Re: Question about LVM and volume images on loop SOLVED
On Wed, Jan 14, 2009 at 09:24:15AM -0500, Brian McKee wrote: > On Tue, Jan 13, 2009 at 8:36 PM, Paul E Condon wrote: > > I do this because I want to > > store large files on a HD whose hardware interface > > limits file sizes to 4Gb, and I want to store larger > > files than 4Gb. ( The HD has 500Gb total capacity.) > > As the others have pointed out - that doesn't seem likely. A drive is > just a bit bucket that doesn't grok it's own contents. This is true, but it appears that the program, mount, does grok the contents, and refuses to mount a drive unless it finds something that it believes is sensible in the bits in that bucket. But see below: > > However, let's say for the sake of argument you don't want to have any > files over 4 gig. Why not shrink the existing partition and create a > series of 4 gig partitions, then use them for physical volumes in LVM. > Much simpler than getting loop devices in there etc. Last night I tried again to put an ext3 file system on the drive. This was something that I had been unable to do almost a year ago and had come to believe was impossible for various 'reasons'. Last night I succeeded. So the whole basis for my question dissolved. I don't know why I failed last year when I tried the obvious solution, I don't think I will ever know. I'm really sorry for having exercised so many people over my crazy question. I hope you thought it was an interesting puzzle and enjoyed thinking about it, but, truth to tell, I must have been doing something mistaken. Thanks, all, for your help, and I hope I haven't worn out my welcome. -- Paul E Condon pecon...@mesanetworks.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On Tue, Jan 13, 2009 at 8:36 PM, Paul E Condon wrote: > I do this because I want to > store large files on a HD whose hardware interface > limits file sizes to 4Gb, and I want to store larger > files than 4Gb. ( The HD has 500Gb total capacity.) As the others have pointed out - that doesn't seem likely. A drive is just a bit bucket that doesn't grok it's own contents. However, let's say for the sake of argument you don't want to have any files over 4 gig. Why not shrink the existing partition and create a series of 4 gig partitions, then use them for physical volumes in LVM. Much simpler than getting loop devices in there etc. Brian -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On Wed, Jan 14, 2009 at 12:27:08AM -0600, Boyd Stephen Smith Jr. wrote: > On Tuesday 13 January 2009, Paul E Condon wrote > about 'Re: Question about LVM and volume images on loop': > >On Tue, Jan 13, 2009 at 08:33:46PM -0600, Boyd Stephen Smith Jr. wrote: > >> On Tuesday 2009 January 13 19:36:22 Paul E Condon wrote: > >> >I do this because I want to > >> >store large files on a HD whose hardware interface > >> >limits file sizes to 4Gb, and I want to store larger > >> >files than 4Gb. ( The HD has 500Gb total capacity.) > >> > >> HD hardware interfaces do not know where files begin and end.[1] You > >> *might* be using a filesystem that limits you to 4Gb. If so, LVM won't > >> buy you anything and you'll have to switch filesystems. > > > >My hope is that I can write a bunch of 4G files mount them as 4G hard > > drives, and group them into a single 'volume group' whose total capacity > > is some large multiple of 4G. Is this impossible? > > It's probably a bad idea. BUT, it should be possible. > > It would be much better to use ext[234], jfs, xfs, or even reiserfs > directly on the drive. You say you have problems, but I'm using reiserfs > across a raid-0 on WD driver and reiserfs across a raid-1 on WD drives > without issues. Could you go into more details. I tried ext3 about a year ago and I could get it onto the disk. I couldn't remember the details of why it didn't work, so I decided to try again to gather some current data with which to confront this community, but --- IT WORKED! Thanks to all for provoking me into new action! -- Paul E Condon pecon...@mesanetworks.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On 01/13/09 23:18, Paul E Condon wrote: [snip] Yes, it's not hardware. Truth is it's vfat. I have had difficulty with Western Digital USB HD when I try to reformat tham for ext3, and they special software from WD to rewrite the vfat format. ? Doesn't Debian see it as an sd device? What errors do you get when you mke2fs the device? A software overlay seemed a good way to go. I'm still hoping. Just too overcomplicated. Sell that device to a Windows-using friend, and for $120 buy a cheap enclosure + OEM 1TBB hdd.) -- Ron Johnson, Jr. Jefferson LA USA "I am not surprised, for we live long and are celebrated poopers." -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On Tuesday 13 January 2009, Paul E Condon wrote about 'Re: Question about LVM and volume images on loop': >On Tue, Jan 13, 2009 at 08:33:46PM -0600, Boyd Stephen Smith Jr. wrote: >> On Tuesday 2009 January 13 19:36:22 Paul E Condon wrote: >> >I do this because I want to >> >store large files on a HD whose hardware interface >> >limits file sizes to 4Gb, and I want to store larger >> >files than 4Gb. ( The HD has 500Gb total capacity.) >> >> HD hardware interfaces do not know where files begin and end.[1] You >> *might* be using a filesystem that limits you to 4Gb. If so, LVM won't >> buy you anything and you'll have to switch filesystems. > >My hope is that I can write a bunch of 4G files mount them as 4G hard > drives, and group them into a single 'volume group' whose total capacity > is some large multiple of 4G. Is this impossible? It's probably a bad idea. BUT, it should be possible. It would be much better to use ext[234], jfs, xfs, or even reiserfs directly on the drive. You say you have problems, but I'm using reiserfs across a raid-0 on WD driver and reiserfs across a raid-1 on WD drives without issues. Could you go into more details. -- Boyd Stephen Smith Jr. ,= ,-_-. =. b...@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/ signature.asc Description: This is a digitally signed message part.
