I'd like to discuss:
FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths How can we use this further, and get it somewhat Red Hat blessed if we need to
FedoraUsers - http://fedorausers.org/ This is a site that hasn't actually been born yet, but I'd like the relevant parties to show up to discuss how we can have a site that makes sense to one of our largest target markets: USERS
Does anyone have contact information for who's behind fedorausers.org ? Whois doesn't tell me too much
Triaging guidelines - http://fedoraproject.org/wiki/TriagingGuidelines How can we market to triagers? How can we make it _absolutely easy_ for triagers to get on the bandwagon? If I didn't know Bugzilla, how'd I use it kind of improvements, all user targetted
Fedora LiveCD - http://fedoraproject.org/wiki/Kadischi We have a LiveCD in the making. LiveCDs as we all know are incredibly useful as a marketing tool (and a rescue tool, a demo tool, etc.). Let's find some use cases, maybe more developers to join, and lets give this a spin if there's techincal prowess among us. I think a lot of inspiration with regards to marketing livecd's can be honed from http://live.gnome.org/GnomeLiveCd
Get people on the CMC lists Greg has created this, I have access to it, so its time to get off my butt and get cracking
All this in addition to our current schedule. Just thought I'd put this up here, so we all have some time to prepare. Yes, hefty plate for that one hour, I don't expect it all to be done then, but hey, I'm giving it a shot =)
On Sat, 2005-09-10 at 23:37 +1000, Colin Charles wrote:
I'd like to discuss: FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths
You need to _expand_, _technically_ on these points. Especially the filesystem comments. Don't be general, be specific.
E.g., ReiserFS does not support SELinux, ACLs and other extents well, let alone NFS and other core functionality/compatibility issues are a "show-stopper" for shipping in Fedora releases. JFS suffers from much of the same compatibility.
As far as XFS, be complementary given the fact that it _does_ have a lot of support -- equaling Ext3 in many areas -- but at this time, most people only consider the official SGI kernel releases to be of "production quality" and not the stock kernel implementation.
Those are very sound facts. If someone would like me to write the section, I can do so very _point-by-point_ in the focus of what Fedora (like Red Hat before) must have in a filesystem.
On Sun, 2005-09-11 at 00:03 -0500, Bryan J. Smith wrote:
On Sat, 2005-09-10 at 23:37 +1000, Colin Charles wrote:
I'd like to discuss: FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths
You need to _expand_, _technically_ on these points. Especially the filesystem comments. Don't be general, be specific.
E.g., ReiserFS does not support SELinux, ACLs ...
Might also be worth mentioning namesys/Hans Reiser are not working on adding support (Although some people at SuSE are.)
Last post I saw from the SuSE people was: http://lkml.org/lkml/2004/12/7/142 which is in our kernel. I understand there's been more issues (which may have been fixed in 2.6.13) reported since.
-Toshio
Bryan J. Smith wrote:
On Sat, 2005-09-10 at 23:37 +1000, Colin Charles wrote:
I'd like to discuss: FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths
You need to _expand_, _technically_ on these points. Especially the filesystem comments. Don't be general, be specific.
E.g., ReiserFS does not support SELinux, ACLs and other extents well, let alone NFS and other core functionality/compatibility issues are a "show-stopper" for shipping in Fedora releases. JFS suffers from much of the same compatibility.
As far as XFS, be complementary given the fact that it _does_ have a lot of support -- equaling Ext3 in many areas -- but at this time, most people only consider the official SGI kernel releases to be of "production quality" and not the stock kernel implementation.
Those are very sound facts. If someone would like me to write the section, I can do so very _point-by-point_ in the focus of what Fedora (like Red Hat before) must have in a filesystem.
I don't know if that is really within the scope of this document. We could very easily write entire pages on some of the points (and indeed we have) and trying to cram it all on this one page would result in a massive document that goes far beyond what we really want to present here. On some of the points that require more elaboration, we can redirect the users to more information, or write sub-pages of the FedoraMyths page to hold the more in-depth information. Please keep in mind that many people will only be interested in one or two of the points, and many novice users will be looking at this document.
On Sun, 2005-09-11 at 01:03 -0500, Patrick Barnes wrote:
I don't know if that is really within the scope of this document. We could very easily write entire pages on some of the points (and indeed we have) and trying to cram it all on this one page would result in a massive document that goes far beyond what we really want to present here. On some of the points that require more elaboration, we can redirect the users to more information, or write sub-pages of the FedoraMyths page to hold the more in-depth information. Please keep in mind that many people will only be interested in one or two of the points, and many novice users will be looking at this document.
Okay, I can thin my comments it out into 3-4 statements total.
But I think the current form is technically misleading. Instead on focusing on saying what is and isn't experimental, why not say that Red Hat requires certain functions to work on a filesystem, JFS and ReiserFS do not, and while XFS does, it's support is still experimental and not as well tested.
On Sun, 2005-09-11 at 04:54 -0500, Bryan J. Smith wrote:
Okay, I can thin my comments it out into 3-4 statements total. But I think the current form is technically misleading. Instead on focusing on saying what is and isn't experimental, why not say that Red Hat requires certain functions to work on a filesystem, JFS and ReiserFS do not, and while XFS does, it's support is still experimental and not as well tested.
Dang, I'm half-asleep. Do-over ...
Say something like ...
Fedora Core, more than most other Linux distributions, requires the filesystem have extensive kernel/application support. Ext2/Ext3 have a long history on all Linux distributions, so kernel and application support is commonly implemented. This includes kernel NFS services, quotas and Extended Attributes (EAs) for things such as Access Control Lists (ACLs) and Mandatory Access Controls (MACs) like SELinux.
ReiserFS is not working on supporting many of these features, which is a show-stopper, and has a long history if compatibility issues with traditional services Fedora Core has been used for. JFS is still missing many of these components, and suffers from the same compatibility history as ReiserFS. While XFS does have extensive support, and SGI has produced releases of XFS for prior Red Hat kernels, Fedora Core has not tested XFS extensive, and there are many known issues with XFS in the Linux kernel (outside of SGI's control).
On Sun, 2005-09-11 at 05:01 -0500, Bryan J. Smith wrote:
Dang, I'm half-asleep. Do-over ... Say something like ...
Fedora Core, more than most other Linux distributions, requires the filesystem have extensive kernel/application support. Ext2/Ext3 have a long history on all Linux distributions, so kernel and application support is commonly implemented. This includes kernel NFS services, quotas and Extended Attributes (EAs) for things such as Access Control Lists (ACLs) and Mandatory Access Controls (MACs) like SELinux.
ReiserFS is not working on supporting many of these features, which is a show-stopper, and has a long history if compatibility issues with traditional services Fedora Core has been used for. JFS is still missing many of these components, and suffers from the same compatibility history as ReiserFS. While XFS does have extensive support, and SGI has produced releases of XFS for prior Red Hat kernels, Fedora Core has not tested XFS extensive, and there are many known issues with XFS in the Linux kernel (outside of SGI's control).
