Hello SuSE people, I know that you can do a dd if=/dev/sda of=/dev/sdb which actually clones an entire drive, partitions, links, etc. Question is can I dd single partitions like dd if=/dev/sda1 of=/dev/sdb1 and still have the same result ? Partition by partition? What I'm trying to do is split an OS off a large drive to a smaller one to gain the partitions back on the larger drive because of that stupid libata decision. I will put it on a drive I have wiped from a much older OS .Will I need to paartition the wiped drive first? or will dd do that for me. Possible? Practical? RANT ON: For the life of me I cannot understand that stupid decision by the devs to use libata with it's 16 partition limitation just as hard drives were getting bigger and bigger. Really really dumb. And someone said they were working hard to get the large number of partitions back. That was a year ago. Nothing yet !! RANT OFF BTW I have copied everything on the "victim" OS with rsycnc. Just in case. Bob S -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Bob S a écrit :
Hello SuSE people,
I know that you can do a dd if=/dev/sda of=/dev/sdb which actually clones an entire drive, partitions, links, etc.
Question is can I dd single partitions like dd if=/dev/sda1 of=/dev/sdb1 and still have the same result ? Partition by partition?
certainly
What I'm trying to do is split an OS off a large drive to a smaller one to gain the partitions back on the larger drive
use of dd gives partitions of the same size on if and of sides. receive pazrtition needs to be at least as big as the source one and the rest is unusable (AFAIK)
.Will I need to paartition the wiped drive first?
certainly or will dd do that for me. no dd only copy (sector by sector in this situation)
devs to use libata with it's 16 partition limitation
I have read than a boot option allows use of the old library, but don't have the problem, so don't know exactly jdd -- Jean-Daniel Dodin Président du CULTe www.culte.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Tuesday 2008-07-15 at 20:47 -0400, Bob S wrote:
Hello SuSE people,
I know that you can do a dd if=/dev/sda of=/dev/sdb which actually clones an entire drive, partitions, links, etc.
Yes, but: if the destination is bigger, you loose all the extra space. It only works perfect if both disks are exactly equall.
Question is can I dd single partitions like dd if=/dev/sda1 of=/dev/sdb1 and still have the same result ? Partition by partition?
Yes, but each partition must be exactly equal size, or a bit bigger the destination (extra space is lost). If source is linux, I prefer to create partitions via fdisk, and copy files using rsync. Boot sectors and mbr (and grub/lilo) is another matter.
What I'm trying to do is split an OS off a large drive to a smaller one to gain the partitions back on the larger drive because of that stupid libata decision. I will put it on a drive I have wiped from a much older OS .Will I need to paartition the wiped drive first? or will dd do that for me.
If you dd the entire disk, no need to partition (it will be overwritten). If you dd partition by partition, you need to create them first.
Possible? Practical?
RANT ON: For the life of me I cannot understand that stupid decision by the devs to use libata with it's 16 partition limitation just as hard drives were getting bigger and bigger. Really really dumb. And someone said they were working hard to get the large number of partitions back. That was a year ago. Nothing yet !! RANT OFF
Agreed :-( - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIfdCEtTMYHG2NR9URAmUqAKCDbmDAwQ+dkyIKyKhhkCxZlo+xYQCcDmCX c8/VeAdI8WVpzN+aRTYBb5U= =Os2a -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Possible? Practical?
RANT ON: For the life of me I cannot understand that stupid decision by the devs to use libata with it's 16 partition limitation just as hard drives were getting bigger and bigger. Really really dumb. And someone said they were working hard to get the large number of partitions back. That was a year ago. Nothing yet !! RANT OFF
Agreed :-(
With 11.0 I assume you still have the option of using the old drivers/IDE interface, correct? So you are ranting about the current state of libata, even though Novell has not forced that on the user unless you choose to use the new driver subsystem (libata). As to libata, Tejun Heo (of Novell) has recently submitted a set of patches to the linux kernel team to address the limit. If accepted, I think the new limit would be 64(?). I have not seen any ack / nack responses to his patches, so I don't know the likelihood of them being accepted, nor of which vanilla kernel they may get into if/when they are accepted. Greg -- Greg Freemyer Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer First 99 Days Litigation White Paper - http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Wednesday 2008-07-16 at 11:53 -0400, Greg Freemyer wrote:
With 11.0 I assume you still have the option of using the old drivers/IDE interface, correct? So you are ranting about the current state of libata, even though Novell has not forced that on the user unless you choose to use the new driver subsystem (libata).
