This content has been marked as final. Show 6 replies
Whether or not a reboot is required to add more storage space and whether or not device names change depend on your storage hardware and volume management software. For instance, are you using ASM, LVM, Multipath, Udev? There is no way to guess your environment. Many file systems can be resized while online, including Ocfs2 1.4 or ASM.
Am I correct to think that you want to resize a LUN that is presented to ASM as an ASM disc? If so, you could use the ALTER DISKGROUP....RESIZE DISK... command, but I have to say that I have never done this. Also, you will then end up with a group consisting of differtent sized discs, which is not optimal. Is not possible to crate a new LUN, and add it to the disc group?
if the second option is applicable. how to avoid LUNs renaming
what I did is that I created a new LUN and mapped it to the servers, now when I do fdisk -l the systems froze... I am afraid damaging the configuration if I reboot the servers and the device naming changes.
how to ensure the system will boot normally if I reboot?
I am sorry but I do not think that I can reliably advise you on this, other than to say that I have found that using either the ASMlib oracleasm utility or setting up udev rules will give device name persistance.
Remember that diskgroups are self describing. As long as Grid Infrastructure can discover the devices, either by interrogating the ASMlib or by reading the diskstring parameter from the profile.xml file, then ASM should be able to work out what is happening. If you are not using ASMlib, then I would set the asm_diskstring to a value that includes wildcards, so that it will still find all devices even if names change.
Perhaps someone else can give more definitive advice.