Actually, I just re-read that and it's too subjective and, worse yet, makes statements on behalf of other entities (which we obviously can't do).
I'll come up with a better set of statements after I get some sleep, as well as after I find a few, solid mailing list links.
On Sun, 2005-09-11 at 05:04 -0500, Bryan J. Smith wrote:
On Sun, 2005-09-11 at 05:01 -0500, Bryan J. Smith wrote:
Dang, I'm half-asleep. Do-over ... Say something like ...
Fedora Core, more than most other Linux distributions, requires the filesystem have extensive kernel/application support. Ext2/Ext3 have a long history on all Linux distributions, so kernel and application support is commonly implemented. This includes kernel NFS services, quotas and Extended Attributes (EAs) for things such as Access Control Lists (ACLs) and Mandatory Access Controls (MACs) like SELinux.
ReiserFS is not working on supporting many of these features, which is a show-stopper, and has a long history if compatibility issues with traditional services Fedora Core has been used for. JFS is still missing many of these components, and suffers from the same compatibility history as ReiserFS. While XFS does have extensive support, and SGI has produced releases of XFS for prior Red Hat kernels, Fedora Core has not tested XFS extensive, and there are many known issues with XFS in the Linux kernel (outside of SGI's control).
Actually, I just re-read that and it's too subjective and, worse yet, makes statements on behalf of other entities (which we obviously can't do).
As has already been said, it may not need much detail. The important points are probably that ReiserFS doesn't yet support key features x, y and z, and XFS isn't suited or reliable for standard setups.
I tagged that myth on after an IRC discussion about unreasonable user requests - people semi-regularly claim that Fedora should support/default to ReiserFS (as SUSE does, I think) because it's supposedly faster or cleverer or whatever, and that XFS is l33t, so it should be a standard installation option.
On Sun, 2005-09-11 at 16:12 +0100, Stuart Ellis wrote:
As has already been said, it may not need much detail. The important points are probably that ReiserFS doesn't yet support key features x, y and z,
Correct, that _needs_ to be in there. The problem with so many ReiserFS advocates is that they've _never_ used it for a production NFS server, or never used EAs. And there are recovery issues with the off-line tools being "out-of-sync" with changes in the filesystem (something that doesn't happen with Ext3 or XFS which have remained unchanged structurally for 10+ years).
and XFS isn't suited or reliable for standard setups.
XFS is _very_reliable_, but _only_ in the official SGI releases. I've found that XFS in the stock kernel is _not_, and most 3rd party rebuilds are incomplete and buggy.
I wish Red Hat would support XFS. It has all the interface/ compatibility, features of Ext3, plus a number more that enterprises need. But until Red Hat takes the time to build and test a "complete" XFS -- I can't trust the stock kernel builds or 3rd parties.
I tagged that myth on after an IRC discussion about unreasonable user requests - people semi-regularly claim that Fedora should support/default to ReiserFS (as SUSE does, I think) because it's supposedly faster or cleverer or whatever,
ReiserFS is innovative. And it utterly _breaks_ standard kernel interfaces and compatibility as a result. Even SuSE developers have told me _not_ to use it for my needs, because their hacks for many things (from NFS to EAs) are suspect.
and that XFS is l33t, so it should be a standard installation option.
Actually, Red Hat needs to realize that Ext3 has scalability issues (I don't like it above 100GB and I do _not_ trust it above 1TB), and _lacks_ a lot of user-space features of XFS like xfsdump, xfs_fsr, etc..., let alone EAs and other meta-data is stored directly in the inode (which xfsdump then retains).
I still have quite a number of Red Hat Linux 7.3 systems in heavy, heavy production with SGI's official XFS 1.2.x release, and one major Red Hat Linux 9 system with SGI's official XFS 1.3.1 release. But that's the key issue, they are the _only_ official SGI releases for any distro.
I've tried to integrate SGI's kernel builds from CVS into Red Hat distros to little avail. And as I mentioned, the stock kernel releases are incomplete -- especially the 2.4 backport that is now in the stock 2.4 kernel. I would _never_ run it, and the stock 2.6 kernel continues to be suspect.
Which means until Red Hat pro-actively develops and tests XFS in newer kernel 2.6 Fedora Core releases, I can't trust XFS either.
WHICH MEANS (AND I HOPE RED HAT IS LISTENING ;-), when I need a large, scalable, high-performance NFS/LAN file server for my Fortune 100 clients -- I deploy Solaris/Opteron now. Ext3 is _not_ cut it, and it _never_ will. I have to agree 100% with Schwartz's comments -- Red Hat is _ignoring_ a significant segment of the enterprise LAN server market.
I still long for the day of the Ext3 + XFS combination Red Hat distribution. Ext3 is better for system and smaller data volumes, XFS is better for larger (especially large file) volumes. But that died once SGI stopped releasing official releases -- the last being 1.3.1 for Red Hat Linux 9.
It's not about "l33t" -- there are serious enterprise features missing in Ext3.
Bryan J. Smith wrote:
On Sun, 2005-09-11 at 16:12 +0100, Stuart Ellis wrote:
As has already been said, it may not need much detail. The important points are probably that ReiserFS doesn't yet support key features x, y and z,
Correct, that _needs_ to be in there. The problem with so many ReiserFS advocates is that they've _never_ used it for a production NFS server, or never used EAs. And there are recovery issues with the off-line tools being "out-of-sync" with changes in the filesystem (something that doesn't happen with Ext3 or XFS which have remained unchanged structurally for 10+ years).
and XFS isn't suited or reliable for standard setups.
XFS is _very_reliable_, but _only_ in the official SGI releases. I've found that XFS in the stock kernel is _not_, and most 3rd party rebuilds are incomplete and buggy.
I wish Red Hat would support XFS. It has all the interface/ compatibility, features of Ext3, plus a number more that enterprises need. But until Red Hat takes the time to build and test a "complete" XFS -- I can't trust the stock kernel builds or 3rd parties.
I tagged that myth on after an IRC discussion about unreasonable user requests - people semi-regularly claim that Fedora should support/default to ReiserFS (as SUSE does, I think) because it's supposedly faster or cleverer or whatever,
ReiserFS is innovative. And it utterly _breaks_ standard kernel interfaces and compatibility as a result. Even SuSE developers have told me _not_ to use it for my needs, because their hacks for many things (from NFS to EAs) are suspect.
and that XFS is l33t, so it should be a standard installation option.
Actually, Red Hat needs to realize that Ext3 has scalability issues (I don't like it above 100GB and I do _not_ trust it above 1TB), and _lacks_ a lot of user-space features of XFS like xfsdump, xfs_fsr, etc..., let alone EAs and other meta-data is stored directly in the inode (which xfsdump then retains).
I still have quite a number of Red Hat Linux 7.3 systems in heavy, heavy production with SGI's official XFS 1.2.x release, and one major Red Hat Linux 9 system with SGI's official XFS 1.3.1 release. But that's the key issue, they are the _only_ official SGI releases for any distro.
