Have problems whit creation VG tjderby / #vgcreate /dev/test1vg /dev/dsk/c5t8d0

[Gustavo Villafaña]

Pls some vary help me i have the next problem

tjderby / #pvcreate -f /dev/rdsk/c5t8d0
Physical volume "/dev/rdsk/c5t8d0" has been successfully created.
tjderby / #mkdir /dev/test1vg
tjderby / #mknod /dev/test1vg/group c 64 0x010000
tjderby / #vgcreate /dev/test1vg /dev/dsk/c5t8d0
vgcreate: Physical Volume "/dev/dsk/c5t8d0" contains no LVM information

tjderby / #vgcreate -f -s 32 /dev/test1vg /dev/dsk/c5t8d0
vgcreate: Physical Volume "/dev/dsk/c5t8d0" contains no LVM information

Tanks-

[Prashanth.D.S]

Hi,

Cause: pvcreate uses raw device file, vgcreate/vgextend block device file so the IO goes via buffer cache. A stale buffer was the cause of stale data being returned to read request of the primary LVM record from vgcreate/vgextend

Get rid of the stale data in the buffer cache by
#dd if= of=/dev/dsk/c1t2d0 bs=1024 count=1000

After this try pvcreate and vgcreate again this should resolve the issue.

Best Regards,
Prashanth

[Prashanth.D.S]

Use..

#dd if=/stand/vmunix /dev/dsk/c1t2d0 bs=1024 count=1000

Best Regards,
Prashanth

[Prashanth.D.S]

Sorry...change the device file

#dd if=/stand/vmunix /dev/dsk/c5t8d0 bs=1024 count=1000

[Gustavo Villafaña]

Hello
I execute this
tjderby / #dd if=/stand/vmunix /dev/dsk/c5t8d0 bs=1024 count=1000
dd: bad arg /dev/dsk/c5t8d0

and execute this
dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024k count=1000

but is just running
As much time as this process delays approximately, carries 5 minutes at a run and not yet finishes .

[Prashanth.D.S]

Hi Gustavo,

Abort using control+c run this dd command on the raw device file like mentioned below..

#dd if=/stand/vmunix of=/dev/rdsk/c5t8d0 bs=1024 count=10

Best Regards,
Prashanth

[Gustavo Villafaña]

The control + c is not working, How kill the processes dd, right now attempt with the command kill - 9 but nonstop the processes.

tjderby / #ps -fea | grep dd
root 24 0 0 May 30 ? 0:00 lvmschedd
root 6151 1 0 Jun 4 ? 0:00 dd if=/dev/dsk/c5t0d0 of=/dev/null
root 29900 29080 1 11:44:54 pts/th 0:00 grep dd
root 27297 1 0 11:33:28 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024k count=1000
root 17616 1 0 10:20:21 ? 0:00 dd if=/dev/dsk/c5t8d0 of=/dev/null bs=1024k count=10
root 28897 1 0 11:39:17 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024 count=1000

[Prashanth.D.S]

Ahh too many dd process have been started normally Ctrl C should get you back to the command prompt..

Wait for sometime and try killing the dd process one by one

#kill -9

[Gustavo Villafaña]

The ctrl + c not working and execute the kill -9
tjderby / #ps -fea | grep dd
root 24 0 0 May 30 ? 0:00 lvmschedd
root 6151 1 0 Jun 4 ? 0:00 dd if=/dev/dsk/c5t0d0 of=/dev/null
root 4034 3186 0 12:09:19 pts/ti 0:00 grep dd
root 27297 1 0 11:33:28 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024k count=1000
root 17616 1 0 10:20:21 ? 0:00 dd if=/dev/dsk/c5t8d0 of=/dev/null bs=1024k count=10
root 28897 1 0 11:39:17 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024 count=1000
tjderby / #kill -9 6151 27297 17616 28897
tjderby / #ps -fea | grep dd
root 24 0 0 May 30 ? 0:00 lvmschedd
root 6151 1 0 Jun 4 ? 0:00 dd if=/dev/dsk/c5t0d0 of=/dev/null
root 4042 3186 1 12:09:46 pts/ti 0:00 grep dd
root 27297 1 0 11:33:28 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024k count=1000
root 17616 1 0 10:20:21 ? 0:00 dd if=/dev/dsk/c5t8d0 of=/dev/null bs=1024k count=10
root 28897 1 0 11:39:17 ? 0:00 dd if=/stand/vmunix of=/dev/dsk/c5t8d0 bs=1024 count=1000
tjderby / #
But the process are running .
I have the same problem whit creation a new VG other
Some other thing to solve this problem?
With who else I can increase this problem, here in Mexico right now we yielded it, but the people of support could have given him solution.

[Prashanth.D.S]

Hi,

Cause for the issue is as mentioned earlier A stale buffer was the cause of stale data being returned to read request of the primary LVM record from vgcreate/vgextend ,

solution is to clear the headers and recreate the pv information.

Now since the dd command is hanging I would recommend you to wait for some time and check as it may take time. If you have a software support with HP i would suggest you to speak to them once.

If the dd command doesnot comeout then i would suggest you to check the hardware (disk) it can be flawed.

Best Regards,
Prashanth

[Gustavo Villafaña]

Have two arrays 20 Hd´s to 36 GB and other array to the problem is in this array to 20 HD's of 73 GB which no record I could have created VG for such motive he was needing help to check this, I have an operating system HPUX seeing 11,0 no himself if this version have some bug with this fellow of records ó what else need making a revision.
Saludos Cordiales