a "restart"
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
I'll check it later
option in Gnome menu.
That is a, ahem, policy decision.
What? 8-| What's next? Shutdown option?
-- Adam Williamson Fedora QA Community Monkey IRC: adamw | Fedora Talk: adamwill AT fedoraproject DOT org http://www.happyassassin.net
-- test mailing list test@lists.fedoraproject.org To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
On Wed, Feb 16, 2011 at 07:32:07AM +0100, Michał Piotrowski wrote:
2011/2/16 Adam Williamson awilliam@redhat.com:
That is a, ahem, policy decision.
What? 8-| What's next? Shutdown option?
<sarcasm> Excellent idea. Who needs to shutdown when you can just suspend/hibernate? </sarcasm>
Actually though, I highly recommend reading the GNOME 3 Design documents before accusing the GNOME team of not doing their homework.
http://gnome3.org/ http://gnome3.org/faq.html http://live.gnome.org/GnomeShell/Design/FAQ http://live.gnome.org/GnomeShell/Design/
2011/2/16 Chuck Anderson cra@wpi.edu:
On Wed, Feb 16, 2011 at 07:32:07AM +0100, Michał Piotrowski wrote:
2011/2/16 Adam Williamson awilliam@redhat.com:
That is a, ahem, policy decision.
What? 8-| What's next? Shutdown option?
<sarcasm> Excellent idea. Who needs to shutdown when you can just suspend/hibernate? </sarcasm>
Going forward Gnome developers can remove hibernate. If you can suspend, you don't need hibernate :) You don't even need suspend - Gnome 3 is so cool that users don't want to shutdown their computers :)
From my POV removing "restart" option because people uses it to
restart to Windows is one of the stupidest ideas in the universe :)
Actually though, I highly recommend reading the GNOME 3 Design documents before accusing the GNOME team of not doing their homework.
Sorry, but I do not have time for it. Gnome developers can do whatever they like with their software. But they should expect that it will meet with critical reactions from users.
Restart option was in Gnome menu since I can remember - and I remember the times of RH 6.0 - not RHEL 6.0 :).
http://gnome3.org/ http://gnome3.org/faq.html http://live.gnome.org/GnomeShell/Design/FAQ http://live.gnome.org/GnomeShell/Design/
-- test mailing list test@lists.fedoraproject.org To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
Michał Piotrowski wrote:
Sorry, but I do not have time for it. Gnome developers can do whatever they like with their software. But they should expect that it will meet with critical reactions from users.
Restart option was in Gnome menu since I can remember - and I remember the times of RH 6.0 - not RHEL 6.0:).
Known bug[1]. CC yourself.
W dniu 16 lutego 2011 16:43 użytkownik Michael Cronenworth mike@cchtml.com napisał:
Michał Piotrowski wrote:
Sorry, but I do not have time for it. Gnome developers can do whatever they like with their software. But they should expect that it will meet with critical reactions from users.
Restart option was in Gnome menu since I can remember - and I remember the times of RH 6.0 - not RHEL 6.0:).
Known bug[1]. CC yourself.
Thanks for the link
-- test mailing list test@lists.fedoraproject.org To unsubscribe: https://admin.fedoraproject.org/mailman/listinfo/test
On 02/16/2011 02:26 PM, Chuck Anderson wrote:
On Wed, Feb 16, 2011 at 07:32:07AM +0100, Michał Piotrowski wrote:
2011/2/16 Adam Williamsonawilliam@redhat.com:
That is a, ahem, policy decision.
What? 8-| What's next? Shutdown option?
<sarcasm> Excellent idea. Who needs to shutdown when you can just suspend/hibernate? </sarcasm>
Actually though, I highly recommend reading the GNOME 3 Design documents before accusing the GNOME team of not doing their homework.
http://gnome3.org/ http://gnome3.org/faq.html http://live.gnome.org/GnomeShell/Design/FAQ http://live.gnome.org/GnomeShell/Design
I think it would be more meaningful to provide links to their actual usability study/research then a links to their design/mockups and FAQ for novice end users.
I think people on this list are more wanting to understand why they are making those changing and less how they are making those changes.
As you can see the most criticism against the Gnome UI Design is removal of existing "features/usage" not about the look and feel about the design and usability of Gnome-shell ( all thou I've come across 1 really annoying usability issue which may or may not be a bug ).
