how to determine if /dev/... is an LV?
Hi all,
This should hopefully be simple. I am modifying some code to deal with LVM. If I am given a device path for an LV, say, /dev/VG/1, programatically, how can I determine that it actually is an LV and not some other kind of device? The best I have come up with is:
lvdisplay -D -c /dev/VG/lvol1 2>&1 | grep "^/dev/VG/lvol1" >/dev/null
and testing "$?".
Seems kind of "hacky" though. Is there a better way?
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
Thanx,
b.
Signature · 1 line
Brian J. Murrell
Well,
You could always check the major number of the device file with the one in /proc/devices, and see whether it is the major number of the lvm driver. Then check whether the minor number is that of a valid logical volume (not 0).
++Jos
And thus it came to pass that Brian J. Murrell wrote:
(on Wed, Aug 22, 2001 at 03:44:25AM -0400 to be exact)
Quoted 27 lines
Hi all,
This should hopefully be simple. I am modifying some code to deal with LVM. If I am given a device path for an LV, say,
/dev/VG/1, programatically, how can I determine that it actually is an LV and not some other kind of device? The best I have come up with is:lvdisplay -D -c
/dev/VG/lvol12>&1 | grep "^/dev/VG/lvol1" >/dev/nulland testing "$?".
Seems kind of "hacky" though. Is there a better way?
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
Thanx,
b.Signature · 7 lines
Brian J. Murrell _______________________________________________ linux-lvm mailing list [email protected] http://lists.sistina.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html
Signature · 1 line
Fear the Muppet! Winner of the HAL2001 Hacker Quiz...
Quoted 7 lines · José Luis Domingo López
Quoted 4 lines · Brian J. Murrell
Hi all,
[...]
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
[ snip ]
Quoted 1 line
Hope it helps. Regards,
Well I do apreciate your efforts (thanx!), but no, it doesn't. You told me everything I already knew but nothing of what I didn't. :-) What I mean is what is the relationship of the MINOR number to LVs.
b.
Signature · 1 line
Brian J. Murrell
Quoted 5 lines · Brian J. Murrell
Hi all,
[...]
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
As /usr/src/linux/Documentation/devices.txt says:
58 char Hayes ESP serial card - alternate devices
0 = /dev/cup0 Callout device for ttyP0
1 = /dev/cup1 Callout device for ttyP1
...
block Reserved for logical volume manager
109 char Reserved for logical volume manager
That is, LVM has block major number 58 reserved for its own use, as well as char major number 109. One of my machines has LVM installed, with a VG called Grupo00 and two LV, "mp3" and "usr" (ls -l /dev/Grupo00):
total 0
crw-r----- 1 root disk 109, 0 jun 4 12:33 group brw-rw---- 1 root disk 58, 0 jun 5 04:48 mp3 brw-rw---- 1 root disk 58, 1 jun 4 17:50 usr
Hope it helps. Regards,
Signature · 5 lines
Jos� Luis Domingo L�pez Linux Registered User #189436 Debian GNU/Linux Potato (P166 64 MB RAM) jdomingo EN internautas PUNTO org => � Spam ? Atente a las consecuencias jdomingo AT internautas DOT org => Spam at your own risk
Quoted 9 lines · Brian J. Murrell
Quoted 7 lines · José Luis Domingo López
Quoted 4 lines · Brian J. Murrell
Hi all,
[...]
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
Quoted 7 lines
[ snip ]
Quoted 1 line
Hope it helps. Regards,
Well I do apreciate your efforts (thanx!), but no, it doesn't. You told me everything I already knew but nothing of what I didn't. :-) What I mean is what is the relationship of the MINOR number to LVs.
# lvdisplay /dev/vg00/u1 --- Logical volume --- LV Name /dev/vg00/u1 <SNIP> Block device 58:0
The last line of the lvdisplay output contains the block major:minor numbers. Internally the LVM driver uses a minor number addressed table (today) to access a particular LVs in core meteadata.
Quoted 11 lines
b.
Signature · 7 lines
Brian J. Murrell _______________________________________________ linux-lvm mailing list [email protected] http://lists.sistina.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html
Signature · 15 lines
Regards,
Heinz -- The LVM Guy --
*** Software bugs are stupid.
Nevertheless it needs not so stupid people to solve them ***
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Heinz Mauelshagen Sistina Software Inc.
Senior Consultant/Developer Am Sonnenhang 11
56242 Marienrachdorf
Germany
[email protected] +49 2626 141200
FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Also, is there any relationship between the output from the lvdisplay and the minor number for the device?