People with SATA disks are forced to use the new interface, limited to 14 partitions. People with PATA disks can still use the old interface and don't have that limit.
As to libata, Tejun Heo (of Novell) has recently submitted a set of patches to the linux kernel team to address the limit. If accepted, I think the new limit would be 64(?).
Good to know.
I have not seen any ack / nack responses to his patches, so I don't know the likelihood of them being accepted, nor of which vanilla kernel they may get into if/when they are accepted.
More waiting... :-) - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIfn4ttTMYHG2NR9URAum+AJ9Upl1vbTcK9sKCBUeBcB8lzy7rLACbB7u3 zJzRcVsC5LLzXQ2ccb/ZwK4= =0VFm -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wed, Jul 16, 2008 at 7:03 PM, Carlos E. R.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
The Wednesday 2008-07-16 at 11:53 -0400, Greg Freemyer wrote:
With 11.0 I assume you still have the option of using the old drivers/IDE interface, correct? So you are ranting about the current state of libata, even though Novell has not forced that on the user unless you choose to use the new driver subsystem (libata).
People with SATA disks are forced to use the new interface, limited to 14 partitions. People with PATA disks can still use the old interface and don't have that limit.
A lot of MB controllers offer legacy mode support for sata drives. I think that should work with drivers/ide as well, so you can get the large partition support that way as well. Greg -- Greg Freemyer Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer First 99 Days Litigation White Paper - http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wed, Jul 16, 2008 at 7:03 PM, Carlos E. R.
wrote: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
The Wednesday 2008-07-16 at 11:53 -0400, Greg Freemyer wrote:
With 11.0 I assume you still have the option of using the old drivers/IDE interface, correct? So you are ranting about the current state of libata, even though Novell has not forced that on the user unless you choose to use the new driver subsystem (libata).
People with SATA disks are forced to use the new interface, limited to 14 partitions. People with PATA disks can still use the old interface and don't have that limit.
A lot of MB controllers offer legacy mode support for sata drives.
I think that should work with drivers/ide as well, so you can get the large partition support that way as well.
Greg -- Oh, so let's go buy a new motherboard so that we can have the functionality
On Wednesday 16 July 2008 07:48:09 pm Greg Freemyer wrote: that we did before, The point is Greg, that all of this was so unneccesary. No forethought, or worse, disregard for the users. Nobody asks or "floats an idea" in this "Community". Just unilaterally invoked by those who know best. Bob S -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Thu, 17 Jul 2008 01:03:06 +0200 (CEST), Carlos E. R. wrote:
People with SATA disks are forced to use the new interface, limited to 14 partitions.
Though I personally doubt that there are many users that really need more then that. And even if, the kpartx trick should help getting more then 14 partitions mounted. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 16 July 2008 08:06:02 pm Philipp Thomas wrote:
On Thu, 17 Jul 2008 01:03:06 +0200 (CEST), Carlos E. R. wrote:
People with SATA disks are forced to use the new interface, limited to 14 partitions.
Though I personally doubt that there are many users that really need more then that. And even if, the kpartx trick should help getting more then 14 partitions mounted.
Well, I think that you would be surprised Phillipp. There are certainly enough of us that we should be considered. And why must we resort to "tricks" to gain the functionality that we had without "tricks". Tell me, what good is a 250 or 500 GB hard drive with only 14 partitions. I run several different OS's (four) plus several different iterations of SuSE. I like to try them all and experiment/compare them. I always make separate partitions for /home, /tmp, /var, and /usr/local because I believe that is the smart way to do it. Let's not get into the why's that I think that is smart. Now add the /swap and the /master partition and start to multiply. Please don't get into the Microsoft mode of thinking that the latest (or previous) SuSE is all that you need or may want. Linux is about freedom. Please don't be telling us what is best for us. Bob S -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wed, 16 Jul 2008 19:04:18 -0400, Bob S wrote:
Well, I think that you would be surprised Phillipp. There are certainly enough of us that we should be considered.
Did I say none? Or did I say those shouldn't be considered?
And why must we resort to "tricks" to gain the functionality that we had without "tricks".
As long as libata has a limit of 14, I'd say it's better to have a workaround that does indeed work then to have no possibility at all, wouldn't you agree? That's not to say that it wouldn't be better to have a way to lift that limit or at least raise that limit.