Now I should point out that the reboot button has had a tendency to "disappear" early in the development cycle ( has occured for few cycles now ) and usually at the same/similar time the keyboard layout settings starts to default to US only then reboot button reappears and keyboard layout settings work again...
JBG
On Wed, Feb 16, 2011 at 03:45:30PM +0000, "Jóhann B. Guðmundsson" wrote:
On 02/16/2011 02:26 PM, Chuck Anderson wrote:
On Wed, Feb 16, 2011 at 07:32:07AM +0100, Michał Piotrowski wrote:
2011/2/16 Adam Williamsonawilliam@redhat.com:
That is a, ahem, policy decision.
What? 8-| What's next? Shutdown option?
<sarcasm> Excellent idea. Who needs to shutdown when you can just suspend/hibernate? </sarcasm>
Actually though, I highly recommend reading the GNOME 3 Design documents before accusing the GNOME team of not doing their homework.
http://gnome3.org/ http://gnome3.org/faq.html http://live.gnome.org/GnomeShell/Design/FAQ http://live.gnome.org/GnomeShell/Design
I think it would be more meaningful to provide links to their actual usability study/research then a links to their design/mockups and FAQ for novice end users.
How about this:
http://www.gnome.org/~mccann/shell/design/GNOME_Shell-20090705.pdf
On Wed, Feb 16, 2011 at 01:55:40PM -0500, Chuck Anderson wrote:
On Wed, Feb 16, 2011 at 03:45:30PM +0000, "Jóhann B. Guðmundsson" wrote:
I think it would be more meaningful to provide links to their actual usability study/research then a links to their design/mockups and FAQ for novice end users.
How about this:
http://www.gnome.org/~mccann/shell/design/GNOME_Shell-20090705.pdf
A quote from this document:
"Quality isn't job one. Being totally fucking amazing is job one."
As far as I can observe somebody mixed up both grammar and syntax and wrote "fucking amazing" in place of "fucked up".
On my 1600x1200 screen I am getting a totally unreadable mess with a bunch of crazy hieroglyphics (icons) with absolutely undecipherable titles. Where some letters can be distinguished they are obviously doubles. What is more there is no obvious way to try to fix this mess (yes, I guessed where some purported configurations may be). When I will show something like that to my wife, which has not the best vision, to put it mildly, and finds anything on dark backgrounds unacceptable under any circumstances, I will be laughed out of town. "Default" fonts are also out of question.
On the top of it my existing keyboard remapping was ignored and so far I did not find a way to remap "Ctrl" key from what it is stuck on current keyboard to a position where I _require_ it to be. Anything which does not allow that is by a definition a reject, as far as I am concerned, regardless what other redeeming features that may have. A plain show stopper.
Does not tells much more although pictures are tad better than a junk on my screen. One has to wholeheartedly agree that here "Quality isn't job one". Or two, or three for that matter.
Just in case somebody thinks that I am exaggerating attached is a picture with a fragment of a "new look" of my screen. Even if something could be done with a rendering if colours cannot be fixed then this is a total reject too.
Michal
On Wed, Feb 16, 2011 at 02:57:43PM -0700, Michal Jaegermann wrote:
On my 1600x1200 screen I am getting a totally unreadable mess with a bunch of crazy hieroglyphics (icons) with absolutely undecipherable titles. Where some letters can be distinguished they are obviously doubles. What is more there is no obvious way to try to fix this mess (yes, I guessed where some purported configurations may be).
The text in your screenshot doesn't look to me like a configuration problem. It looks like a bug, perhaps specific to your hardware, perhaps not. FWIW, on my screen, it's actually readable (running F15 as of this morning).
HTH,
Nalin
On Wed, Feb 16, 2011 at 05:06:25PM -0500, Nalin Dahyabhai wrote:
On Wed, Feb 16, 2011 at 02:57:43PM -0700, Michal Jaegermann wrote:
On my 1600x1200 screen I am getting a totally unreadable mess with a bunch of crazy hieroglyphics (icons) with absolutely undecipherable titles. Where some letters can be distinguished they are obviously doubles. What is more there is no obvious way to try to fix this mess (yes, I guessed where some purported configurations may be).
The text in your screenshot doesn't look to me like a configuration problem. It looks like a bug, perhaps specific to your hardware, perhaps not. FWIW, on my screen, it's actually readable (running F15 as of this morning).
You are extremely generous calling that "text".
