I doubt this has anything to do with bugs regarding multi desktop. The same issue exists on all platforms with or without multiple desktops. You can't have two firefox instances running at the same time with the same firefox profile, which is why the have a locking mechanism so they can pop up that dialog.
Did you really use File->New Window for this? Or, did you try and start a new process? Starting a new process will fail, opening a new window from the existing process will succeed. The question is, do popups from the 2nd window (now on a different desktop) open in the correct desktop.
Don't use it on all ISOs without first checking if it is required. For example, AFAIK recent Fedora and Mandriva ship hybrid ISOs, using unetbootin is both unnecessary (dd is sufficient to transfer to the ISO to a usb stick such that it will boot, Mandriva provides a GUI tool for Windows and Linux for those who forget to dd to the entire block device, not the partition...) and harmful (if you run unetbootin, it will break the feature, and *not* boot from the USB stick).
If you mean for the pure same (or more) functionality (but not necessarily concerned with a GUI), lsof has been around for ever.
Although to get the exact functionality (choose a process, find what files it has open) you need to run 'ps' to get the pid of interest, then lsof -p $pid (which may be a little bit more effort), the advantage of lsof is if you don't know exactly what file, or exactly what process, because then you can pipe the output for all processes to grep/awk/sed/perl to filter on a user, file, specific block device etc. (using patterns/regular expressions) or script killing the potentially multiple process with open files in a filesystem you have to take offline.
lsof shows the current working directory, the binary itself (useful in case a process has changed it's process label), the shared libraries, any open files (which includes real files, stdin/stdout/stderr, and open sockets).
If you don't mean exactly the same functionality, but which could be used for some similar problems, see the Linux Trace Toolkit - Next Generation, which provides a similar GUI for a slightly different purpose.
IBM might be a big US company but it is hardly focused "almost entirely" on open source. Pretty much all of their hardware is proprietary,
As much as Sun's.
their own Unix is proprietary,
But you have a choice of different OS's supported on their proprietary hardware, unlike Sun. This includes Linux, for which IBM has made the contributions for the hardware-specific features.
they sell all sorts of proprietary tools like Lotus Notes, ClearCase and the rest of the Rational tools, etc.
Sun sells many proprietary products.
Sure you can point out a number of projects and things they work on and support that are open source, but those hardly make up even a significant portion of their portfolio.
The same can be said for Sun. Of course, Oracle also makes contributions (check the stats for recent kernels, Oracle is normally somewhere around number 3-5 corporate contributor, behind Red Hat, Novell, IBM.
Red Hat Enterprise Virtualization isn't even funny. It's $750/socket for 24x7 phone support, per year. Hyper-V is free. Xen is free. KVM is free. ESXi is free. If you don't want Hyper-V without support, buy Windows Server Standard, it's $350 a year for the same support, and it's per server. You don't get dinged by the processor. If you're going to pay through the nose, you better be getting VMWare because at least they have the crazy advanced feature-set to back it up.
I think you've misunderstood RH's virtualisation strategy, the RHEV product *is* the competitor to VMWare's "crazy advanced feature-set" (competing with their vCenter product set), but (AFAIK) there is no "cheap" VMWare solution that provides HA, whereas RHEL Advanced Platform (without RHEV) does.
On RHEL standard, you're limited to 4 supported free Red Hat VMs, and 2 sockets on the host, starting at $349/year.
Because the solution is as simple as changing the default policy. Make it so that the default behavior is to notify only.
No, the solution is to not give blanket sudo access to the first user, but instead allow specific trusted applications, possibly running in a restricted mode (e.g. rurpmi on Mandriva), or to run elevated commands via a daemon running as root with specific authorization (the PolicyKit model).
This is why one reason I don't recommend Ubuntu.
(and this specific problem is not a "Linux" problem, it's almost exclusively an Ubuntu problem)
To have true high-availability, even 2 VMware servers isn't enough, you need a reliable shared storage system that both servers can access.