Tell me, what good is a 250 or 500 GB hard drive with only 14 partitions.
Oh, I can tell you why four 1 TB drives in a software RAID5 setup with essentially one partition makes sense :) This setup is used to mostly store TV recordings. For different Windows versions, Linux versions, OS/2 and Netware I use vmware and virtualbox.
run several different OS's (four) plus several different iterations of SuSE.
I do too, but I use virtualization and disk image files, serves most of my needs. Only situation that vm doesn't suffice is when I need good accelerated 3D graphics.
I like to try them all and experiment/compare them. I always make separate
partitions for /home, /tmp, /var, and /usr/local because I believe that is the smart way to do it.
There are pros and cons, but this IMHO mostly a matter of personal taste and thus not really something to discuss.
Now add the /swap and the /master partition and start to multiply.
/swap can be shared between different Linuxes (unless you want to be able to suspend to disc in all Linux versions). And out of interest, what's /master for?
Please don't get into the Microsoft mode of thinking that the latest (or previous) SuSE is all that you need or may want. Please don't be telling us what is best for us.
I did none of that, so why such a defensive mode?
Linux is about freedom.
Yes, but that includes the freedom of those that put together a distribution as to what they support. Linux, for instance, supports a wide range of architectures, yet openSUSE is only offered for a small subset. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Wednesday 16 July 2008 11:59:53 pm Philipp Thomas wrote:
On Wed, 16 Jul 2008 19:04:18 -0400, Bob S wrote:
Well, I think that you would be surprised Phillipp. There are certainly enough of us that we should be considered.
Hello Phillipp, I really don't know why I am bothering to answer this. You are very difficult to convince when you have your mind made up.
Did I say none? Or did I say those shouldn't be considered?
No you didn't. Not directly. Here is what you said," Though I personally doubt that there are many users that really need more then that." It is in the contest of your statement.
And why must we resort to "tricks" to gain the functionality that we had without "tricks".
As long as libata has a limit of 14, I'd say it's better to have a workaround that does indeed work then to have no possibility at all, wouldn't you agree? That's not to say that it wouldn't be better to have a way to lift that limit or at least raise that limit.
Yes you're right. It would be better to have a work-around, To use "tricks", because what was done, is done. period! The point is that these things were done unnecessarily and without consideration. ie: arbitrarily.
Tell me, what good is a 250 or 500 GB hard drive with only 14 partitions.
Oh, I can tell you why four 1 TB drives in a software RAID5 setup with essentially one partition makes sense :) This setup is used to mostly store TV recordings. For different Windows versions,
Makes sense to me for those "specialized" uses. That's not me though, nor many on this list.
run several different OS's (four) plus several different iterations of SuSE.
I do too, but I use virtualization and disk image files, serves most of my needs. Only situation that vm doesn't suffice is when I need good accelerated 3D graphics.
So what you are saying is that we should run our systems via the gospel of Phillipp.
I like to try them all and experiment/compare them. I always make separate partitions for /home, /tmp, /var, and /usr/local because I believe that is the smart way to do it.
There are pros and cons, but this IMHO mostly a matter of personal taste and thus not really something to discuss.
It is a matter for discussion if that is a matter of preference or personal taste. An arbitrary decision now dictates the way we must run our systems.
Now add the /swap and the /master partition and start to multiply.
/swap can be shared between different Linuxes (unless you want to be able to suspend to disc in all Linux versions).
As a matter of fact I do share swap between some OS's. Now we are nit-picking.
And out of interest, what's /master for?
OK the "/" partition. The one where you install everything.
Please don't get into the Microsoft mode of thinking that the latest (or previous) SuSE is all that you need or may want. Please don't be telling us what is best for us.
I did none of that, so why such a defensive mode?
No you haven't directly. It is your opinions which are espoused to be the truth revealed. You are the one who is defensive of some bad or unpopular decisions. I have been on this list for many many years. You have never seen me arbitrarily post my opinions as a rant for the sake of a rant. You are responding to something I posted as a problem with an explanation of the problem. Am I upset about it? Yes, of course, from the very beginning, over something that was arbitrary and uncalled for.
Linux is about freedom.
Yes, but that includes the freedom of those that put together a distribution as to what they support. Linux, for instance, supports a wide range of architectures, yet openSUSE is only offered for a small subset.