The hardware is SyncMaster 213T LCD Samsung monitor, 1600x1200, driven by a DVI-0 digital output of a radeon card using kernel modesetting. With something other than gnome-shell this has a very nice and stable picture. I am afraid of even think how this may look on a lesser hardware.
Besides, as I already mentioned, color schemes and other settings are absolutely unacceptable (but at this moment at least they do not look like adjustable - I am afraid that this may be another "design decision").
Michal
On Wed, 2011-02-16 at 14:57 -0700, Michal Jaegermann wrote:
Just in case somebody thinks that I am exaggerating attached is a picture with a fragment of a "new look" of my screen. Even if something could be done with a rendering if colours cannot be fixed then this is a total reject too.
Well, I mean, that's obviously a bug. A bug. There's no need to go off on a huge rant as if that's how someone actually intended it to look, when it clearly isn't. It's just a bug. File it and it should get fixed.
On Wed, Feb 16, 2011 at 02:12:18PM -0800, Adam Williamson wrote:
On Wed, 2011-02-16 at 14:57 -0700, Michal Jaegermann wrote:
Just in case somebody thinks that I am exaggerating
Well, I mean, that's obviously a bug. A bug.
So which of the issues I was talking about is "a bug"? Or do you mean all of that together?
Michal
On Wed, 2011-02-16 at 18:02 -0700, Michal Jaegermann wrote:
On Wed, Feb 16, 2011 at 02:12:18PM -0800, Adam Williamson wrote:
On Wed, 2011-02-16 at 14:57 -0700, Michal Jaegermann wrote:
Just in case somebody thinks that I am exaggerating
Well, I mean, that's obviously a bug. A bug.
So which of the issues I was talking about is "a bug"? Or do you mean all of that together?
The corrupted text.
On Wed, Feb 16, 2011 at 06:59:26PM -0800, Adam Williamson wrote:
On Wed, 2011-02-16 at 18:02 -0700, Michal Jaegermann wrote:
On Wed, Feb 16, 2011 at 02:12:18PM -0800, Adam Williamson wrote:
Well, I mean, that's obviously a bug. A bug.
So which of the issues I was talking about is "a bug"? Or do you mean all of that together?
The corrupted text.
How about a keyboard and colour schemes? On the long run these are _much_ more important.
Michal
How about a keyboard and colour schemes? On the long run these are _much_ more important.
Yes, send in bug reports if you are having problems with them.
What you are testing is TC and not even Alpha so you should expect a lot of broken things. Total System Quality doesn't come straight from the developers themselves. It gets tested and sent back to the developers to fix. The testing part is what we're trying to do here. You should expect to download an image that once installed will probably break in at least 14 different ways. That ensures on release day an end user will download an image that is of high quality.
-- John Watzke
On Thu, Feb 17, 2011 at 3:49 PM, John Watzke watzkej@gmail.com wrote:
What you are testing is TC and not even Alpha so you should expect a lot of broken things. Total System Quality doesn't come straight from the developers themselves. It gets tested and sent back to the developers to fix. The testing part is what we're trying to do here. You should expect to download an image that once installed will probably break in at least 14 different ways. That ensures on release day an end user will download an image that is of high quality.
I am having real problems trying to get a gnome3 session in either tc2 or the last few nightlies on my test laptop - I can boot the live image (off a usbkey), but not to a desktop, and only to a VT. I can mess with /etc/gdm/custom.conf and stop prefdm.service and restart it but although it tries to start a session I get a brief flashing mutter crash box that stays on the gnome3 screen a fraction of a second - and then goes back to the background only - and this reappears briefly again every few seconds. I can kill the session by going to another VT - but no matter how I try to restart using systemctl stop/start prefdm.service, or even trying telinit 3/5 which presumably does the same -(and killing off custom.conf from gdm) nothing works - and although I have seen suggestions to switchuser I can't get this as I am not in a desktop, and the gdm greeter does not have that option - this is wireless so I can't save the log files as I am running off a live image from a liveusb key....
Any magic incantations anyone knows about to work around these issues for a live image and get to a gnome3 desktop (I had gnome3 running from the gnome3 testday image so I know gnome3 will work in this machine)
Thanks
On Thu, 2011-02-17 at 20:35 +0000, mike cloaked wrote:
On Thu, Feb 17, 2011 at 3:49 PM, John Watzke watzkej@gmail.com wrote:
What you are testing is TC and not even Alpha so you should expect a lot of broken things. Total System Quality doesn't come straight from the developers themselves. It gets tested and sent back to the developers to fix. The testing part is what we're trying to do here. You should expect to download an image that once installed will probably break in at least 14 different ways. That ensures on release day an end user will download an image that is of high quality.