Even then, the storage chassis itself will be a central point of failure.
With Linux, DRBD, GFS and either KVM or Xen, you don't need shared storage, as DRBD does the replication for you between physical nodes, GFS does the "VMFS"-type concurrently accessible filesystem, and you get live migration free.
To have true HA you need a pair of independent shared storage units with continuous synchronous replication and some reliable mechanism of failover.
If you're looking at that level, most decent storage arrays have redundant controllers, you shouldn't need a second array for HA, mainly for DR (where D in DR stands for disaster, the kind where nothing in the vicinity of the first array works).
If the administration 'team' has equal access to all the services today on disparate servers, I don't think virtualization is necessarily a good idea, the services can be consolodated in a single OS instance.
Even if they all can run on the same OS instance, do you really want a large database query killing your DNS recursion. If they were separate VMs, then memory pressure on the database VM wouldn't impact the DNS VM (or, as much).
If you require different OSs for other reasons (e.g., some Windows, some Unix) then virtualisation is a requirement if you want to reduce box count.
In terms of HA, put two relatively low end boxes in each branch (you said 7 year old servers were fine, so high end is overkill). Read up on linux HA which is free, and use DRBD to get total redundancy in your storage as well as a cheap software mirror or raid 5. Some may rightfully question the need for HA, but this approach is pretty dirt cheap at low scale.
1)Just install CentOS, or the distribution of your choice that ships Red Hat Cluster and a suitable hypervisor 2)Install DRBD, cluster, and configure GFS on top of DRBD for storage of VM base images and VM configuration files 3)Choose Xen or KVM as hypervisor 4)Install VMs (Windows, Linux etc.) using the virt-manager GUI tool 5)See that you can now migrate VMs between physical servers without service interruption, and VM recovery can occur in seconds (if a physical server failed). CentOS probably won't have it quite yet, but Xen can now do real-time state replication, so in future even unplanned downtime on a physical machine will be without impact
If you can fit it in your budget (which you should be able to, having spent nothing on virtualisation software), buy decent servers which have remote management cards (e.g. HP iLO, Dell DRAC, Sun ILOM). Not only is it convenient (e.g. being able to boot into recovery remotely if you ever need it), but cluster operation will be more reliable if you use these for fencing.
While this may be a bit more complex than typical "Linux HA", the benefits are worth it. In an environment I was involved in until recently, we had a virtualisation cluster running VM pairs which were clustered. In the past 6 months, the virtualisation layer (including GFS, cluster on the physicals, Xen etc.) has not failed, while the clustered service running on the VMs has numerous times. The most likely action that will be taken to fix this is to remove the clustering between VMs, to rely almost exclusively on virtualisation for HA.
This might not be "Best Practice", but it can provide best of breed and bang for buck for a small investment of time, which can be recovered for the next site.
Of course, which is why you can't log in as root in any display manager in Mandriva. However, you can log in as root on the console, or su to root. However, the more insecure "blanket sudo to root to the first user" is *not* present by default. What *is* present however in some cases, are "restricted" versions of some tools, e.g. the rurpmi version of urpmi, which restricts what you can do with packages, specifically so you can safely give users access to installing software themselves. Most GUI configuration tools are handled by consolehelper (in 2010.0 this might now be the *kit replacement).
Oh, and the reason why I left Mandrake for SuSE (and later Ubuntu) in the first place: RPM Hell. Mandrake is only slightly better than SuSE (and moreso than Fedora) at this, but I don't want to fight with my package manager because of bizarre internal consistency issues that prevent upgrading packages, or adding third-party software. I've never had this problem on Ubuntu or Debian or Elive or any other.deb-based distribution.
And I haven't had this on Mandrake/Mandriva since 7.0, which was the first release to ship urpmi, before apt was in a stable release of Debian.
Ubuntu also has a method of installing patent-encumbered software, and it's built-in, not an external repository
So, Ubuntu is violating DMCA?