And I agree that the devs should have the freedom to make decisions, after all they are the ones doing the work. And their work is appreciated by me and many others, but Ido think that they should consider the consequences of those decisions,or at least seek input from the community, especially with something as radical as this. Maybe not radical to you, but radical to many in the community. That's all I have to say. No more from me about this. Bob S -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* Bob S (911@sanctum.com) [20080718 04:24]:
The point is that these things were done unnecessarily and without consideration. ie: arbitrarily.
They definitely weren't done unnecessarily and I doubt they were done without consideration! libATA was developed to offer a ATA driver API, class transports for ATA and ATAPI devices, and SCSI/ATA translation for ATA devices. It seems that the SCSI limitations didn't concern the developers or weren't seen critical. After a time, the upstream kernel developers decided to not only support SATA controllers but also ATA ones, I guess in order to unify the API and concentrate maintenance on one set of drivers. So if you want to protest, do it upstreams on lkml.
So what you are saying is that we should run our systems via the gospel of Phillipp.
No, I don't do that. I simply stated how I manage it. If you keep on trying to second guess me, we can stop this discussion here and now.
An arbitrary decision now dictates the way we must run our systems.
It's *not* arbirary!
over something that was arbitrary and uncalled for.
I wouldn't call it either arbitrary or uncalled. Something like libata was needed and for understandable reasons they choose to make (S/P)ATA devices appear like SCSI devices as SCSI and ATA share quite a few details. Heck, even ATAPI was mostly copied from the SCSI specs, just using a different physical connection.
or at least seek input from the community, especially with something as radical as this.
I'd really be interested in a survey as to how many people really do need/use more then those 16 partitions! Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On 2008/07/18 12:43 (GMT+0200) Philipp Thomas apparently typed:
I'd really be interested in a survey as to how many people really do need/use more then those 16 partitions!
14, not 16: Only 0-15 fit within the SCSI major/minor dev constraint, of which 0 is the whole disk, and one of 1-4 is the extended, which also can host no filesystem. Under this roof there are 15 functional systems with more than 14 partitions (up to 67 per, most with only 1 HD), 10 with fewer than 15 (some with more than 1 HD), and 4 USB/eSATA disks used to backup with up to 74 each. LVM is used on 1 system used mainly for testing LVM matters. -- "Do not let the sun go down while you are still angry." Ephesians 4:26 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Friday 2008-07-18 at 12:43 +0200, Philipp Thomas wrote:
* Bob S () [20080718 04:24]:
The point is that these things were done unnecessarily and without consideration. ie: arbitrarily.
They definitely weren't done unnecessarily and I doubt they were done without consideration! libATA was developed to offer a ATA driver API, class transports for ATA and ATAPI devices, and SCSI/ATA translation for ATA devices. It seems that the SCSI limitations didn't concern the developers or weren't seen critical. After a time, the upstream kernel developers decided to not only support SATA controllers but also ATA ones, I guess in order to unify the API and concentrate maintenance on one set of drivers.
Scsi didn't traditionally need more partitions because the disks were smaller.
So if you want to protest, do it upstreams on lkml.
I wouldn't know where. Do they have a bugzilla?
An arbitrary decision now dictates the way we must run our systems.
It's *not* arbirary!
Yes, it is arbitrary. The HD hardware/firmware can hold thousands of partitions, there is no limit. The limitation is in software, in the way used to distribute the major/minor number over 8 bits in unix/linux, or in the available letters of the alphabet in windows. That is arbitrary by definition. Like "1 megabyte is enough". What will you do with peta storage, put it all in the same partition?
I'd really be interested in a survey as to how many people really do need/use more then those 16 partitions!
Yes, me too. (And it is not 16, it is 14) Novell sometimes does surveys. They did one on disks not long ago, and they forgot to ask this question. Now they are doing another one on 11.0, and again some of the questions are wrong. They should better post a sample of the intended survey, ask for comments, adjust the questions properly, then go ahead with it. And create the survey in several languages, too. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIgJXutTMYHG2NR9URAmqgAJ0UheTarfOMOluoEZcBTtWwsWeUGwCglvCx 8FcHhcqK6qU2x3iCneaue4c= =5ktW -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* Carlos E. R. (robin.listas@telefonica.net) [20080718 15:09]:
So if you want to protest, do it upstreams on lkml.
I wouldn't know where. Do they have a bugzilla?
I think not. I deed mean the linux-kernel mailing list.
It's *not* arbirary!
Yes, it is arbitrary. The HD hardware/firmware can hold thousands of partitions, there is no limit.
