- Posted In
- Red Hat Gluster Storage
I have a thin pool that has the metadata 100% full. I have read support article: https://access.redhat.com/solutions/3251681 I also read lvmthin(7) man page.
They appear to disagree with each other AFAIK.
RH KB says: "However, if metadata has gone full causing the corruption, don't run thin_repair or lvconvert --repair because it's no use trying to repair this metadata into a similarly sized metadata LV. This may cause a corruption, and if this happens, the only way to then recover the metadata is open a new Bugzilla or use existing BZ1497178 created for these situations. Please open a support ticket with Red Hat Technical Support in order to open a new Bugzilla."
man page says:
Command to repair a thin pool: lvconvert --repair VG/ThinPoolLV
Repair performs the following steps:
1. Creates a new, repaired copy of the metadata. lvconvert runs the thin_repair command to read damaged metadata from the existing pool metadata LV, and writes a new repaired copy to the VG's pmspare LV.
I am comfortable with LVM but I am new to lvm thin. I have never needed to perform a repair of it before. What is the right thing to do here?