Even if you don't do that, any time you try to play an MP3 or WMV, etc, it will ask you if you want to download and install the codec required, on the spot, just warning you that it's patent-encumbered.
MP3 playback is available out-the-box, the codeina thingy is also available by default.
How is that not *better* than your solution in Mandriva?
Anyway, this is not a technical decision, Mandriva has a policy of not being involved in the distribution of any patent-infringing software. If software patents disappeared, it would take a few hours to get the packages into Mandriva.
it means "both architectures, but just a base system with no software whatsoever, although the installer pretends to have it
Which it would, as the intention is to use it for a network-assisted install. It is mainly intended for people who are familiar with the distro, so it isn't even listed on the comparison page.
The ease-of-use / great desktop mantle long ago moved from Mandriva to PCLinuxOS, essentially a fork of Mandriva.
PCLinuxOS primarily as a Mandrake release (except that the SRPMS were rebuilt with the only change being the removal of the changelog) with all the DMCA-violating software from PLF included, a different theme, and shipping apt-rpm by default.
Since these days the default themes are actually quite good, proprietary drivers (which were always available on the Mandrake Powerpack releases) have been available on the mirrors in the non-free repository for the past few releases, and apt-rpm is abandoned upstream, what does PCLinuxOS actually bring to the table?
If there's something else missing from Mandriva, why don't the PCLinuxOS people contribute to Mandriva? Contributors are welcome (and have been for years).
My main objection is simply that it doesn't have a vanilla (x)nethack package
If you file a bug with component "Package request", it is possible that this objection could be removed... nethack_falconseye is available in contrib though, so you might want to motivate (what the differences/features/benefits are).
kde 4 really kicked mandrivas usability... I currently use 2009 Spring and kde 4.3 is a big improvement over older kde 4 versions, but quite often I regret switching from 2008 Spring. many features, that worked in 2008 spring are now broken
akregator and kmail now have problems with some servers
I've been using kmail quite a bit, and haven't had problems. I don't use akregator much...
kile and kate's scripting feature don't work anymore
I think it should be back in KDE 4.4, but this is of course an upstream issue.
kaffeine can't handle non-square pixels anymore, so DVD playback is stretched on my 16:9 TV - and my bugreports are just ignored)
i get errors from PulseAudio all the time
dragon player is working quite well for me on KDE 4.2 on Mandriva 2009.1. The only thing I am missing in dragon is a decent playlist.
I cant mount encrypted harddrives at boot-time, not even with initscripts or using crypttab (i have to mount them manually after booting
If this is your bug, it may have workarounds for 2009.1, and is fixed in 2010.0 by the switch to plymouth (splashy was the cause in 2009.0 and 2009.1). If you have a different bug, you need to provide means to reproduce it...
the one thing that's really improved is kdenlive)
I tried to install Mandriva 2010, but aparently its installer doesn't think my SSD is a harddrive... although all previous mandriva versions installed on it just fine... maybe I'll switch the ports where my harddrives are plugged in - that may change something, but then again i'll have to reinstall grub manually (mandrivas bootloader repair tool never worked for me)
I didn't try 2010.0 on my Acer Aspire One, so I can't comment here, but I didn't see any bugs filed on this.
mandriva 2009 was completely unusable with kde 4.1...
Which is why KDE3 was still available for it, unlike other distributions that were released at the same time.
Ubuntu does XFS, (as well as ext*, JFS, MurderFS and so on) through the standard installer.
XFS has been available on Mandrake/Mandriva since Mandrake 8.2 if I remember correctly. Since that time it has been possible for users to resize system filesystems (e.g./usr) using a graphical interface. This is still not possible on many distributions.
mdraid, lvm and truecrypt only work through the alternate installer disc (but the curses interface ain't that much more difficult than the GUI, so it oughtn't be an issue.
The Mandriva installer supports RAID, LVM, and LUKS encryption in the graphical installer. This GUI tool is also available after installation.