I've tried to integrate SGI's kernel builds from CVS into Red Hat distros to little avail. And as I mentioned, the stock kernel releases are incomplete -- especially the 2.4 backport that is now in the stock 2.4 kernel. I would _never_ run it, and the stock 2.6 kernel continues to be suspect.
Which means until Red Hat pro-actively develops and tests XFS in newer kernel 2.6 Fedora Core releases, I can't trust XFS either.
WHICH MEANS (AND I HOPE RED HAT IS LISTENING ;-), when I need a large, scalable, high-performance NFS/LAN file server for my Fortune 100 clients -- I deploy Solaris/Opteron now. Ext3 is _not_ cut it, and it _never_ will. I have to agree 100% with Schwartz's comments -- Red Hat is _ignoring_ a significant segment of the enterprise LAN server market.
I still long for the day of the Ext3 + XFS combination Red Hat distribution. Ext3 is better for system and smaller data volumes, XFS is better for larger (especially large file) volumes. But that died once SGI stopped releasing official releases -- the last being 1.3.1 for Red Hat Linux 9.
It's not about "l33t" -- there are serious enterprise features missing in Ext3.
Suggestion: Why don't you go ahead and write a wiki page about the assorted filesystems, their strengths and weaknesses, and why some aren't currently in Fedora. You could create the page at http://fedoraproject.org/wiki/FAQ/FileSystems and let me know when it is ready for review. Once we have a good version in place, we can write in just a small statement on the FedoraMyths page and direct the curious to the new page. If you would like to follow the progress of filesystem support and keep that page a living document, that would be great. We'll also create a link to the new page directly off of the FAQ.
On Sun, 2005-09-11 at 15:53 -0500, Patrick Barnes wrote:
Suggestion: Why don't you go ahead and write a wiki page about the assorted filesystems, their strengths and weaknesses, and why some aren't currently in Fedora. You could create the page at http://fedoraproject.org/wiki/FAQ/FileSystems and let me know when it is ready for review. Once we have a good version in place, we can write in just a small statement on the FedoraMyths page and direct the curious to the new page. If you would like to follow the progress of filesystem support and keep that page a living document, that would be great. We'll also create a link to the new page directly off of the FAQ.
Actually, I've written a _lot_ on XFS since I produced some RPMs back in early 2001 (just months prior to the official SGI XFS 1.0 release).
I have a larger, technical discussion on Ext3 and XFS is in my blog here which covers a _lot_ of recent status: http://thebs413.blogspot.com/2005/08/filesystem-fundamentals-and-practices.h...
I could probably dig up my posts from 2000, 2001, 2002 on-ward on my usage of different filesystems. One legacy one from June 2001 is here: http://www.geocities.com/thebs413/ITEC_JFS_2001Jun13.pdf
In a nutshell, I adopted Ext3 in 2000 after SuSE told me that ReiserFS would not be a feasible filesystem for kernel NFS. I have a few colleagues with SuSE that are "frank" on the continuing issues with ReiserFS for such applications.
As far as JFS, the problem was that it was ported from OS/2, and _not_ the AIX implementation. So the Linux JFS port was basically *0* ready for an UNIX-like platform, and the resulting "re-write" occurred. It had more to do with IBM honoring their Non-Compete clause with SCO in their Monterey agreement. This was prior to IBM breaking and violating the agreement (which SCO sued over years later), so they were honoring it at the time the Linux JFS fork came about.
Since then, I've been deploying XFS. Once the SGI XFS 1.2 release hit, I put it into serious production. But the lack of good integration in the kernel, even 2.6 (although much better than 2.4) has left me _not_ trusting anything except official SGI XFS kernels. And I've had trouble integrating anything but the old 1.2.x for RHL7.3 and 1.3.1 for RHL9 releases.
On Sun, 2005-09-11 at 15:53 -0500, Patrick Barnes wrote:
Suggestion: Why don't you go ahead and write a wiki page about the assorted filesystems, their strengths and weaknesses, and why some aren't currently in Fedora. You could create the page at http://fedoraproject.org/wiki/FAQ/FileSystems and let me know when it is ready for review. Once we have a good version in place, we can write in just a small statement on the FedoraMyths page and direct the curious to the new page. If you would like to follow the progress of filesystem support and keep that page a living document, that would be great. We'll also create a link to the new page directly off of the FAQ.
Yes Bryan, if you think the filesystem issue is /that/ important, that users of Fedora must know it all, then please by all means, create a sub-page on the wiki for this. We'll be happy to review it, and I'm sure we can get more technical heads from fedora-devel-list interested
In the meantime, telling the marketing list about filesystem limitations does not help us in our cause in any way. If you're unhappy about the XFS file support, maybe you want to escalate working on it, cleaning it up, and submitting patches in Bugzilla (and about now you're going to point to eon old bugs open asking for xfs to be updated...) or better still, the upstream kernel
Best regards
On Mon, 2005-09-12 at 11:54 +1000, Colin Charles wrote:
Yes Bryan, if you think the filesystem issue is /that/ important, that users of Fedora must know it all, then please by all means, create a sub-page on the wiki for this. We'll be happy to review it, and I'm sure we can get more technical heads from fedora-devel-list interested
And I said I _would_ do this over the next few days. I'm working 7 days a week right now, and will get you it ASAP. It will be polished and professional in tone, focus as well as legal-base.
In the meantime, telling the marketing list about filesystem limitations does not help us in our cause in any way.
I was just pointing out the fact that Fedora/Red Hat has continually hurt itself on the filesystems point because it does _not_ directly address why ReiserFS and JFS are not included, plays games in general it does _not_ have to -- but even worse -- goes one step further, acting like all of us long-time integrators and Ext3 proponents don't exist when we say "it's time to get serious about XFS."
If you're unhappy about the XFS file support,
No. That's not it at all! Do _not_ belittle my statements as such because you are belittling a _lot_ of consultants at Fortune 100 companies when you say such.
What I'm unhappy about the "marketing" aspects of why XFS is _not_ supported. I would much rather Fedora/Red Hat say, "we can't afford the personnel to develop and support a 2nd filesystem." That would be direct, fair and understanding.
But the comments that Ext3 does everything XFS does, Ext3 does not have limitations that XFS solves, refer to it as "experimental" and other comments insult the intelligence of those of us who actually deploy Fedora/Red Hat solutions.
That's the problem. Again, if Red Hat says it does not want to support a 2nd filesystem, that's one thing. But 99% of the comments I see on XFS, and why Fedora/Red Hat is not looking at it for the future, are full of FUD.
Sadly enough, it's in Fedora/Red Hat's own, best interest. Even Sun has basically all but said it.
maybe you want to escalate working on it, cleaning it up, and submitting patches in Bugzilla (and about now you're going to point to eon old bugs open asking for xfs to be updated...) or better still, the upstream kernel
You're kidding me, right? Do you _know_ the effort that would be involved?!
There is already a company, SGI, and their OSS team that is _more_than_willing_ to work with Red Hat and the Fedora developers. But at this point, from what I've seen, it's been a 1-way street. That's the problem.