Yes.
example from lvdisplay for lvol1:
Block device 58:0
major 58: minor:0
If you look at the devices for, you may notice, hey, they match!
crw-r----- 1 root disk 109, 0 Aug 23 21:00 group brw-rw---- 1 root disk 58, 0 Aug 23 21:04 lvol1 brw-rw---- 1 root disk 58, 1 Aug 23 21:03 lvol2
other correlation:
all minor numbers for char 109 (group files) are unique across the system. all minor numbers for block 58 (lvols) are unique across the system.
without looking at the source, when it comes to recovering/rebuilding vg info in the file system, I am assuming that lvm doesn't really care about device files when it is recording info into the vg headers (vgda & such), and since the advent of UUIDs, I doubt if it really cares about minor numbers either.
here is an test layout I created just for the heck of it.
I created vga across /dev/sda1 & sda4, vgb on sda2, vgc on sda3. I then lvcreated 3 times on vga,vgb,vgc in turn. [lvcreate -l 1 vg(a-c)] I then lvremoved vga/lvol1.
I then lvcreated 3 times on vga,vgb,vgc in turn. (again)
the lvcreate makes default names for the lvols.
/dev/vga:
total 46
dr-xr-xr-x 2 root root 161 Aug 23 21:04 . drwxr-xr-x 17 root root 46444 Aug 23 21:01 .. crw-r----- 1 root disk 109, 0 Aug 23 21:00 group brw-rw---- 1 root disk 58, 0 Aug 23 21:04 lvol1 brw-rw---- 1 root disk 58, 1 Aug 23 21:03 lvol2 brw-rw---- 1 root disk 58, 2 Aug 23 21:03 lvol3 brw-rw---- 1 root disk 58, 9 Aug 23 21:04 lvol4 brw-rw---- 1 root disk 58, 10 Aug 23 21:04 lvol5
interestingly enough, it recreated lvol1 here on the fourth lvcreate and started with the lowest minor number available (0), even though higher numbers had already been used.
/dev/vgb:
total 46
dr-xr-xr-x 2 root root 182 Aug 23 21:04 . drwxr-xr-x 17 root root 46444 Aug 23 21:01 .. crw-r----- 1 root disk 109, 1 Aug 23 21:00 group brw-rw---- 1 root disk 58, 3 Aug 23 21:03 lvol1 brw-rw---- 1 root disk 58, 4 Aug 23 21:03 lvol2 brw-rw---- 1 root disk 58, 5 Aug 23 21:03 lvol3 brw-rw---- 1 root disk 58, 11 Aug 23 21:04 lvol4 brw-rw---- 1 root disk 58, 12 Aug 23 21:04 lvol5 brw-rw---- 1 root disk 58, 13 Aug 23 21:04 lvol6
/dev/vgc:
total 46
dr-xr-xr-x 2 root root 182 Aug 23 21:04 . drwxr-xr-x 17 root root 46444 Aug 23 21:01 .. crw-r----- 1 root disk 109, 2 Aug 23 21:00 group brw-rw---- 1 root disk 58, 6 Aug 23 21:03 lvol1 brw-rw---- 1 root disk 58, 7 Aug 23 21:03 lvol2 brw-rw---- 1 root disk 58, 8 Aug 23 21:03 lvol3 brw-rw---- 1 root disk 58, 14 Aug 23 21:04 lvol4 brw-rw---- 1 root disk 58, 15 Aug 23 21:04 lvol5 brw-rw---- 1 root disk 58, 16 Aug 23 21:04 lvol6
This was all gone on a SuSE 7.1 system with the stock 2.2.18 kernel. LVM 0.9 13/11/2000 (IOP 10)
Lost in Tokyo,
Keith
P.S. My, what a big e-mail address you have Grandma!
Quoted 1 line · Brian J. Murrell
lvdisplay -D -c
/dev/VG/lvol12>&1 | grep "^/dev/VG/lvol1" >/dev/null
lvdisplay -D -c /dev/VG/lvol1 2>&1 | grep -q ^/dev/VG/lvol1 && do whatever you want :-)
Signature · 4 lines
Harald Milz | [email protected] | "He was a modest, Linux New Media AG | phone +49 (0) 89 993411-20 | good-humored boy. It Stefan-George-Ring 24 | fax +49 (0) 89 993411-99 | was Oxford that made D-81929 Muenchen | http://www.linux-magazin.de | him insufferable."