AFAIR, the PC BIOS specification only knows of a maximum of 8 boot devices (0x80-0x87), that's also arbitrary.
The limitation is in software, in the way used to distribute the major/minor number over 8 bits in unix/linux, or in the available letters of the alphabet in windows. That is arbitrary by definition. Like "1 megabyte is enough".
Seems like we're talking past each other :( I took it to mean that the decision to use libata was done arbitrarily. I'll agree that the partition limit is kind of arbitrarily. But every scheme has its limitations, some hardware, some software. I didn't defend the limit for partions on SCSI devices, which is indeed arbitrarily choosen.
What will you do with peta storage, put it all in the same partition?
I don't know. Though it'll probably still be some rather distant time when disks reach that capacity (then again, I could be wrong as I wouldn't have expected 1 TB (*not* 1 TiB) disks so soon.
I'd really be interested in a survey as to how many people really do need/use more then those 16 partitions!
Yes, me too.
(And it is not 16, it is 14)
Like Felix said, it's 0-15 with only 14 usable, I should have been more precise.
They should better post a sample of the intended survey, ask for comments, adjust the questions properly, then go ahead with it. And create the survey in several languages, too.
That's IMO a *very* good question to ask - respecticely proposal to make - over on the opensuse-project mailing list. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On 2008/07/18 15:36 (GMT+0200) Philipp Thomas apparently typed:
* Carlos E. R. (robin.listas@telefonica.net) [20080718 15:09]:
So if you want to protest, do it upstreams on lkml.
I wouldn't know where. Do they have a bugzilla?
I think not.
Of course it does: http://bugzilla.kernel.org/
It's *not* arbirary!
Yes, it is arbitrary. The HD hardware/firmware can hold thousands of partitions, there is no limit.
AFAIR, the PC BIOS specification only knows of a maximum of 8 boot devices (0x80-0x87), that's also arbitrary.
I'm pretty sure his use of arbitrary was meant to apply to the decision to terminate direct kernel support for previously supported devices. With libata, that support must now come from userland in the form of dm/kpartx, but it seems that doesn't yet work for any but a select few developers, https://bugzilla.novell.com/show_bug.cgi?id=397816 being one manifestation of failure by mere mortal users. -- "Do not let the sun go down while you are still angry." Ephesians 4:26 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Friday 2008-07-18 at 09:53 -0400, Felix Miata wrote:
On 2008/07/18 15:36 (GMT+0200) Philipp Thomas apparently typed:
* Carlos E. R. () [20080718 15:09]:
So if you want to protest, do it upstreams on lkml.
I wouldn't know where. Do they have a bugzilla?
I think not.
Of course it does: http://bugzilla.kernel.org/
I'll have to think of it. If there is such a feature request I'll add my vote. I'm not happy about opening one myself there, as I don't know the "language", and I don't test kernels. I would rather prefer suse to represent its users in all flavors.
It's *not* arbirary!
Yes, it is arbitrary. The HD hardware/firmware can hold thousands of partitions, there is no limit.
AFAIR, the PC BIOS specification only knows of a maximum of 8 boot devices (0x80-0x87), that's also arbitrary.
I'm pretty sure his use of arbitrary was meant to apply to the decision to terminate direct kernel support for previously supported devices.
Exactly. We had 64 partitions, now we have 16 (14). And the decission to support 8, 16, 255 whatever partitions is also arbitrary. If a limit must be had, then choose one more than big enough. It is only big enough to those that do not use partitions. We should have a large limit on the number of disks and the number of partitions. Freedom.
With libata, that support must now come from userland in the form of dm/kpartx, but it seems that doesn't yet work for any but a select few developers, https://bugzilla.novell.com/show_bug.cgi?id=397816 being one manifestation of failure by mere mortal users.
- -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIgKoTtTMYHG2NR9URAjXSAKCOP+DK8IOzTya7iewSa7eDsOq8WwCfXjh4 UFipwtIERX1wspuKdkQ/TRo= =/KTg -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, Jul 18, 2008 at 10:34 AM, Carlos E. R.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
The Friday 2008-07-18 at 09:53 -0400, Felix Miata wrote:
On 2008/07/18 15:36 (GMT+0200) Philipp Thomas apparently typed:
* Carlos E. R. () [20080718 15:09]:
So if you want to protest, do it upstreams on lkml.
I wouldn't know where. Do they have a bugzilla?
I think not.