Re: Question about LVM and volume images on loop
On Tue, Jan 13, 2009 at 08:33:46PM -0600, Boyd Stephen Smith Jr. wrote: > On Tuesday 2009 January 13 19:36:22 Paul E Condon wrote: > >I do this because I want to > >store large files on a HD whose hardware interface > >limits file sizes to 4Gb, and I want to store larger > >files than 4Gb. ( The HD has 500Gb total capacity.) > > HD hardware interfaces do not know where files begin and end.[1] You *might* > be using a filesystem that limits you to 4Gb. If so, LVM won't buy you > anything and you'll have to switch filesystems. My hope is that I can write a bunch of 4G files mount them as 4G hard drives, and group them into a single 'volume group' whose total capacity is some large multiple of 4G. Is this impossible? -- Paul E Condon pecon...@mesanetworks.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On Tue, Jan 13, 2009 at 06:08:05PM -0800, Jeff D wrote: > On Tue, 13 Jan 2009, Paul E Condon wrote: > > > I want to allocate several large blank files on HD, > > and make them into virtual hard disk which I mount > > using loop. > > > > I then want to treat them as 'physical volumes' under > > LVM, and collect them into a single volume group. > > > > Then, I intend to define a few larger logical volumes > > on this 'volume group'. I do this because I want to > > store large files on a HD whose hardware interface > > limits file sizes to 4Gb, and I want to store larger > > files than 4Gb. ( The HD has 500Gb total capacity.) > > Maybe I'm missing something here, but how is the hardware interface > stopping you from writing files larger than 4G? That's just something I > haven't ran into. How is the drive formatted now? Can't you just format > the drive with ext3 or something else that supports large files? I don't > think that LVM is going to solve your problem here though. > Yes, it's not hardware. Truth is it's vfat. I have had difficulty with Western Digital USB HD when I try to reformat tham for ext3, and they special software from WD to rewrite the vfat format. A software overlay seemed a good way to go. I'm still hoping. -- Paul E Condon pecon...@mesanetworks.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: Question about LVM and volume images on loop
On Tuesday 2009 January 13 19:36:22 Paul E Condon wrote: >I do this because I want to >store large files on a HD whose hardware interface >limits file sizes to 4Gb, and I want to store larger >files than 4Gb. ( The HD has 500Gb total capacity.) HD hardware interfaces do not know where files begin and end.[1] You *might* be using a filesystem that limits you to 4Gb. If so, LVM won't buy you anything and you'll have to switch filesystems. -- Boyd Stephen Smith Jr. ,= ,-_-. =. b...@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/ [1] How could they? Each filesystem stores files differently. Also, mdadm and LVM between the physical drives and the filesystem would change that. signature.asc Description: This is a digitally signed message part.
Re: Question about LVM and volume images on loop
On Tue, 13 Jan 2009, Paul E Condon wrote: > I want to allocate several large blank files on HD, > and make them into virtual hard disk which I mount > using loop. > > I then want to treat them as 'physical volumes' under > LVM, and collect them into a single volume group. > > Then, I intend to define a few larger logical volumes > on this 'volume group'. I do this because I want to > store large files on a HD whose hardware interface > limits file sizes to 4Gb, and I want to store larger > files than 4Gb. ( The HD has 500Gb total capacity.) Maybe I'm missing something here, but how is the hardware interface stopping you from writing files larger than 4G? That's just something I haven't ran into. How is the drive formatted now? Can't you just format the drive with ext3 or something else that supports large files? I don't think that LVM is going to solve your problem here though. Jeff -- 8 out of 10 Owners who Expressed a Preference said Their Cats Preferred Techno. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Question about LVM and volume images on loop
I want to allocate several large blank files on HD, and make them into virtual hard disk which I mount using loop. I then want to treat them as 'physical volumes' under LVM, and collect them into a single volume group. Then, I intend to define a few larger logical volumes on this 'volume group'. I do this because I want to store large files on a HD whose hardware interface limits file sizes to 4Gb, and I want to store larger files than 4Gb. ( The HD has 500Gb total capacity.) My problem is that I can't figure out how to execute pvcreate on my 'image physical volumes'. If I mount the image, I think I should apply pvcreate to /dev/loopxx, but that doesn't work, giving a message that the volume is mounted. If I unmount the image, I get (blk0 is my file to be initialized) : big:/media/WDP-5/blks# pvcreate -f blk0 Device blk0 not found (or ignored by filtering). How can I associate a loop device node with a particular file, without mounting it? Or how can I get pvcreate to write whatever it puts on a phisical volume into a file on disk? Or how can I mount a loop device without specifing a fils system type? Or, etc.? TIA -- Paul E Condon pecon...@mesanetworks.net -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
