From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.0.6) Gecko/20060728 Firefox/1.5.0.6 Description of problem: volume is a concat of two 250GB disks: Output of df: /dev/mapper/VolGroup00-LogVol00 225G 126G 88G 59% / Output of lvdisplay: --- Logical volume --- LV Name /dev/VolGroup00/LogVol00 VG Name VolGroup00 LV UUID N7X93V-p83L-ALvM-mSIV-LZvJ-uncW-ow6d6A LV Write Access read/write LV Status available # open 1 LV Size 464.69 GB Current LE 14870 Segments 3 Allocation inherit Read ahead sectors 0 Block device 253:0 Version-Release number of selected component (if applicable): 2.6.17-1.2174_FC5 How reproducible: Always Steps to Reproduce: Setup system with concatenated disks Actual Results: df seems to report only the first piece of the volume Expected Results: df should report complete volume size Additional info:
The sizes reported by lvdisplay and df do not need to coincide, first reports size of the volume, second size of the filesystem residing there. If you formatted the volume and then used lvextend on it, you still need to resize the filesystem. To verify, you can run (as root): blockdev --getsz /dev/mapper/VolGroup00-LogVol00 which gives size in 512 sectors, so divide by 2 to get size in kilobytes. If the actual block device size is different from what lvdisplay says, please reopen the bug. (You can also try writing data to the volume or reformatting it, which of course destroys the filesystem and all data).
The size of the filesystem was not changed.. It was created to fill the two disks but df reports it at HALF it's actual size.
In that case can you give blockdev --getsz output?
I knew someone would ask something like that, unfortunately we had to migrate this to solaris 10 w/ SDS - I'll try to build a copy of this machine to run this test on when I have some free time.