Of course it does: http://bugzilla.kernel.org/
I'll have to think of it. If there is such a feature request I'll add my vote. I'm not happy about opening one myself there, as I don't know the "language", and I don't test kernels. I would rather prefer suse to represent its users in all flavors.
If you want to track it a couple key patch series are: http://lkml.org/lkml/2008/7/3/167 http://kerneltrap.org/mailarchive/linux-kernel/2008/7/14/2453374 Note the dates are very recent. I have not seen anyone respond to these with meaningful ack / nack / question type responses. Greg -- Greg Freemyer Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer First 99 Days Litigation White Paper - http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* Felix Miata (mrmazda@ij.net) [20080718 15:53]:
I'm pretty sure his use of arbitrary was meant to apply to the decision to terminate direct kernel support for previously supported devices.
But that decision was made upstreams and we try to be as close as possible to the mainstream kernel in order to minimize the maintenance effort.
libata, that support must now come from userland in the form of dm/kpartx, but it seems that doesn't yet work for any but a select few developers, https://bugzilla.novell.com/show_bug.cgi?id=397816 being one manifestation of failure by mere mortal users.
Lets begin by stating that developers are mere mortals too :) And I didn't know it still doesn't work :( If that is so, there currently is no solution and that is indeed a bad thing. So Users with such disks have to stay with older distributions, which I'd have to agree isn't really satisfying. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On 2008/07/18 19:03 (GMT+0200) Philipp Thomas apparently typed:
* Felix Miata (mrmazda@ij.net) [20080718 15:53]:
I'm pretty sure his use of arbitrary was meant to apply to the decision to terminate direct kernel support for previously supported devices.
But that decision was made upstreams and we try to be as close as possible to the mainstream kernel in order to minimize the maintenance effort.
Granted, the more enlightened among us understand it was purely kernel dev fiat over which SUSE/Novell had no control
libata, that support must now come from userland in the form of dm/kpartx, but it seems that doesn't yet work for any but a select few developers, https://bugzilla.novell.com/show_bug.cgi?id=397816 being one manifestation of failure by mere mortal users.
Lets begin by stating that developers are mere mortals too :) And I didn't know it still doesn't work :( If that is so, there currently is no solution and that is indeed a bad thing.
I think Hannes is on track to have us a fix before long. I thought he had it 3 days ago, but either he or I screwed up, and I'd be little surprised if it was me.
So Users with such disks have to stay with older distributions, which I'd have to agree isn't really satisfying.
For the time being, that's not actually necessary. Unlike SMBFS, legacy IDE didn't get stripped from kernel sources prior to replacement by an as yet uncapable successor. So, some more enlightened distros, like *SUSE*, retain the legacy drivers in their release kernels, and make it possible at install time to prefer them to libata. I think most partition limit complaints are from those who don't catch that fact in the release notes, those who pick an unenlightened distro (Fedora & Ubuntu among them), and those with newer systems for which legacy IDE is not compatible with their hardware. -- "Do not let the sun go down while you are still angry." Ephesians 4:26 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, 18 Jul 2008 13:25:52 -0400, Felix Miata wrote:
Granted, the more enlightened among us understand it was purely kernel dev fiat over which SUSE/Novell had no control
Either I'm missing a smiley or it's unnecessary sarcasm. Do you really think, that SUSE/Novell kernel devs alone can block a decision when the majority of the rest wants it? I won't deny we do have influence, but I think it's far below what's needed for such a decision. And there are other examples, like stopping to export the LSM (Linux Security Modules) interface.
I think Hannes is on track to have us a fix before long.
I think so too.
Unlike SMBFS, legacy IDE didn't get stripped from kernel sources prior to replacement by an as yet uncapable successor.
That won't help you if you have SATA discs only supported by libata, will it? And it's those I had in mind when writing 'no solution'. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Unlike SMBFS, legacy IDE didn't get stripped from kernel sources prior to replacement by an as yet uncapable successor.
That won't help you if you have SATA discs only supported by libata, will it? And it's those I had in mind when writing 'no solution'.
