On 10/10/2017 09:53 AM, KM wrote:> Thanks for all of the input, not really sure what if anything I will do.? i was hoping it would be easy and i could just create a /boot in root, and copy the actual boot contents to it and use it.?? wishful thinking i guess.? just to give a complete picture here is the current partitioning on the server....in case anyone wants to say anymore.? Thanks in advance. > Filesystem??????????? Size? Used Avail Use% Mounted on > /dev/mapper/vg_bldsrv-lv_root > ?????????????????????? 50G?? 26G?? 22G? 55% / > tmpfs???????????????? 9.0G? 156K? 9.0G?? 1% /dev/shm > /dev/sda1????????????? 96M?? 33M?? 59M? 36% /boot > /dev/mapper/vg_bldsrv-lv_home > ????????????????????? 861G? 371G? 447G? 46% /homeYour root filesystem is in an LVM volume. CentOS 6 is still using GRUB legacy, which does not support /boot in LVM. For you, there really is no way around the messy and delicate process of shrinking and relocating a filesystem and the LVM volumes to make space for a larger /boot partition. Frankly, I would hesitate to do that in place on my own system, and I have quite a bit of experience with doing unspeakable things with LVM volumes. You really need to do that resizing in the context of moving everything to another disk. If it's a server that you don't want to take down for the time it takes for that procedure, you can do amazing things with pvmove while your system continues to run, but you still need another disk to hold those volumes temporarily. -- Bob Nichols "NOSPAM" is really part of my email address. Do NOT delete it.
On 10/10/2017 6:20 PM, Robert Nichols wrote:>> Filesystem Size? Used Avail Use% Mounted on >> /dev/mapper/vg_bldsrv-lv_root >> ??????????????????????? 50G?? 26G?? 22G? 55% / >> tmpfs???????????????? 9.0G? 156K? 9.0G?? 1% /dev/shm >> /dev/sda1????????????? 96M?? 33M?? 59M? 36% /boot >> /dev/mapper/vg_bldsrv-lv_home >> ?????????????????????? 861G? 371G? 447G? 46% /home > > Your root filesystem is in an LVM volume. CentOS 6 is still using GRUB > legacy, which does not support /boot in LVM.says up there, /boot is /dev/sda1, this is almost exactly the config of my C6 servers. -- john r pierce, recycling bits in santa cruz
On 10/10/2017 6:50 PM, John R Pierce wrote:>> Your root filesystem is in an LVM volume. CentOS 6 is still using >> GRUB legacy, which does not support /boot in LVM. > > > says up there, /boot is /dev/sda1, this is almost exactly the config > of my C6 servers.never mind, I realized after I sent this, you were talking about him MOVING his /boot to / -- john r pierce, recycling bits in santa cruz
On 10/10/2017 09:20 PM, Robert Nichols wrote:> For you, there really is no way around the messy and delicate process > of shrinking and relocating a filesystem and the LVM volumes to make > space for a larger /boot partition. Frankly, I would hesitate to do > that in place on my own system, and I have quite a bit of experience > with doing unspeakable things with LVM volumes. You really need to do > that resizing in the context of moving everything to another disk.Agreed.? If / and /home are on xfs you can't shrink anyway.? I'm not sure if ext4 can be shrunk while mounted (I seem to remember that it can't).> > If it's a server that you don't want to take down for the time it > takes for that procedure, you can do amazing things with pvmove while > your system continues to run, but you still need another disk to hold > those volumes temporarily. >As long as there is enough slack space in the volume group you can do this.? If there is no slack space you have real problems, especially with XFS (one reason I still use ext4 for many things, and one reason I never fill the volume group to 100%). I have done the pvmove and filesystem resize dance before, live, with the second hard disk attached via iSCSI.? The least fun piece is then resizing the /boot partition and its filesystem.? But I had enough slack space in the volume group.? What can be done here is unmounting /home, shrinking /home the appropriate amount, and then you have enough slack space to do the shrink and move (not fully live, but semi-live, and you can't have any logged-in users with open files in /home).? Shrinking from the end of the filesystem and pv is easy; shrinking from the beginning is hard and prone to errors.? (gparted and similar do the move of the end of a partition fine; moving the start is much much harder). However, if you can shrink enough from the end you can put /boot on the last partition on the disk instead of the first, although you will have to do some grub stanza editing to get rid of /dev/sda1 and replace with the appropriate device for the new /boot.? So you could shrink /home, shrink the pv, shrink the partition holding the pv (this is the risky part), then add a partition to the end of the disk for the new /boot.? If you've never done this sort of thing before you may want to get someone who has done this sort of thing to do it. Otherwise, if you feel at all uncomfortable doing this it may just be easier to pull a backup and reinstall.
I searched a bit to see if there is a way to upgrade from CentOS 7 directly to CentOS 8.? I found RHEL instructions but not CentOS. ? Although they probably should be/would be similar, the instructions I found enable a rhel repository to get the leap command, which I can't seem to do in CentOS. Does anyone know if you can do an upgrade yet.? I know they had been working on it in the past. also - when they say upgrade (for example on the rhel pages), is it in place meaning I can leave my files/data there, or is it strictly a way of installing the OS that is going to wipe out my files? Thanks in advance. KM