My point has been, and will continue to be, that Red Hat, including those who work on Fedora Core as their paid job functions, needs to put people on integration and support of XFS for its _own_ future. It's not "XFS v. Ext3" -- it's "XFS to complement Ext3."
99% of the comments I see are "versus" comments. Hence the continued problem, hence the actual reversal _away_ from Fedora Core and Red Hat Enterprise Linux for some applications.
What I'm unhappy about the "marketing" aspects of why XFS is _not_
supported. I would much rather Fedora/Red Hat say, "we can't afford the personnel to develop and support a 2nd filesystem." That would be direct, fair and understanding.
The official answer for RHEL is here: http://kbase.redhat.com/faq/FAQ_80_5737.shtm
But the comments that Ext3 does everything XFS does, Ext3 does not have limitations that XFS solves, refer to it as "experimental" and other comments insult the intelligence of those of us who actually deploy Fedora/Red Hat solutions.
That's the problem. Again, if Red Hat says it does not want to support a 2nd filesystem, that's one thing. But 99% of the comments I see on XFS, and why Fedora/Red Hat is not looking at it for the future, are full of FUD.
Again, You should be specific about what you consider FUD in the Fedora Myths page. Avoid long rants. If anyone is willing to step up and maintain XFS or any other filesystem for that matter in Fedora, it can be send to the fedora development list.
regards Rahul
On Mon, 2005-09-12 at 09:00 +0530, Rahul Sundaram wrote:
The official answer for RHEL is here: http://kbase.redhat.com/faq/FAQ_80_5737.shtm
And while it is applicable to ReiserFS and JFS, it has _not_ been applicable to XFS. There have been numerous points during the 2.4 kernel where XFS was _better_ than Ext3 in support areas Red Hat sorely needed -- from quotas to ACLs to (and still today) filesystem backup (especially on-line).
This answer is oblivious to reality on XFS. I've read the same comments again and again -- others have done the same -- and everytime it's FUD. People making the comments are actually unfamiliar with all the issues solved by XFS.
Again, that viewpoint _is_ applicable to ReiserFS and JFS. As a proponent of Ext3 myself, I _do_ advocate why Red Hat does not support ReiserFS and JFS. But that advocacy _falls_flat_ when it comes to XFS. Anyone who knows the history of XFS' development and release on Linux knows this.
It's a _total_joke_ in the XFS group when Fedora/Red Hat come back and say "oh, there's nothing XFS does that Ext3 doesn't do" and there are all these "exceptions" they end up agreeing to -- such as the lack of so many "standard" on-line user-space tools in traditional UNIX filesystems that XFS has (since day 1 on Linux), the varying EA support and history of quota support, the removal on relying on the LVM2/DM stack to solve so many things (such as the lack of on-line user-space tools) which introduce more race conditions, etc...
That's what I'm talking about -- key enterprise features that are _expected_ in a multi-TB UNIX filesystem. Which forces us to send our clients to Solaris, instead of Fedora Core or Red Hat Enterprise Linux.
Again, You should be specific about what you consider FUD in the Fedora Myths page. Avoid long rants.
I've got plenty of specifics, and I've tried to address them here, as well as in posts before. But apparently you don't want to read what you consider a "long rant." That's just sad, because it is _not_ a "long rant."
If anyone is willing to step up and maintain XFS or any other filesystem for that matter in Fedora, it can be send to the fedora development list.
Try SGI and the XFS team. Unfortunately, it takes some level of "formal engagement" from Red Hat/Fedora kernel developers too. That hasn't happened from what I've seen over 4 years and, sadly enough, it's in Red Hat's own, best interest.
Bryan J. Smith wrote:
On Mon, 2005-09-12 at 09:00 +0530, Rahul Sundaram wrote:
The official answer for RHEL is here: http://kbase.redhat.com/faq/FAQ_80_5737.shtm
And while it is applicable to ReiserFS and JFS, it has _not_ been applicable to XFS. There have been numerous points during the 2.4 kernel where XFS was _better_ than Ext3 in support areas Red Hat sorely needed -- from quotas to ACLs to (and still today) filesystem backup (especially on-line).
That FAQ doesnt talk about specific features and hence applies to all filesystems
regards Rahul
On Mon, 2005-09-12 at 09:27 +0530, Rahul Sundaram wrote:
That FAQ doesnt talk about specific features and hence applies to all filesystems
Again, that's the blanket problem. And it's not about "features."
So from the Fedora Myths page, it's about the reality that Fedora Core ships as a platform that does X, Y and Z.
Because ReiserFS and JFS don't support X, Y and/or Z, they are not options -- and likely never will be until they do. The lack of such support is the "show stopper" for inclusion.
But then that's a problem for FS, because it _does_ do X, Y and Z -- EA ACL/SELinux, Quotas, NFS, etc... So there is another reason why it's not included by default. E.g., XFS has known issues with 4KiB stacks, and some NFS support issues have been introduced in kernel 2.6 (that didn't exist with SGI's kernel 2.4 implementations).
I can be very, very technically accurate, but not "over-assuming" at the same time -- especially from the legal aspect of claiming things that might be disputable.
Hi
doOn Mon, 2005-09-12 at 09:27 +0530, Rahul Sundaram wrote:
That FAQ doesnt talk about specific features and hence applies to all filesystems
Again, that's the blanket problem. And it's not about "features."
Its also about resources. Fedora myths page can simply say this " Red Hat does not have the resources require to support other filesystems besides ext3 though it is provided as a option during the installation with specific parameters. If the community has sufficient interest in other filesystems to maintain it, send in a proposal to the Fedora development list with the details."
I would recommend not talking about specific features unless someone is willing to keep it updated regards Rahul
WHICH MEANS (AND I HOPE RED HAT IS LISTENING ;-), when I need a large, scalable, high-performance NFS/LAN file server for my Fortune 100 clients -- I deploy Solaris/Opteron now.
Ext3 is _not_ cut it, and it _never_ will. I have to agree 100% with Schwartz's comments -- Red Hat is _ignoring_ a significant segment of the enterprise LAN server market. \\\
Never is a long time. Ext3 is under active development and will continue to get more enhancements . If you want to send in customer feedback using the appropriate support channels would be a better idea. If you do believe that the page can highlight things in a better way that is invasive, make a alternative wiki page as a draft and point that to this list for dicussions
regards Rahul
On Mon, 2005-09-12 at 04:13 +0530, Rahul Sundaram wrote:
Never is a long time. Ext3 is under active development and will continue to get more enhancements .
Let me be the _first_ to thank Red Hat and Tweedie for the Ext3 filesystem. I trust it implicitly. It started saving my bacon in early 2000, and I have trusted it ever since. It was a much needed, evolutionary filesystem from _trusted_ Ext2 which has not changed since the mid-'90s. There's nothing like knowing you can always do a full Ext2 fsck -- that's something I trust.
