emo/thin-provisioning-tools
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[thin-repair, thin_dump] When repairing we now hunt for the best btree roots.

Browse Source 
We've had a trickle of users who accidentally activate the same pool on a
VM and host at the same time.  Typically the host doesn't do any IO, but
the kernel will still rewrite the superblock on shutdown.  This leaves
the superblock pointing to very out of date btree roots and so we get
massive metadata loss.

This patch changes thin_repair, and thin_dump --repair.  They now hunt
for the most recent, undamaged and consistent roots of the device and
mapping trees, and use that as the starting point of the repair.
This commit is contained in:
Joe Thornber
2019-04-17 12:17:13 +01:00
parent b027a1039f
commit 9e20465fd1
4 changed files with 560 additions and 362 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
thin-provisioning/thin_repair.cc
View File

@@ -22,14 +22,8 @@ namespace {
block_manager<>::ptr new_bm = open_bm(new_path, block_manager<>::READ_WRITE);
metadata::ptr new_md(new metadata(new_bm, metadata::CREATE, 128, 0));
emitter::ptr e = create_restore_emitter(new_md);
block_manager<>::ptr old_bm = open_bm(old_path, block_manager<>::READ_ONLY);
metadata::ptr old_md(new metadata(old_bm, false));
dump_options opts;
opts.repair_ = true;
metadata_dump(old_md, e, opts);
metadata_repair(old_bm, e);
} catch (std::exception &e) {
cerr << e.what() << endl;