I am having real problems trying to get a gnome3 session in either tc2 or the last few nightlies on my test laptop - I can boot the live image (off a usbkey), but not to a desktop, and only to a VT. I can mess with /etc/gdm/custom.conf and stop prefdm.service and restart it but although it tries to start a session I get a brief flashing mutter crash box that stays on the gnome3 screen a fraction of a second - and then goes back to the background only - and this reappears briefly again every few seconds. I can kill the session by going to another VT - but no matter how I try to restart using systemctl stop/start prefdm.service, or even trying telinit 3/5 which presumably does the same -(and killing off custom.conf from gdm) nothing works - and although I have seen suggestions to switchuser I can't get this as I am not in a desktop, and the gdm greeter does not have that option - this is wireless so I can't save the log files as I am running off a live image from a liveusb key....
Any magic incantations anyone knows about to work around these issues for a live image and get to a gnome3 desktop (I had gnome3 running from the gnome3 testday image so I know gnome3 will work in this machine)
Simplest test would be to boot to runlevel 3, log in as 'liveuser' (no password), and run startx. If that fails, it looks like that crasher in mutter is your problem and you should report that (it may turn out to be a graphics card issue rather than a mutter issue).
On Thu, Feb 17, 2011 at 8:51 PM, Adam Williamson awilliam@redhat.com wrote:
Simplest test would be to boot to runlevel 3, log in as 'liveuser' (no password), and run startx. If that fails, it looks like that crasher in mutter is your problem and you should report that (it may turn out to be a graphics card issue rather than a mutter issue).
Yes I have tried that and the mutter crasher is indeed the issue! However I can't get at the log files (I mean I can't save them as it is a live system) I can see loads of error messages on a VT and could take a photo - but it would be nice if I could save messages and/or the xorg log file and attach that to a bz.
On Thu, Feb 17, 2011 at 8:59 PM, mike cloaked mike.cloaked@gmail.com wrote:
On Thu, Feb 17, 2011 at 8:51 PM, Adam Williamson awilliam@redhat.com wrote:
Simplest test would be to boot to runlevel 3, log in as 'liveuser' (no password), and run startx. If that fails, it looks like that crasher in mutter is your problem and you should report that (it may turn out to be a graphics card issue rather than a mutter issue).
Yes I have tried that and the mutter crasher is indeed the issue! However I can't get at the log files (I mean I can't save them as it is a live system) I can see loads of error messages on a VT and could take a photo - but it would be nice if I could save messages and/or the xorg log file and attach that to a bz.
I have managed to plug it into a wire and get the messages file out via rsync to another machine on my LAN - I will bz it.
On Thu, Feb 17, 2011 at 9:12 PM, mike cloaked mike.cloaked@gmail.com wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=678418
On Thu, Feb 17, 2011 at 9:12 PM, mike cloaked mike.cloaked@gmail.com wrote:
Argh, just missed giving you the current bug number. This is the issue I reported earlier in this thread. It will probably be marked as a dup of https://bugzilla.redhat.com/show_bug.cgi?id=677842. Please fill in some of your experiences and hardware info in the bug to help debug the problem.
-- John Watzke
On Thu, 2011-02-17 at 21:12 +0000, mike cloaked wrote:
On Thu, Feb 17, 2011 at 8:59 PM, mike cloaked mike.cloaked@gmail.com wrote:
On Thu, Feb 17, 2011 at 8:51 PM, Adam Williamson awilliam@redhat.com wrote:
Simplest test would be to boot to runlevel 3, log in as 'liveuser' (no password), and run startx. If that fails, it looks like that crasher in mutter is your problem and you should report that (it may turn out to be a graphics card issue rather than a mutter issue).
Yes I have tried that and the mutter crasher is indeed the issue! However I can't get at the log files (I mean I can't save them as it is a live system) I can see loads of error messages on a VT and could take a photo - but it would be nice if I could save messages and/or the xorg log file and attach that to a bz.
I have managed to plug it into a wire and get the messages file out via rsync to another machine on my LAN - I will bz it.
Great. What I usually do is just plug in another USB key and write error logs out to that. You can mount a USB key just fine when booted live.
