I haven't investigated more than reading Chris Mason's post in the thread: "On our end, many of these Btrfs warts are getting solved. The 3.19 merge window fixes some very hard to find corruption problems that we've been chasing down, and Josef Bacik has developed a slick power-fail testing target that makes it much easier to prevent similar bugs in the future. 3.19 will also fix rare corruptions with block group removal, making both balance and the new auto-blockgroup cleanup feature much more reliable.
We've hit a few performance problems deploying Btrfs here at Facebook, and fixes for these are making it into upstream kernels. We've also now caught two storage cards returning either stale or corrupt data, and the Btrfs crcs saved us from replicating the bad copies out across the cluster." On Wed, Apr 22, 2015 at 10:54 AM, Robie Basak <robie.ba...@ubuntu.com> wrote: > On Wed, Apr 22, 2015 at 10:47:32AM -0400, Bryan Quigley wrote: >> >Is it really ready yet? For example, CoreOS reported having issues and >> >switched away in December/January: >> >> >https://groups.google.com/forum/m/#!topic/coreos-dev/NDEOXchAbuU >> >> That's what the previous UDS sessions discussed. AFAICT from that >> list the btrfs devs have fixed the issues identified. > > The discussion dates from December 2014, which is after our most recent > UDS. It seems to suggest the issues still existed at the time. Are you > saying that the issues are fixed and they hadn't noticed or didn't > consider this relevant to their discussion, or that the issues have been > fixed upstream this year? > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
