1

All, I appreciate in advance any useful responses to the following:

  1. One (1) 800GB LUN from external SAN was presented to 4 RHL servers (2.6.9-78.0.8 x86_64)
  2. Initial < fdisk -l > before reboot indicated no disk with < doesn't contain a valid partition table >
  3. After reboot, < fdisk -l > displays three (3) disks with < doesn't contain a valid partition table > /dev/sdo /dev/sdac /dev/sdlmmac [ with slightly different disk names on the other 3 RHL servers ]
  4. Checking /proc/partitions, I see all three of the above disk listings, each with 800GB
  5. When I did the same steps above (1-3) 4 months ago with a different clustered set (5 RHL servers) with another 800GB LUN step #3 only displayed one (1) disk with < doesn't contain a valid partition table > . I then fdisk formatted that disk on the 1st RHL server, confirmed on the other 4 RHL servers that the fdisk worked, and was then able to attach to my Oracle ASM for extra storage to all 5 instances of the clustered set.
  6. How should I handle what's occuring now with the display of the 3 disks (step #3 above)? Should I just fdisk format the /dev/sdlmmac ?
    Why did this happen? Is there a config issue with this 4 RHL set than with the other 5 RHL set that I worked with 4 months ago that could be causing this?

again, Thanks in advance Ed

EdLee
  • 11
  • 1
  • What volume manager are you using with the 4-node cluster? Is it configured the same way as the previous 5-node cluster? – mfinni May 28 '13 at 13:11
  • The LVM is the same version for both clusters. I just checked the /etc/lvm/lvm.conf files and they are the same. Is there a specific file that I should check other than the lvm.conf file? Sorry it took this long to reply, but I had to learn about the LVM to respond to your comment. – EdLee May 28 '13 at 17:21
  • I don't know much of the details to configuring LVM, so I can't provide much more info. Are you sure that this volume in question is being managed by LVM on the new 4-node cluster? – mfinni May 28 '13 at 17:23
  • Well, I brought up the LVM on both clusters and I can see all three (3) disks in the LVM display - with confirmation that they are 800GB. – EdLee May 28 '13 at 17:39
  • I mean, the 3 disks on the 4-node cluster – EdLee May 28 '13 at 17:40
  • mfinni - I have other fires to put out and this issue has just been put on the back burner due to low impact - I will probably just < fdisk create > one of the disks at a later time - I can always unpresent and retry this later. I will update. Thanks. – EdLee May 28 '13 at 17:54

0 Answers0