*BUT* I have 2 standing rules on my Ext3 deployments:
1. No Ext3 filesystem is ever larger than 1TB. I try to keep them 100GB or smaller. I still use Ext3 for system volumes, and I would _never_ use another filesystem for /, /tmp, /var and several other filesystems.
2. Avoid using Extended Attributes (EAs) on Ext3 for data filesystems, except for SELinux. This is more of a backup consideration than anything else. There is no "reliable" way to backup/restore EAs from Ext3. Sorry, although I personally like Jorg's work, star is not what I consider "enterprise quality."
Again, Ext3 will _continue_ to be a _good_ filesystem for Red Hat and its integrators like myself. But I can_not_ entrust it as a solution for multi-TB filesystems. Unfortunately, there's not much I can in Linux, and that's the problem -- one Red Hat should help solve.
XFS is a no-brainer.
XFS' structure was designed for 64-bit, with extents, with delayed writes, with EAs and other meta-data in the inodes, with a _full_suite_ of user-space tools from xfsdump to xfs_fsr (defragmentor) to the off- line xfs_repair tool.
KEY POINT: This structure has _not_ changed since 1995.
Everything has been built around that structure since 1995. There are no new "hacks" to "extend" the filesystem's design. It was not left "incomplete." And XFS was a _direct_port_ from Irix to Linux, bringing a _lot_ of capability to the kernel itself.
That included POSIX EA support from day 1, _official_ quota support even _before_ Ext3, excellent kernel NFS support in the SGI XFS releases for kernel 2.4 (although this is no longer the case) and many other details. That is what I trusted from 2001-2004 -- especially the XFS 1.2 release for Red Hat Linux 7.3 and 1.3.1 release for Red Hat Linux 9.
Unfortunately, there is _no_ other XFS release I can trust. It's not the filesystem itself, it's the _lack_ of consideration on several levels. XFS and SGI could really, really, _really_ use Red Hat's assistance on developing XFS as a "standard" filesystem for 2.6 kernel distributions. It's not that XFS is "experimental," it's just the changes that have occurred in the kernel since the filesystem first came over, and was extremely mission critical.
Including Red Hat changes like 4KiB stacks -- which I _do_ believe is a good idea. In fact, I have been extremely complementary and supportive of Red Hat's decision making versus SuSE, Debian and other distributions. 9 times out of 10, Red Hat makes the best choice time and time again. Ext3 was one of them. But Ext3 cannot continue to be extended and still leverage its trusted structure that was _not_ designed for the scale and features that enterprises now need.
The problem is that everytime I bring this up, people treat me like I'm some ReiserFS enthusiast who has never run large-scale NFS/SMB servers, or a JFS advocate who doesn't know it's history on Linux. I _do_ use Ext3 and I will _continue_ to deploy Ext3 for many filesystems. But Ext3 is not "cutting it" for some large data filesystems, and it will never offer what XFS has -- and more importantly -- has had since the mid-'90s.
ReiserFS and JFS are basically "no way" on Red Hat, with Red Hat's services focus, and I explain this to people regularly. Even SuSE developers admit what ReiserFS cannot support, and most understand why Red Hat does not support it. But XFS quite different in its compatibility, and it is only because of kernel changes and lack of distro support why issues are now occurring.
If you want to send in customer feedback using the appropriate support channels would be a better idea. If you do believe that the page can highlight things in a better way that is invasive, make a alternative wiki page as a draft and point that to this list for dicussions
I will. Until then, I invite people to read my blog.
I'm an Ext3 system integrator. The problem is that the lack of XFS as a complement to Ext3 is causing me to deploy Solaris/Opteron instead of Linux/Opteron for LAN servers now. Again, I'm not so pundit with little experience deploying mission critical LAN file servers claiming superior performance out of ReiserFS, or JFS advocate who is oblivious to the origins of Linux's JFS which has caused it's support issues.
I really don't prefer Sun. But because of Red Hat's stance, I've gone from moving from Solaris to Linux only to be moving back towards Solaris for some services. Linux is great for web servers, database servers (using "raw" volumes for Oracle, or smaller MySQL/PostgresSQL filesystems), grid computing and other areas. But for large file servers, the lack of a turnkey Linux distro with XFS (since the older SGI releases), has pushed me back to UNIX, and Solaris is where it's at for LAN fileservers -- especially on Opteron.
Bryan J. Smith wrote:
On Mon, 2005-09-12 at 04:13 +0530, Rahul Sundaram wrote:
Never is a long time. Ext3 is under active development and will continue to get more enhancements .
Let me be the _first_ to thank Red Hat and Tweedie for the Ext3 filesystem. I trust it implicitly. It started saving my bacon in early 2000, and I have trusted it ever since. It was a much needed, evolutionary filesystem from _trusted_ Ext2 which has not changed since the mid-'90s. There's nothing like knowing you can always do a full Ext2 fsck -- that's something I trust.
*BUT* I have 2 standing rules on my Ext3 deployments:
- No Ext3 filesystem is ever larger than 1TB. I try to keep them
100GB or smaller. I still use Ext3 for system volumes, and I would _never_ use another filesystem for /, /tmp, /var and several other filesystems.
- Avoid using Extended Attributes (EAs) on Ext3 for data filesystems,
except for SELinux. This is more of a backup consideration than anything else. There is no "reliable" way to backup/restore EAs from Ext3. Sorry, although I personally like Jorg's work, star is not what I consider "enterprise quality."
Again, Ext3 will _continue_ to be a _good_ filesystem for Red Hat and its integrators like myself. But I can_not_ entrust it as a solution for multi-TB filesystems. Unfortunately, there's not much I can in Linux, and that's the problem -- one Red Hat should help solve.
XFS is a no-brainer.
XFS' structure was designed for 64-bit, with extents, with delayed writes, with EAs and other meta-data in the inodes, with a _full_suite_ of user-space tools from xfsdump to xfs_fsr (defragmentor) to the off- line xfs_repair tool.
KEY POINT: This structure has _not_ changed since 1995.
Everything has been built around that structure since 1995. There are no new "hacks" to "extend" the filesystem's design. It was not left "incomplete." And XFS was a _direct_port_ from Irix to Linux, bringing a _lot_ of capability to the kernel itself.
That included POSIX EA support from day 1, _official_ quota support even _before_ Ext3, excellent kernel NFS support in the SGI XFS releases for kernel 2.4 (although this is no longer the case) and many other details. That is what I trusted from 2001-2004 -- especially the XFS 1.2 release for Red Hat Linux 7.3 and 1.3.1 release for Red Hat Linux 9.
Unfortunately, there is _no_ other XFS release I can trust. It's not the filesystem itself, it's the _lack_ of consideration on several levels. XFS and SGI could really, really, _really_ use Red Hat's assistance on developing XFS as a "standard" filesystem for 2.6 kernel distributions. It's not that XFS is "experimental," it's just the changes that have occurred in the kernel since the filesystem first came over, and was extremely mission critical.