Someone should post a story everytime they emerge --sync && emerge world
No, otherwise people running Mandriva cooker should post every time they 'urpmi --auto-update', or people running Debian testing should every time they run 'apt-get upgrade', or users on Fedora rawhide every time they run 'yum update' (ok, for Fedora, maybe not *every* time...).
Just because you compiled it, doesn't mean you got it sooner thananyoneelse. (Note, the Mandriva build system is currently not accepting build submissions for "cooker" as cooker is still in freeze, the build system will be back to the usual 50+ packages per day by the end of the week).
As much as Debian died by dpkg, in other words, not at all. I guess you didn't try urpmi (which was in a released version of Mandriva before apt was in a stable release of Debian)?
Mandriva's not even run by the guy that founded Mandrake. So everyone that remembers the old Mandrake should remember that this is just somebody else with sorta the same name doing the distro now.
So, when no more founders of Microsoft are employed by Microsoft, they should change their name, or their customers should consider switching?
What really made Mandrake, and continues to make Mandriva, is not one person, but the combination of employees and contributors. While many of both have come and gone, a lot of the contributors from the Mandrake era still use and contribute to the distro, and new contributors join quite often.
If you bothered to look, you would probably find that Mandriva is more open than Ubuntu or Fedora (not sure about "Open"SUSE).
Which model? Some models e.g. F4?) only have a very limited (WIndows only) software set, which only upload the data to the Polar web service (https://www.polarpersonaltrainer.com/). I don't remember any way to get the data out.
There are open-source utilities to download the data, but I have not had much success with them, and so could not use PolarViewer etc. with them.
I guess the more expensive models allow use from other software.
In a kerberized environment, kdestroy (or click on the "Remove Credentials Cache" option on the krb5-auth-dialog applet) is enough to ensure that all your access will not work until you re-authenticate. Remote shell sessions, web sites, email etc.
Now, if KDE had an option to kdestroy on screen saver lock, and if it correctly got tickets on unlock, it would be a lot more usable./me logs some bugs...
It's been mentioned elsewhere (but not here as far as I can tell) that this development is particularly notable, given that Windows doesn't support Exchange out of the box.
But, most linux distributions have for the past 6 months (thanks to evolution-mapi). And that's not even counting the previous support (based on the OWA API) that was available via evolution-exchange... and neither requires you upgrade to Exchange 2007 (but evolution-mapi supports Exchange 2007, where the old OWA API is no longer available).
I did think of going back to mandriva, after gentoo - but that was right after the mandriva 9.2 'cd drive killer' kernel bug. OK, it turned out to be LG's fault in the end, but it did rather put me off!
Ironically, Gentoo had killed a lot of drives with the same patch months before (in their live Gaming CD), but didn't manage to figure out the cause, and left lots of toasted drives behind.
In fact, it was Mandrake (at the time) that tracked down this bug, got a resolution from the hardware vendor to restore drives to a working state, and ensure that the packet writing patch could get into mainline with some testing.
but then they went to KDE4 far to early with MDV 2009.0
As opposed to many other distros that shipped KDE 4.0 as the only option 6 months earlier, where 2009.0 shipped 4.1 by default, but still had 3.5 available?
I couldn't even re-find solutions to problems I'd previously had solved (e.g. an odd xorg.conf for my monitor). Its a lot better today, but there is still an odd gap between the main site [mandriva.com] and the place all the really useful [mandriva.com] information is kept.
Well, you should have been able to find a bug report... and these days the forum is open to anyone, or, what do you think should fill the gap ?
Slashdot Mirror
I doubt this has anything to do with bugs regarding multi desktop. The same issue exists on all platforms with or without multiple desktops. You can't have two firefox instances running at the same time with the same firefox profile, which is why the have a locking mechanism so they can pop up that dialog.
Did you really use File->New Window for this? Or, did you try and start a new process? Starting a new process will fail, opening a new window from the existing process will succeed. The question is, do popups from the 2nd window (now on a different desktop) open in the correct desktop.
