On Thu, Nov 3, 2011 at 14:20, Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote: > XFS is just waiting for I/O to complete - xfs_bwrite is a very simple > funtion that synchronously submits I/O, and xfs_buf_iowait is called > to wait for the completion. This is a typical symptom for a broken > I/O subsystem. What kind of setup do you use? Especially hardware > RAID controllers tends to produce these kinds of issues. Oh yeah, I didn't mean to necessarily implicate XFS as such, more to say "I don't think that's a Ceph bug", look lower in the stack. I think I'm just likely to call out fs bugs at this point because usually with btrfs, that same hung message from the kernel tends to mean the fs broke, not the disk. Broken block device drivers and firmware happens too. Failure is always an option, at every layer of the stack. -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