Including Red Hat changes like 4KiB stacks -- which I _do_ believe is a good idea. In fact, I have been extremely complementary and supportive of Red Hat's decision making versus SuSE, Debian and other distributions. 9 times out of 10, Red Hat makes the best choice time and time again. Ext3 was one of them. But Ext3 cannot continue to be extended and still leverage its trusted structure that was _not_ designed for the scale and features that enterprises now need.
The problem is that everytime I bring this up, people treat me like I'm some ReiserFS enthusiast who has never run large-scale NFS/SMB servers, or a JFS advocate who doesn't know it's history on Linux. I _do_ use Ext3 and I will _continue_ to deploy Ext3 for many filesystems. But Ext3 is not "cutting it" for some large data filesystems, and it will never offer what XFS has -- and more importantly -- has had since the mid-'90s.
ReiserFS and JFS are basically "no way" on Red Hat, with Red Hat's services focus, and I explain this to people regularly. Even SuSE developers admit what ReiserFS cannot support, and most understand why Red Hat does not support it. But XFS quite different in its compatibility, and it is only because of kernel changes and lack of distro support why issues are now occurring.
If you want to send in customer feedback using the appropriate support channels would be a better idea. If you do believe that the page can highlight things in a better way that is invasive, make a alternative wiki page as a draft and point that to this list for dicussions
I will. Until then, I invite people to read my blog.
I'm an Ext3 system integrator. The problem is that the lack of XFS as a complement to Ext3 is causing me to deploy Solaris/Opteron instead of Linux/Opteron for LAN servers now. Again, I'm not so pundit with little experience deploying mission critical LAN file servers claiming superior performance out of ReiserFS, or JFS advocate who is oblivious to the origins of Linux's JFS which has caused it's support issues.
I really don't prefer Sun. But because of Red Hat's stance, I've gone from moving from Solaris to Linux only to be moving back towards Solaris for some services. Linux is great for web servers, database servers (using "raw" volumes for Oracle, or smaller MySQL/PostgresSQL filesystems), grid computing and other areas. But for large file servers, the lack of a turnkey Linux distro with XFS (since the older SGI releases), has pushed me back to UNIX, and Solaris is where it's at for LAN fileservers -- especially on Opteron.
1. You're ranting. 2. You're doing it in the wrong place. 3. If you turn your passions into action, rather than ranting, you might be pleasantly surprised by the results.
Your opinions regarding the assorted filesystems really are irrelevant to this list, as are the technical details behind those reasons. It is not that nobody cares, but nobody here is prepared to do anything about it. You should offer assistance to the developers of XFS on Linux. If upstream support for XFS improves, Red Hat and/or the Fedora Foundation will be more inclined to support it. As Rahul and I have suggested, you are welcome to create an *objective* wiki page, describing the reasons behind the inclusion/exclusion of different filesystem types. Anything else is off-topic on this list, and is likely to be met only with hostility or frustration.
Bryan J. Smith wrote:
On Mon, 2005-09-12 at 04:13 +0530, Rahul Sundaram wrote:
Never is a long time. Ext3 is under active development and will continue to get more enhancements .
Let me be the _first_ to thank Red Hat and Tweedie for the Ext3 filesystem. I trust it implicitly. It started saving my bacon in early 2000, and I have trusted it ever since. It was a much needed, evolutionary filesystem from _trusted_ Ext2 which has not changed since the mid-'90s. There's nothing like knowing you can always do a full Ext2 fsck -- that's something I trust.
*BUT* I have 2 standing rules on my Ext3 deployments:
- No Ext3 filesystem is ever larger than 1TB. I try to keep them
100GB or smaller. I still use Ext3 for system volumes, and I would _never_ use another filesystem for /, /tmp, /var and several other filesystems.
While I dont really want to take this further on this list, let me point to you the RHEL 4 U1 release notes.
"The ext2 and ext3 filesystems have an internal limit of 8 TB. Devices up to this limit have been tested in Red Hat Enterprise Linux 4 Update 1."
- Avoid using Extended Attributes (EAs) on Ext3 for data filesystems,
except for SELinux. This is more of a backup consideration than anything else. There is no "reliable" way to backup/restore EAs from Ext3. Sorry, although I personally like Jorg's work, star is not what I consider "enterprise quality."
This is not a filesystem limitation
regards Rahul
On Mon, 2005-09-12 at 06:06 +0530, Rahul Sundaram wrote:
While I dont really want to take this further on this list, let me point to you the RHEL 4 U1 release notes. "The ext2 and ext3 filesystems have an internal limit of 8 TB. Devices up to this limit have been tested in Red Hat Enterprise Linux 4 Update 1."
And I understand it has even been raised to 17.6TB (16TiB) now. That's _not_ what I'm talking about.
This is not a filesystem limitation
But lack of a good set of safe, trusted user-space utilities for the filesystem are. Red Hat offers a non-filesystem utility, star. That's not good for enterprises. And the more Red Hat tries to distract people from that, the more they are denying what they cannot offer.
That's _exactly_ the self-defeating marketing I'm talking about.
Hi
And I understand it has even been raised to 17.6TB (16TiB) now. That's _not_ what I'm talking about.
Your personal choices are not a list concern
This is not a filesystem limitation
That's _exactly_ the self-defeating marketing I'm talking about.
We are not talking about RHEL marketing here
regards Rahul
On Mon, 2005-09-12 at 06:26 +0530, Rahul Sundaram wrote:
Your personal choices are not a list concern
That's fine. Of course, what you call my "personal choices" is just one of the _many_ voices. If Fedora and Red Hat want to ignore these voices, then they make their own issues.
Again, I'm not some garden variety punk who thinks Fedora should have this, Fedora should have that. I'm telling you not only what people who believe in Fedora Core, like Red Hat Linux before it, are feeling, but what others are picking up on.
We are not talking about RHEL marketing here
It's all the same. What doesn't go into Fedora Core rarely makes it into Red Hat Enterprise Linux. Frankly, the tunnel vision I see regularly is what bothers me most.
But the thing I always remember, which is the reason I love Red Hat, is the developers. Red Hat is a GPL company first and foremost, and until that changes, tunnel vision -- let alone the larger issue of marketing -- really doesn't matter to me.
But I'm still forced to deploy Sun when I'd rather deploy Fedora/Red Hat.
Hi
We are not talking about RHEL marketing here
It's all the same. What doesn't go into Fedora Core rarely makes it into Red Hat Enterprise Linux. Frankly, the tunnel vision I see regularly is what bothers me most.
Its not tunnel vision. RHEL features and bugfixes are driven by the customer requests and support calls in a very real sense and not rants in a mailing list especially not ones in Fedora marketing list. If you want Red Hat to listen in relationship to RHEL, try http://redhat.com/support.
For Fedora, here is how you contribute. Note that the amount of oppurtunities is much more than RHEL
* File bug reports * File feature requests with good rationale. I would like to have more than 8 TB filesystems for such and such scenarios might be better than "I want XFS" * Write code and send in patches * Write documentation * Translate UI and docs * Package in Fedora Extras * Art work, themes and sound themes * Advocate * Last but not the least, use it
There might be a few I have missed
regards Rahul
On Mon, 2005-09-12 at 08:51 +0530, Rahul Sundaram wrote:
Its not tunnel vision.