Don't use it on all ISOs without first checking if it is required. For example, AFAIK recent Fedora and Mandriva ship hybrid ISOs, using unetbootin is both unnecessary (dd is sufficient to transfer to the ISO to a usb stick such that it will boot, Mandriva provides a GUI tool for Windows and Linux for those who forget to dd to the entire block device, not the partition ...) and harmful (if you run unetbootin, it will break the feature, and *not* boot from the USB stick).
So it turns out that there is a GUI for lsof.
I've never needed a GUI for lsof, but I guess some people would say it is required if there is a similar GUI tool for Windows ...
If you mean for the pure same (or more) functionality (but not necessarily concerned with a GUI), lsof has been around for ever.
Although to get the exact functionality (choose a process, find what files it has open) you need to run 'ps' to get the pid of interest, then lsof -p $pid (which may be a little bit more effort), the advantage of lsof is if you don't know exactly what file, or exactly what process, because then you can pipe the output for all processes to grep/awk/sed/perl to filter on a user, file, specific block device etc. (using patterns/regular expressions) or script killing the potentially multiple process with open files in a filesystem you have to take offline.
lsof shows the current working directory, the binary itself (useful in case a process has changed it's process label), the shared libraries, any open files (which includes real files, stdin/stdout/stderr, and open sockets).
If you don't mean exactly the same functionality, but which could be used for some similar problems, see the Linux Trace Toolkit - Next Generation, which provides a similar GUI for a slightly different purpose.
IBM might be a big US company but it is hardly focused "almost entirely" on open source. Pretty much all of their hardware is proprietary,
As much as Sun's.
their own Unix is proprietary,
But you have a choice of different OS's supported on their proprietary hardware, unlike Sun. This includes Linux, for which IBM has made the contributions for the hardware-specific features.
they sell all sorts of proprietary tools like Lotus Notes, ClearCase and the rest of the Rational tools, etc.
Sun sells many proprietary products.
Sure you can point out a number of projects and things they work on and support that are open source, but those hardly make up even a significant portion of their portfolio.
The same can be said for Sun. Of course, Oracle also makes contributions (check the stats for recent kernels, Oracle is normally somewhere around number 3-5 corporate contributor, behind Red Hat, Novell, IBM.
Red Hat Enterprise Virtualization isn't even funny. It's $750/socket for 24x7 phone support, per year. Hyper-V is free. Xen is free. KVM is free. ESXi is free. If you don't want Hyper-V without support, buy Windows Server Standard, it's $350 a year for the same support, and it's per server. You don't get dinged by the processor. If you're going to pay through the nose, you better be getting VMWare because at least they have the crazy advanced feature-set to back it up.
But, Red Hat Advanced Platform (starting at $1499/year, with unlimited sockets/memory includes clustering, GFS (to support live VM migration) and unlimited supported Red Hat guests, which inherit the base features (clustering and GFS). You don't get the RHEV management platform, but the server supports being managed by the RHEV management platform (which is a separate "per managed socket subscription of $500).
I think you've misunderstood RH's virtualisation strategy, the RHEV product *is* the competitor to VMWare's "crazy advanced feature-set" (competing with their vCenter product set), but (AFAIK) there is no "cheap" VMWare solution that provides HA, whereas RHEL Advanced Platform (without RHEV) does.
On RHEL standard, you're limited to 4 supported free Red Hat VMs, and 2 sockets on the host, starting at $349/year.
If Linux malware is unheard of, why does McAfee sell LinuxSheld?
The question isn't why they sell it, but why customers buy it, and that is most likely for "Benefit" 4:
LinuxShield protects Microsoft Windows systems by blocking Microsoft Windows viruses from passing through the Linux environment
Because the solution is as simple as changing the default policy. Make it so that the default behavior is to notify only.
No, the solution is to not give blanket sudo access to the first user, but instead allow specific trusted applications, possibly running in a restricted mode (e.g. rurpmi on Mandriva), or to run elevated commands via a daemon running as root with specific authorization (the PolicyKit model).