On Thu, Feb 17, 2011 at 9:24 PM, Adam Williamson awilliam@redhat.com wrote:
Great. What I usually do is just plug in another USB key and write error logs out to that. You can mount a USB key just fine when booted live.
True - I would have liked to write to the same mounted key running the live image but of course directories in it are mounted read only - there is a second usb slot though!
On Wed, 2011-02-16 at 17:23 +0100, Christoph Frieben wrote:
2011/2/16 Chuck Anderson:
Excellent idea. Who needs to shutdown when you can just suspend/hibernate?
Actually, you can **not** suspend: item "Suspend.." merely opens a "Shutdown" dialog. ~C
You can suspend by shutting the lid of your laptop, or by using the suspend key. The case of desktops with no suspend key and no lid wasn't sufficiently considered. This is another one that's under discussion, I can't find the bug, but the suspend option may come back. I think they're re-re-considering the whole 'shutdown management' design now.
W dniu 16 lutego 2011 07:32 użytkownik Michał Piotrowski mkkp4x4@gmail.com napisał:
2011/2/16 Adam Williamson awilliam@redhat.com:
[..]
We definitely need reports on such 'odd' (i.e. not US ASCII...) character set issues, i18n issues etc - can you double-check it and file the issue if it's reproducible? Thanks.
I'll check it later
https://bugzilla.redhat.com/show_bug.cgi?id=678070
https://bugzilla.redhat.com/show_bug.cgi?id=678075
On Mon, 14 Feb 2011 14:34:43 -0500, Andre wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
The x86_64 DVD.iso still segfaults for "askmethod" installation from harddisk.
Just as with the image for the GNOME test day, I'm unable to install this. Booting the vmlinuz/initrd.img from the isolinux enters text mode and wants to start installation from network. Using askmethod and pointing it at /dev/sda6 path /15/64 crashes.
:-(
On Wed, 2011-02-16 at 12:35 +0100, Michael Schwendt wrote:
On Mon, 14 Feb 2011 14:34:43 -0500, Andre wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
The x86_64 DVD.iso still segfaults for "askmethod" installation from harddisk.
Just as with the image for the GNOME test day, I'm unable to install this. Booting the vmlinuz/initrd.img from the isolinux enters text mode and wants to start installation from network. Using askmethod and pointing it at /dev/sda6 path /15/64 crashes.
Is there a bug filed for this segfault?
Also, it's possible you don't need to boot with askmethod. Can you explain in some detail your use case that needs 'askmethod'?
Thanks, James
On Wed, 16 Feb 2011 08:07:18 -0500, James wrote:
On Wed, 2011-02-16 at 12:35 +0100, Michael Schwendt wrote:
On Mon, 14 Feb 2011 14:34:43 -0500, Andre wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
The x86_64 DVD.iso still segfaults for "askmethod" installation from harddisk.
Just as with the image for the GNOME test day, I'm unable to install this. Booting the vmlinuz/initrd.img from the isolinux enters text mode and wants to start installation from network. Using askmethod and pointing it at /dev/sda6 path /15/64 crashes.
Is there a bug filed for this segfault?
No. I won't file a bug for something that is either not supported in the test images or known to be broken. When I brought this up for the GNOME test day images, it was said that the images were never meant to be used as install images and that most testers would use them for creating live media.
The linked Wiki pages are like a maze where I go in circles trying to find info about how the DVD ISO image may be used.
Also, it's possible you don't need to boot with askmethod. Can you explain in some detail your use case that needs 'askmethod'?
Without askmethod I'm not asked whether to perform an install from harddisk. Instead, the installer asks me to select the network card which to use. With the GNOME test day image, it actually proceeded to downloading all packages from the network.
On Wed, 2011-02-16 at 15:17 +0100, Michael Schwendt wrote:
On Wed, 16 Feb 2011 08:07:18 -0500, James wrote:
On Wed, 2011-02-16 at 12:35 +0100, Michael Schwendt wrote:
On Mon, 14 Feb 2011 14:34:43 -0500, Andre wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Installation:
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
The x86_64 DVD.iso still segfaults for "askmethod" installation from harddisk.
Just as with the image for the GNOME test day, I'm unable to install this. Booting the vmlinuz/initrd.img from the isolinux enters text mode and wants to start installation from network. Using askmethod and pointing it at /dev/sda6 path /15/64 crashes.
Is there a bug filed for this segfault?
No. I won't file a bug for something that is either not supported in the test images or known to be broken.