Philipp
I believe every SATA disk is supported via the old drivers/ide if you have the right controller and controller config. The problem is sata controllers. Most onboard motherboard controllers offer legacy mode. In that mode they use the a traditional hardware interface and drivers/ide works. I don't think any of the add-in card controllers offer legacy mode. FYI: To get optimum performance you likely need to use the newer sata specific modes that do require libata. (ie. AHCI mode is pretty common, but libata only.) Greg -- Greg Freemyer Litigation Triage Solutions Specialist http://www.linkedin.com/in/gregfreemyer First 99 Days Litigation White Paper - http://www.norcrossgroup.com/forms/whitepapers/99%20Days%20whitepaper.pdf The Norcross Group The Intersection of Evidence & Technology http://www.norcrossgroup.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Friday 18 July 2008 15:08:52 Carlos E. R. wrote:
Novell sometimes does surveys. They did one on disks not long ago, and they forgot to ask this question. Now they are doing another one on 11.0, and again some of the questions are wrong.
I forwarded all feedback to the team who was making the survey.
They should better post a sample of the intended survey, ask for comments, adjust the questions properly, then go ahead with it. And create the survey in several languages, too.
Sounds like a good plan. -- with kind regards, Martin Lasarsch, Core Services SUSE LINUX Products GmbH, Maxfeldstr. 5 90409 Nürnberg GF: Markus Rex, HRB 16746 (AG Nürnberg) martin.lasarsch@suse.de - http://www.opensuse.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Thursday 2008-07-24 at 15:46 +0200, Martin Lasarsch wrote:
On Friday 18 July 2008 15:08:52 Carlos E. R. wrote:
Novell sometimes does surveys. They did one on disks not long ago, and they forgot to ask this question. Now they are doing another one on 11.0, and again some of the questions are wrong.
I forwarded all feedback to the team who was making the survey.
Thanks. I intended to write about this in the project list, but I kept forgetting. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIiJ3RtTMYHG2NR9URAjWHAJ9qHSL61IVbtatYituBA0qsA2VaHQCcDCpn 3yHVCPUinbeQqaor312VPJQ= =RFfA -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
* Philipp Thomas
I'd really be interested in a survey as to how many people really do need/use more then those 16 partitions!
I have 31 -- Patrick Shanahan Plainfield, Indiana, USA HOG # US1244711 http://wahoo.no-ip.org Photo Album: http://wahoo.no-ip.org/gallery2 Registered Linux User #207535 @ http://counter.li.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On 2008/07/18 19:06 (GMT+0200) Philipp Thomas apparently typed:
* Patrick Shanahan (paka@opensuse.org) [20080718 16:23]:
I have 31
That makes it three users (Bob, Felix and you).
If you expect a reasonable response to a topic, the summary must match the topic. Probably Carlos and others haven't yet been clued as to the actual current topic in this thread. Also, hardly all affected users are subscribers to this list, or even use SUSE. Try some Googling for multiboot >15 partitions or libata partition limit or similar if you want a more realistic sample of the number of affected users. -- "Do not let the sun go down while you are still angry." Ephesians 4:26 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On Fri, 18 Jul 2008 13:35:03 -0400, Felix Miata wrote:
If you expect a reasonable response to a topic, the summary must match the topic.
How about once assuming tongue-in-cheek? But I admit I did forget a suitable emoticon on that, sorry 'bout that. Philipp -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
On 2008/07/18 19:06 (GMT+0200) Philipp Thomas apparently typed:
* Patrick Shanahan (paka@opensuse.org) [20080718 16:23]:
I have 31
That makes it three users (Bob, Felix and you).
The count should be systems affected, not people. Not everyone is responsible for a single system. -- "Do not let the sun go down while you are still angry." Ephesians 4:26 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Friday 2008-07-18 at 19:06 +0200, Philipp Thomas wrote:
* Patrick Shanahan () [20080718 16:23]:
I have 31
That makes it three users (Bob, Felix and you).
You forget me, and many others that do not read this list (or any at all), or this particular thread. There are hundreds of thousands of people using Linux, hopefully millions, and there are hardly thousands of different subscribers here. We are simply a qualified minority :-) But then, there are many minorities that use Linux thinking that Linux supports all classes of old hardware in the kernel, contrary to the competition that forces users to have the newest hardware. - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFIgPaatTMYHG2NR9URArpBAJ9W/F2Io9dM+ZwH/p+hEpZCk+XmUACeNsjw JvDuryhe9iIyLQjX9xWN0aw= =BZKB -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
participants (9)
-
Bob S
-
Carlos E. R.
-
Felix Miata
-
Greg Freemyer
-
jdd sur free
-
Martin Lasarsch
-
Patrick Shanahan
-
Philipp Thomas
-
Philipp Thomas