This is why one reason I don't recommend Ubuntu.
(and this specific problem is not a "Linux" problem, it's almost exclusively an Ubuntu problem)
To have true high-availability, even 2 VMware servers isn't enough, you need a reliable shared storage system that both servers can access.
Even then, the storage chassis itself will be a central point of failure.
With Linux, DRBD, GFS and either KVM or Xen, you don't need shared storage, as DRBD does the replication for you between physical nodes, GFS does the "VMFS"-type concurrently accessible filesystem, and you get live migration free.
To have true HA you need a pair of independent shared storage units with continuous synchronous replication and some reliable mechanism of failover.
If you're looking at that level, most decent storage arrays have redundant controllers, you shouldn't need a second array for HA, mainly for DR (where D in DR stands for disaster, the kind where nothing in the vicinity of the first array works).
If the administration 'team' has equal access to all the services today on disparate servers, I don't think virtualization is necessarily a good idea, the services can be consolodated in a single OS instance.
Even if they all can run on the same OS instance, do you really want a large database query killing your DNS recursion. If they were separate VMs, then memory pressure on the database VM wouldn't impact the DNS VM (or, as much).
If you require different OSs for other reasons (e.g., some Windows, some Unix) then virtualisation is a requirement if you want to reduce box count.
In terms of HA, put two relatively low end boxes in each branch (you said 7 year old servers were fine, so high end is overkill). Read up on linux HA which is free, and use DRBD to get total redundancy in your storage as well as a cheap software mirror or raid 5. Some may rightfully question the need for HA, but this approach is pretty dirt cheap at low scale.
1)Just install CentOS, or the distribution of your choice that ships Red Hat Cluster and a suitable hypervisor
2)Install DRBD, cluster, and configure GFS on top of DRBD for storage of VM base images and VM configuration files
3)Choose Xen or KVM as hypervisor
4)Install VMs (Windows, Linux etc.) using the virt-manager GUI tool
5)See that you can now migrate VMs between physical servers without service interruption, and VM recovery can occur in seconds (if a physical server failed). CentOS probably won't have it quite yet, but Xen can now do real-time state replication, so in future even unplanned downtime on a physical machine will be without impact
If you can fit it in your budget (which you should be able to, having spent nothing on virtualisation software), buy decent servers which have remote management cards (e.g. HP iLO, Dell DRAC, Sun ILOM). Not only is it convenient (e.g. being able to boot into recovery remotely if you ever need it), but cluster operation will be more reliable if you use these for fencing.
While this may be a bit more complex than typical "Linux HA", the benefits are worth it. In an environment I was involved in until recently, we had a virtualisation cluster running VM pairs which were clustered. In the past 6 months, the virtualisation layer (including GFS, cluster on the physicals, Xen etc.) has not failed, while the clustered service running on the VMs has numerous times. The most likely action that will be taken to fix this is to remove the clustering between VMs, to rely almost exclusively on virtualisation for HA.
This might not be "Best Practice", but it can provide best of breed and bang for buck for a small investment of time, which can be recovered for the next site.
Legal requirement.
A full GUI root account is a terrible idea
Of course, which is why you can't log in as root in any display manager in Mandriva. However, you can log in as root on the console, or su to root. However, the more insecure "blanket sudo to root to the first user" is *not* present by default. What *is* present however in some cases, are "restricted" versions of some tools, e.g. the rurpmi version of urpmi, which restricts what you can do with packages, specifically so you can safely give users access to installing software themselves. Most GUI configuration tools are handled by consolehelper (in 2010.0 this might now be the *kit replacement).
Oh, and the reason why I left Mandrake for SuSE (and later Ubuntu) in the first place: RPM Hell. Mandrake is only slightly better than SuSE (and moreso than Fedora) at this, but I don't want to fight with my package manager because of bizarre internal consistency issues that prevent upgrading packages, or adding third-party software. I've never had this problem on Ubuntu or Debian or Elive or any other .deb-based distribution.