From the answers I've seen, it is -- it damn is.
RHEL features and bugfixes are driven by the customer requests and support calls in a very real sense
Because Red Hat is getting web servers and Oracle servers (raw slices) and grid computing sales. Red Hat is now _losing_ multi-TB file server sales because they can_not_ deliver on it with just Ext3 -- especially given its lack of user-space filesystem support.
Sun is eating that up. I guess Red Hat has decided it is not worth doing? Or is it because the Fedora developments, including those by its own developers who obliviously think "Ext3 doesn't do anything XFS does."
That's marketing -- however you slice it -- and a major issue.
and not rants in a mailing list
This isn't a "rant." This is me, a system integrator at Fortune 100 companies over the last 4 years, telling you not only what I have experienced, but what others have experienced. We cannot offer Fedora Core and Red Hat Enterprise Linux solutions -- and it starts with Fedora developments.
The fact that you're taking it as a "rant" instead of -- "hmmm, maybe there is a 4-year 'suggestion' this guy is trying to make" -- goes to the heart of the matter. I'm not the only "ranting lunatic" out there having to "move back" to Solaris for solutions.
especially not ones in Fedora marketing list.
I honestly don't know where else to turn. And when I read the rather "say nothing" portion that just feeds the ReiserFS and JFS proponents their "Red Hat is unfairly biased towards Ext3" non-sense, I have to agree with them from the XFS mis-conceptions presented.
If you want Red Hat to listen in relationship to RHEL, try http://redhat.com/support.
Been there -- a good 4 years.
For Fedora, here is how you contribute. Note that the amount of oppurtunities is much more than RHEL
Yes, I know.
- File bug reports
- File feature requests with good rationale. I would like to have more
than 8 TB filesystems for such and such scenarios might be better than "I want XFS"
You didn't hear a word I said. That is so sad.
- Write code and send in patches
I'll get farther with SGI than I will Red Hat. I'm asking -- at the Fedora level -- to try to get Red Hat involved. I've been trying various avenues for 4 years now to no avail.
And it's only getting more pertinent as Solaris is now on Opteron.
- Write documentation
Way, way -- WAY AHEAD -- of you there.
- Translate UI and docs
- Package in Fedora Extras
I'm actually looking to submit several Fedora Extras packages.
- Art work, themes and sound themes
- Advocate
I'm well known as a "Red Hat apologist" (even though I hate the title). So I find it humorous that you would suggest that to me. ;->
- Last but not the least, use it
Considering I've been integrating Fedora Core in companies as large as one major Fortune 20 company in just the last 2 years, and several other Fortune 100 companies as well -- again, WAY AHEAD OF YOU THERE.
There might be a few I have missed
Again, I'm not some "I want filesystem X" puke. But I know you'll feel free to take my comments as such.
Boys, boys. You're both pretty. :)
Bryan, I get your point. Lemme talk to some smart folks and see what the general feeling is around XFS "inside the walls." I'm kinda curious myself now.
--g
_____________________ ____________________________________________ Greg DeKoenigsberg ] [ the future masters of technology will have Community Relations ] [ to be lighthearted and intelligent. the Red Hat ] [ machine easily masters the grim and the ] [ dumb. --mcluhan
On Sun, 11 Sep 2005, Bryan J. Smith wrote:
On Mon, 2005-09-12 at 08:51 +0530, Rahul Sundaram wrote:
Its not tunnel vision.
From the answers I've seen, it is -- it damn is.
RHEL features and bugfixes are driven by the customer requests and support calls in a very real sense
Because Red Hat is getting web servers and Oracle servers (raw slices) and grid computing sales. Red Hat is now _losing_ multi-TB file server sales because they can_not_ deliver on it with just Ext3 -- especially given its lack of user-space filesystem support.
Sun is eating that up. I guess Red Hat has decided it is not worth doing? Or is it because the Fedora developments, including those by its own developers who obliviously think "Ext3 doesn't do anything XFS does."
That's marketing -- however you slice it -- and a major issue.
and not rants in a mailing list
This isn't a "rant." This is me, a system integrator at Fortune 100 companies over the last 4 years, telling you not only what I have experienced, but what others have experienced. We cannot offer Fedora Core and Red Hat Enterprise Linux solutions -- and it starts with Fedora developments.
The fact that you're taking it as a "rant" instead of -- "hmmm, maybe there is a 4-year 'suggestion' this guy is trying to make" -- goes to the heart of the matter. I'm not the only "ranting lunatic" out there having to "move back" to Solaris for solutions.
especially not ones in Fedora marketing list.
I honestly don't know where else to turn. And when I read the rather "say nothing" portion that just feeds the ReiserFS and JFS proponents their "Red Hat is unfairly biased towards Ext3" non-sense, I have to agree with them from the XFS mis-conceptions presented.
If you want Red Hat to listen in relationship to RHEL, try http://redhat.com/support.
Been there -- a good 4 years.
For Fedora, here is how you contribute. Note that the amount of oppurtunities is much more than RHEL
Yes, I know.
- File bug reports
- File feature requests with good rationale. I would like to have more
than 8 TB filesystems for such and such scenarios might be better than "I want XFS"
You didn't hear a word I said. That is so sad.
- Write code and send in patches
I'll get farther with SGI than I will Red Hat. I'm asking -- at the Fedora level -- to try to get Red Hat involved. I've been trying various avenues for 4 years now to no avail.
And it's only getting more pertinent as Solaris is now on Opteron.
- Write documentation
Way, way -- WAY AHEAD -- of you there.
- Translate UI and docs
- Package in Fedora Extras
I'm actually looking to submit several Fedora Extras packages.
- Art work, themes and sound themes
- Advocate
I'm well known as a "Red Hat apologist" (even though I hate the title). So I find it humorous that you would suggest that to me. ;->
- Last but not the least, use it
Considering I've been integrating Fedora Core in companies as large as one major Fortune 20 company in just the last 2 years, and several other Fortune 100 companies as well -- again, WAY AHEAD OF YOU THERE.
There might be a few I have missed
Again, I'm not some "I want filesystem X" puke. But I know you'll feel free to take my comments as such.
-- Bryan J. Smith b.j.smith@ieee.org http://thebs413.blogspot.com
The best things in life are NOT free - which is why life is easiest if you save all the bills until you can share them with the perfect woman
-- Fedora-marketing-list mailing list Fedora-marketing-list@redhat.com http://www.redhat.com/mailman/listinfo/fedora-marketing-list
On Sun, 2005-09-11 at 23:49 -0400, Greg DeKoenigsberg wrote:
Bryan, I get your point.
Thanx. Remember, I'm an "outsider." I do a lot of marketing of Fedora Core, and I've always stuck right by Red Hat's decisions because of all the trademark mess (no good deed goes unpunished, and that's exactly what Red Hat's goodwill has always been taken as by so many companies).
