I started xfs_repair a few days ago with the "-o ihash", "-o bhash", and "-o ag_stride" to help speed it up. While there was a drastic increase in it working through the different phases (compared to a previous attempt), it got to Phase 6 and is now outputting "rebuild AG headers and trees - 73 of 73 allocation groups done" every 15 minutes. That's message has been the same for just a little over 2 days now.

I've been watching atop and iostat and it seems to be doing something because the r/s, w/s, and rsec/s values (for sdc, sdc1, and dm-0) in iostat are decreasing as are the "busy" percentages for the raid partition and sdc listings in atop. I'm just unsure of what is actually happening if anything. strace hasn't really shown anything when watching the xfs_repair process. The only time it does is when it outputs that message noted above to a log file.

Any help or suggestions on getting more insight into what xfs_repair is doing, if anything, is greatly appreciated.