And I haven't had this on Mandrake/Mandriva since 7.0, which was the first release to ship urpmi, before apt was in a stable release of Debian.
Ubuntu also has a method of installing patent-encumbered software, and it's built-in, not an external repository
So, Ubuntu is violating DMCA?
Even if you don't do that, any time you try to play an MP3 or WMV, etc, it will ask you if you want to download and install the codec required, on the spot, just warning you that it's patent-encumbered.
MP3 playback is available out-the-box, the codeina thingy is also available by default.
How is that not *better* than your solution in Mandriva?
Can you play AMR audio?
Anyway, this is not a technical decision, Mandriva has a policy of not being involved in the distribution of any patent-infringing software. If software patents disappeared, it would take a few hours to get the packages into Mandriva.
it means "both architectures, but just a base system with no software whatsoever, although the installer pretends to have it
Which it would, as the intention is to use it for a network-assisted install. It is mainly intended for people who are familiar with the distro, so it isn't even listed on the comparison page.
Either way, running Mandriva Control Center->Software Management etc. should allow you to quite easily come right.
The ease-of-use / great desktop mantle long ago moved from Mandriva to PCLinuxOS, essentially a fork of Mandriva.
PCLinuxOS primarily as a Mandrake release (except that the SRPMS were rebuilt with the only change being the removal of the changelog) with all the DMCA-violating software from PLF included, a different theme, and shipping apt-rpm by default.
Since these days the default themes are actually quite good, proprietary drivers (which were always available on the Mandrake Powerpack releases) have been available on the mirrors in the non-free repository for the past few releases, and apt-rpm is abandoned upstream, what does PCLinuxOS actually bring to the table?
If there's something else missing from Mandriva, why don't the PCLinuxOS people contribute to Mandriva? Contributors are welcome (and have been for years).
My main objection is simply that it doesn't have a vanilla (x)nethack package
If you file a bug with component "Package request", it is possible that this objection could be removed ... nethack_falconseye is available in contrib though, so you might want to motivate (what the differences/features/benefits are).
kde 4 really kicked mandrivas usability... I currently use 2009 Spring and kde 4.3 is a big improvement over older kde 4 versions, but quite often I regret switching from 2008 Spring. many features, that worked in 2008 spring are now broken
I've been using kmail quite a bit, and haven't had problems. I don't use akregator much ...
kile and kate's scripting feature don't work anymore
I think it should be back in KDE 4.4, but this is of course an upstream issue.
kaffeine can't handle non-square pixels anymore, so DVD playback is stretched on my 16:9 TV - and my bugreports are just ignored)
i get errors from PulseAudio all the time
dragon player is working quite well for me on KDE 4.2 on Mandriva 2009.1. The only thing I am missing in dragon is a decent playlist.
I cant mount encrypted harddrives at boot-time, not even with initscripts or using crypttab (i have to mount them manually after booting
If this is your bug, it may have workarounds for 2009.1, and is fixed in 2010.0 by the switch to plymouth (splashy was the cause in 2009.0 and 2009.1). If you have a different bug, you need to provide means to reproduce it ...
the one thing that's really improved is kdenlive)
I tried to install Mandriva 2010, but aparently its installer doesn't think my SSD is a harddrive... although all previous mandriva versions installed on it just fine... maybe I'll switch the ports where my harddrives are plugged in - that may change something, but then again i'll have to reinstall grub manually (mandrivas bootloader repair tool never worked for me)
I didn't try 2010.0 on my Acer Aspire One, so I can't comment here, but I didn't see any bugs filed on this.
mandriva 2009 was completely unusable with kde 4.1...
Which is why KDE3 was still available for it, unlike other distributions that were released at the same time.
Ubuntu does XFS, (as well as ext*, JFS, MurderFS and so on) through the standard installer.
