TopoLVM is quite aggressive at grabbing storage and this should be more explicitly advertised/configurable #707
-
Hi all, Here's what happened.. I am turning an unused MacOS Intel Mac Pro into a single node openshift. Since I wanted to keep dual-booting the machine, I proceeded that way:
At this point, I had a working Mac Pro with:
Then things went downhill from there:
seeing that, I edited lvmclusters/lvmcluster-sample and removed the 'datavg' section. Then I proceeded to re-image /dev/sda with the BM backup I had taken a few hours earlier. When I booted back into RHCOS, TopoLVM woke up again, carved a PV out of /dev/sda2 and added it to 'vg1', then it extended the thinpool onto /dev/sda2. For the next 30 minutes, I tried to take /dev/sda2 out of the LVM2 config (thin pools cannot be shrunk, unfortunately) so I ended up uninstalling the operator and tried to get rid of the "nvme0n1+/dev/sda2" 'vg1'. In the end, I ended up destroyin my SNO install, restored /dev/sda (MacOS) and installed a fresh SNO node with 4.12 and ToopoLVM (I will not try to change the name of the vg, this time...) That's only my 2c but the overall experience was that TopoLVM was quite aggressive and not too configurable:
I would love to see if:
|
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
Hi @ElCoyote27 |
Beta Was this translation helpful? Give feedback.
-
Thank you for your feedback. |
Beta Was this translation helpful? Give feedback.
Hi @ElCoyote27
I assume you used lvm-operator. If so, could you report this issue to Red Hat?
TopoLVM organization and project are NOT maintained by Red Hat. They developed lvm-operator and added topolvm to their suites solely.