If you add "askmethod" and it segfaults/crashes ... that's a bug. I'd like to see this issue fixed. I don't believe "askmethod" is going away, but I'm still trying to understand how askmethod, repo= and stage2= will change now that we no longer have a stage2 install.img file.
In the meantime, can you provide some more information on the segfault/crash? Even better, please file a bug with the crash information. This is one that needs to be resolved.
When I brought this up for the GNOME test day images, it was said that the images were never meant to be used as install images and that most testers would use them for creating live media.
That's news to me, live images are, and will be, a supported means of installing Fedora. In fact, F15Alpha won't be released until they work ... https://bugzilla.redhat.com/show_bug.cgi?id=672265 (barring an unlikely decision from FESCO to release without them).
The linked Wiki pages are like a maze where I go in circles trying to find info about how the DVD ISO image may be used.
That's likely because this is still before the Alpha. Adjusting the documentation to match reality is something we'll need to get lined up for the installation guide. Which pages were confusing?
Also, it's possible you don't need to boot with askmethod. Can you explain in some detail your use case that needs 'askmethod'?
Without askmethod I'm not asked whether to perform an install from harddisk.
Personally, I'm still understanding how the install.img removal will impact our testing. I believe Hurry may have reached out on this topic as well, to determine how this impacts the 'installation source' tests [1].
One item to note, is that loader is *not* responsible for determining where the install.img is ... there is no install.img ... only packages. Directing the installer to different package repos, happens later in the install during repo setup. That said, some of the supported package repositories may require some preparation, or don't have a corresponding graphical setup in the installer. For those scenarios, the existing 'repo=' and 'askmethod' boot arguments [2] may be used.
[1] https://fedoraproject.org/wiki/Category:Installation_Source [2] https://fedoraproject.org/wiki/Anaconda_Boot_Options
Instead, the installer asks me to select the network card which to use.
https://bugzilla.redhat.com/show_bug.cgi?id=677773 ?
With the GNOME test day image, it actually proceeded to downloading all packages from the network.
Thanks, James
On Wed, 16 Feb 2011 11:30:31 -0500, James wrote:
If you add "askmethod" and it segfaults/crashes ... that's a bug. I'd like to see this issue fixed. I don't believe "askmethod" is going away,
In the meantime, can you provide some more information on the segfault/crash? Even better, please file a bug with the crash information. This is one that needs to be resolved.
https://bugzilla.redhat.com/678086
The linked Wiki pages are like a maze where I go in circles trying to find info about how the DVD ISO image may be used.
That's likely because this is still before the Alpha. Adjusting the documentation to match reality is something we'll need to get lined up for the installation guide. Which pages were confusing?
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
already is intimidating. Even more so with the growing "References" section at the bottom and when uncollapsing the tables. Those sections could really be put on its own page, so the primary page linked in the TC2 announcement would be clear and concise.
Instead, the installer asks me to select the network card which to use.
Dunno. It asks me *very* early in text mode already => https://bugzilla.redhat.com/678087
On Wed, 2011-02-16 at 18:21 +0100, Michael Schwendt wrote:
That's likely because this is still before the Alpha. Adjusting the documentation to match reality is something we'll need to get lined up for the installation guide. Which pages were confusing?
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
already is intimidating. Even more so with the growing "References" section at the bottom and when uncollapsing the tables. Those sections could really be put on its own page, so the primary page linked in the TC2 announcement would be clear and concise.
The test results are the whole point of the page. Splitting each section into a separate page would be possible, I guess, but would just make things more unwieldy for a pretty small benefit (make each page smaller).
On Wed, 2011-02-16 at 18:21 +0100, Michael Schwendt wrote:
On Wed, 16 Feb 2011 11:30:31 -0500, James wrote:
If you add "askmethod" and it segfaults/crashes ... that's a bug. I'd like to see this issue fixed. I don't believe "askmethod" is going away,
In the meantime, can you provide some more information on the segfault/crash? Even better, please file a bug with the crash information. This is one that needs to be resolved.
https://bugzilla.redhat.com/678086
The linked Wiki pages are like a maze where I go in circles trying to find info about how the DVD ISO image may be used.
That's likely because this is still before the Alpha. Adjusting the documentation to match reality is something we'll need to get lined up for the installation guide. Which pages were confusing?
https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test
already is intimidating. Even more so with the growing "References" section at the bottom and when uncollapsing the tables.