XFS has been available on Mandrake/Mandriva since Mandrake 8.2 if I remember correctly. Since that time it has been possible for users to resize system filesystems (e.g. /usr) using a graphical interface. This is still not possible on many distributions.
mdraid, lvm and truecrypt only work through the alternate installer disc (but the curses interface ain't that much more difficult than the GUI, so it oughtn't be an issue.
The Mandriva installer supports RAID, LVM, and LUKS encryption in the graphical installer. This GUI tool is also available after installation.
about new gentoo releases?
According to the release engineering page, there was.
Someone should post a story everytime they emerge --sync && emerge world
No, otherwise people running Mandriva cooker should post every time they 'urpmi --auto-update', or people running Debian testing should every time they run 'apt-get upgrade', or users on Fedora rawhide every time they run 'yum update' (ok, for Fedora, maybe not *every* time ...).
Just because you compiled it, doesn't mean you got it sooner than anyone else. (Note, the Mandriva build system is currently not accepting build submissions for "cooker" as cooker is still in freeze, the build system will be back to the usual 50+ packages per day by the end of the week).
As much as Debian died by dpkg, in other words, not at all. I guess you didn't try urpmi (which was in a released version of Mandriva before apt was in a stable release of Debian)?
So, when no more founders of Microsoft are employed by Microsoft, they should change their name, or their customers should consider switching?
What really made Mandrake, and continues to make Mandriva, is not one person, but the combination of employees and contributors. While many of both have come and gone, a lot of the contributors from the Mandrake era still use and contribute to the distro, and new contributors join quite often.
If you bothered to look, you would probably find that Mandriva is more open than Ubuntu or Fedora (not sure about "Open"SUSE).
Which model? Some models e.g. F4?) only have a very limited (WIndows only) software set, which only upload the data to the Polar web service (https://www.polarpersonaltrainer.com/). I don't remember any way to get the data out.
There are open-source utilities to download the data, but I have not had much success with them, and so could not use PolarViewer etc. with them.
I guess the more expensive models allow use from other software.
In a kerberized environment, kdestroy (or click on the "Remove Credentials Cache" option on the krb5-auth-dialog applet) is enough to ensure that all your access will not work until you re-authenticate. Remote shell sessions, web sites, email etc.
Now, if KDE had an option to kdestroy on screen saver lock, and if it correctly got tickets on unlock, it would be a lot more usable. /me logs some bugs ...
It's been mentioned elsewhere (but not here as far as I can tell) that this development is particularly notable, given that Windows doesn't support Exchange out of the box.
But, most linux distributions have for the past 6 months (thanks to evolution-mapi). And that's not even counting the previous support (based on the OWA API) that was available via evolution-exchange ... and neither requires you upgrade to Exchange 2007 (but evolution-mapi supports Exchange 2007, where the old OWA API is no longer available).
3 years and 22 days ago?
(Only Americans would still stick to a date format that is so ambiguous ....)
I did think of going back to mandriva, after gentoo - but that was right after the mandriva 9.2 'cd drive killer' kernel bug. OK, it turned out to be LG's fault in the end, but it did rather put me off!
Ironically, Gentoo had killed a lot of drives with the same patch months before (in their live Gaming CD), but didn't manage to figure out the cause, and left lots of toasted drives behind.
In fact, it was Mandrake (at the time) that tracked down this bug, got a resolution from the hardware vendor to restore drives to a working state, and ensure that the packet writing patch could get into mainline with some testing.
but then they went to KDE4 far to early with MDV 2009.0
As opposed to many other distros that shipped KDE 4.0 as the only option 6 months earlier, where 2009.0 shipped 4.1 by default, but still had 3.5 available?
I couldn't even re-find solutions to problems I'd previously had solved (e.g. an odd xorg.conf for my monitor). Its a lot better today, but there is still an odd gap between the main site [mandriva.com] and the place all the really useful [mandriva.com] information is kept.
Well, you should have been able to find a bug report ... and these days the forum is open to anyone, or, what do you think should fill the gap ?