Operating System - HP-UX
1753724 Members
5065 Online
108799 Solutions
New Discussion

Re: /etc/lvmtab is out of date with the running kernel

 
atul2701
Frequent Advisor

Re: /etc/lvmtab is out of date with the running kernel

pls provide the output of below command also.

strings /etc/lvmtab
ioscan -fnC disk

atul
Atul Gupta
raghu_14
Advisor

Re: /etc/lvmtab is out of date with the running kernel

The one stale extent indicates the problem is either with the disk or IO card or cable fault.

Do the disk in question has multiple paths (PVlinks)? If in case one link has failed, then we can make this problem correct by configuring the VG with pvlinks. (vgextend vg pvlink1 pvlink2 etc)

what is the o/p of below :
lvdisplay -v /dev/vgxx/lvolxx
diskinfo
ioscan -fnC disk
pvdisplay disk_name | grep "PV status"
dd if=disk_name of=/dev/null bs=1024 count=64

May be you need to try reading the contents of the stale extent with the dd command specifically (by first finding the Physical Extent 0th location, and then skipping all the PE's till you reach the correct extent where it is showing stale). If the dd is successful then there is some issue with LVM. If there is some error, then you may need to go back check the IO card/ cable/disk. If the disk has gone bad, then you may need to follow the procedure to replace the bad disk http://bizsupport2.austin.hp.com/bc/docs/support/SupportManual/c01911837/c01911837.pdf

Hope this helps,
Raghu

Regards,
Raghu
A ship in the harbor is safe, but that is not what ships are built for.
dsausers
Frequent Visitor

Re: /etc/lvmtab is out of date with the running kernel

Sorry to revive such an old thread, but I came across this same problem, and it doesn't seem that the thread completely resolved it.  So I thought I'd put my solution in here for future folks searching the webs for this issue.  I found that running the following command cleaned up the erroneous disk the kernel thought was in the volume group:

 

# vgreduce -f {volumegroup}

 

I then recieved the following message:

PV with key 0 successfully deleted from vg {volumegroup}
Repair completed, please perform the following steps..:
1. vgscan -k -f {volumegroup}
2. vgcfgbackup {volumegroup}

 

I ran the commands and it resolved the issue.  My vgdisplay now shows 2 current PVs, and 2 active PVs.  However, be careful as a vgscan, I think prior to 11.31 would not scan in inactive VG's.  I'm running 11.31, so this was not a problem, but just beware.