Thanks for the feedback. Testing the installer, among many components, can be complicated and involve a lot of different scenarios (not all of which we articulate on the wiki). Conveying the plan through the wiki presents a challenge. The method currently used has matured over several releases based on feedback from participants, but we know it's not perfect. If you have ideas for improvement ... feel free to copy'n'paste content into your personal space (e.g. User:Mschwendt/draft) we can play around with ideas.
The intended audience for that page is not the casual tester. It's aimed at those who are actively involved in contributing test feedback for the installer. And as you point out, relies on some experience modifying wiki. Hurry is working on something that we can use instead of the wiki to manage that workflow (https://fedorahosted.org/fedora-qa/ticket/152). But in the meantime, we do our best to monitor install test runs on the wiki.
The "References" section intentionally expands as we do more testing. While we aren't dead-set on this approach, it was something we tried in F-14 to eliminate listing duplicate issues on the page. The References section grows as test results with bugs are posted.
The collapsed sections can default to expanded, I don't recall at this time why we chose to intentionally collapse them by default.
Another way you can contribute is by providing bugs, or feedback on the list. I think the order listed would also be my preference in terms of order.
Those sections could really be put on its own page, so the primary page linked in the TC2 announcement would be clear and concise.
We used to do it that way ... and if you can believe it, it was even harder for us to track what the heck is going on, and to direct install-test contributors to a consistent location. But again, ideas/suggestions/wiki_patches welcome!
Instead, the installer asks me to select the network card which to use.
Dunno. It asks me *very* early in text mode already => https://bugzilla.redhat.com/678087
That's a little different, I'll stay tuned to that bug and we can adjust our test cases as needed based on any behavior changes.
Thanks, James
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Where would I report the following? (which bugzilla component?)
Keyboard is still not working for me in X. With TC2 I've tried no further than to enter something into the graphical login screen from GNOME. It doesn't work. Switching to virtual consoles doesn't work either. For GNOME Test Day I confirmed that they keyboard doesn't work in XFCE either.
Today's X log is this: http://mschwendt.fedorapeople.org/Xorg.0.log.20110216
On Wed, 2011-02-16 at 22:41 +0100, Michael Schwendt wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Where would I report the following? (which bugzilla component?)
Keyboard is still not working for me in X. With TC2 I've tried no further than to enter something into the graphical login screen from GNOME. It doesn't work. Switching to virtual consoles doesn't work either. For GNOME Test Day I confirmed that they keyboard doesn't work in XFCE either.
Today's X log is this: http://mschwendt.fedorapeople.org/Xorg.0.log.20110216
Does it work if you boot 'runlevel 3'?
On Wed, 16 Feb 2011 13:53:23 -0800, Adam wrote:
On Wed, 2011-02-16 at 22:41 +0100, Michael Schwendt wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Where would I report the following? (which bugzilla component?)
Keyboard is still not working for me in X. With TC2 I've tried no further than to enter something into the graphical login screen from GNOME. It doesn't work. Switching to virtual consoles doesn't work either. For GNOME Test Day I confirmed that they keyboard doesn't work in XFCE either.
Today's X log is this: http://mschwendt.fedorapeople.org/Xorg.0.log.20110216
Does it work if you boot 'runlevel 3'?
Yes.
And contrary to test-day, I can startx manually and keep a working keyboard inside GNOME. Available updates haven't changed anything. In gdm, keyboard isn't recognized.
On Wed, 2011-02-16 at 23:28 +0100, Michael Schwendt wrote:
On Wed, 16 Feb 2011 13:53:23 -0800, Adam wrote:
On Wed, 2011-02-16 at 22:41 +0100, Michael Schwendt wrote:
Fedora 15 Alpha TC2 is now available [1]. Please refer to the following pages for download links and testing instructions.
Where would I report the following? (which bugzilla component?)
Keyboard is still not working for me in X. With TC2 I've tried no further than to enter something into the graphical login screen from GNOME. It doesn't work. Switching to virtual consoles doesn't work either. For GNOME Test Day I confirmed that they keyboard doesn't work in XFCE either.
Today's X log is this: http://mschwendt.fedorapeople.org/Xorg.0.log.20110216
Does it work if you boot 'runlevel 3'?
Yes.
And contrary to test-day, I can startx manually and keep a working keyboard inside GNOME. Available updates haven't changed anything. In gdm, keyboard isn't recognized.
Remove the variant line from /etc/sysconfig/keyboard and reboot.