Lemme talk to some smart folks and see what the general feeling is around XFS "inside the walls."
I'm sure the big issue is the adoption of 4KiB stacks, something I agree very much with Red Hat on, but I know SGI has had issues.
I'm kinda curious myself now.
Again, I'll put together something for the wiki by Tuesday, possibly tomorrow night. I just started with a new company this week -- and we're deep into hurricane deployments of our technology for communications.
Hi
I honestly don't know where else to turn. And when I read the rather "say nothing" portion that just feeds the ReiserFS and JFS proponents their "Red Hat is unfairly biased towards Ext3" non-sense, I have to agree with them from the XFS mis-conceptions presented.
If you are willing to take up the initiative to get other filesystem supported then send in a mail to fedora-devel with your detailed proposal. Red Hat is certainly biased towards ext3 since Red Hat has the necessary expertise to support it and enhance it as required as opposed to other filesystems. The only difference in relationship with Fedora is that the community can participate and take over maintenance other filesystems and things that Red Hat isnt able to allocate resources towards. I dont see Fedora myths page talking about feature comparisons for precisely this reason
regards Rahul
On Mon, 2005-09-12 at 09:20 +0530, Rahul Sundaram wrote:
If you are willing to take up the initiative to get other filesystem supported then send in a mail to fedora-devel with your detailed proposal.
I'll discuss this over on the XFS list. I need to get a better feel on what they believe would be required involvement on the Fedora end. But I know it's going to bring up some old history and/or bad tastes.
Red Hat is certainly biased towards ext3 since Red Hat has the necessary expertise to support it and enhance it as required as opposed to other filesystems.
And I'm telling you until Red Hat puts forth the interest to see XFS as "the complement" to Ext3 it really needs, then it matters really little. Which is why I've been trying to see Red Hat get serious about its future by putting forth the effort to integrate XFS formally.
Especially when there were 2-4 years ago, as there are even some still now, various things that XFS supports that Ext3 does not. And no, I'm not talking about "speed" or other "performance" non-sense. I'm talking about enterprise features -- like on-line user-space tools, etc...
The only difference in relationship with Fedora is that the community can participate and take over maintenance other filesystems and things that Red Hat isnt able to allocate resources towards.
Then _that's_ what needs to be said -- that Red Hat isn't able to allocate resources towards a 2nd filesystem. I'd more than accept that.
But anyone who knows anything about the histories of both Ext3 and XFS on Linux -- especially those of us who have deployed both over the years -- do _not_ accept the oblivious "XFS doesn't do anything Ext3 can" non- sense.
It's an argument that _is_ valid against ReiserFS and JFS for Fedora/Red Hat's focus that also gets, quite inaccurately, applied against XFS.
I dont see Fedora myths page talking about feature comparisons for precisely this reason
It's not about "features."
It's about things like lack of EA support which makes it impossible for ReiserFS and JFS to support SELinux, various history of kernel NFS, quote and other "standard" kernel interface/support issues with ReiserFS and JFS, etc... Things that are _core_ to Fedora Core/Red Hat releases that are "show stoppers."
But things that are _not_ true in the case of XFS.
Hi
Then _that's_ what needs to be said -- that Red Hat isn't able to allocate resources towards a 2nd filesystem. I'd more than accept that.
As far as I know no formal statements that ext3 supports all features that other filesystems do have been made by Red Hat. What has been repeatedly pointed out is the question of resources vs gain and requests in specific features that customers want based on their scenarios. If you do send in such a list to Red Hat support or partners , due consideration will given to those. For Fedora, get a detailed realistic plan to the fedora development list. Rest of the discussion doesnt really matter
regards Rahul
Bryan J. Smith wrote:
On Mon, 2005-09-12 at 06:26 +0530, Rahul Sundaram wrote:
Your personal choices are not a list concern
That's fine. Of course, what you call my "personal choices" is just one of the _many_ voices. If Fedora and Red Hat want to ignore these voices, then they make their own issues.
Again, I'm not some garden variety punk who thinks Fedora should have this, Fedora should have that. I'm telling you not only what people who believe in Fedora Core, like Red Hat Linux before it, are feeling, but what others are picking up on.
You have mentioned the 1 TB limit on how you would setup your systems. However since this isnt a technical limitation of the filesystem it doesnt make a case for other filesystems. It merely states your opinion on what you consider the limit which is entirely subjective..
regards Rahul
On 09/10/2005 09:37 AM, Colin Charles wrote:
I'd like to discuss:
FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths How can we use this further, and get it somewhat Red Hat blessed if we need to
Drive-by feedback: I see a whole bunch of negative items in bold, and then the word FACT in bold right underneath it. Someone quickly skimming will get the wrong idea. Highlight the positive as a contra to the negative. Don't highlight the negative assuming people will read the rebuttal.
Christopher Aillon wrote:
On 09/10/2005 09:37 AM, Colin Charles wrote:
I'd like to discuss:
FedoraMyths - http://fedoraproject.org/wiki/FedoraMyths How can we use this further, and get it somewhat Red Hat blessed if we need to
Drive-by feedback: I see a whole bunch of negative items in bold, and then the word FACT in bold right underneath it. Someone quickly skimming will get the wrong idea. Highlight the positive as a contra to the negative. Don't highlight the negative assuming people will read the rebuttal.
The current appearance of the document is largely dictated by the perspective and the form of the wiki. The 'myths' are the focus of the document. Dispelling them with the facts is its purpose. It is not meant to be read by drive-by reviewers. People who view the page are looking for reasons. That said, I will try to go back and improve the contrast to better highlight the facts.
On 09/11/2005 02:07 AM, Patrick Barnes wrote:
The current appearance of the document is largely dictated by the perspective and the form of the wiki. The 'myths' are the focus of the document. Dispelling them with the facts is its purpose. It is not meant to be read by drive-by reviewers. People who view the page are looking for reasons. That said, I will try to go back and improve the contrast to better highlight the facts.
The point is that your document is capable of being read drive-by, thanks to the nature of the internet (and e.g. Google). The people who are susceptible to myth and most need to read that document are also the most likely to be confused with the wrong message if they happen to come across it the same way I did.
Disregarding that, there is still something even AFTER i read the document through that makes the human (at least this one) focus on the bullet points. Those being negative really hurts. Let's focus on the positive. Perhaps something like:
* <b>Q: Is Fedora Core unstable and unreliable?</b> * <b>A: No. In fact, many businesses rely on Fedora Core for day-to-day operation and, in some cases, critical infrastructure.</b> The misconception that Fedora is unstable is driven by two things:
Basically, give the reader the bottom line information up front, and then if they want to read the full response, they can. This helps ensure the information we provide will be construed in a manner desirable to us.
marketing@lists.fedoraproject.org
-
Bryan Smith -
Christopher Aillon -
Colin Charles -
Greg DeKoenigsberg -
Patrick W. Barnes -
Rahul Sundaram -
Stuart Ellis -
Toshio Kuratomi