diff options
author | Kent Overstreet <kent.overstreet@gmail.com> | 2018-12-04 11:07:22 -0500 |
---|---|---|
committer | Kent Overstreet <kent.overstreet@gmail.com> | 2018-12-04 11:07:22 -0500 |
commit | 0fa30f074e7f6b34b2248717915efa376abd2556 (patch) | |
tree | 217f38c3883cfbe0ea00fd985e7a15a52f8f7783 | |
parent | e62e43e76e832a9f28327e781e0338403d323ace (diff) |
Status update
-rw-r--r-- | index.mdwn | 29 |
1 files changed, 18 insertions, 11 deletions
@@ -99,15 +99,18 @@ necessary, but I'm not expecting any more incompatible disk format changes. - Full data checksumming - Fully supported and enabled by default. We do need to implement scrubbing, - once we've got replication and can take advantage of it. + Fully supported and enabled by default; checksum errors will cause IOs to be + retried if there's another replica available. - Compression - Not _quite_ finished - it's safe to enable, but there's some work left - related to copy GC before we can enable free space accounting based on - compressed size: right now, enabling compression won't actually let you store - any more data in your filesystem than if the data was uncompressed + Done - LZ4, gzip and ZSTD are currently supported. ZSTD support unfortunately + still seems to be slightly buggy, but LZ4 is stable and well tested. + + - Multiple device support + + Done - you can add and remove devices at runtime while the filesystem is in + use, migrating data off the device if necessary. - Tiering/writeback caching: @@ -120,12 +123,16 @@ necessary, but I'm not expecting any more incompatible disk format changes. Basic caching functionality works, but it's not (yet) as configurable as bcache's caching (e.g. you can't specify writethrough caching). - - Replication + - Replication (i.e. RAID1/10) + + Done - you can yank out a disk while a filesystem is in use and it'll keep + working, transparently handling IO errors. You can then use the rereplicate + command to write out another copy of all the degraded data to another device. + + - Erasure coding - All the core functionality is complete, and it's getting close to usable: you - can create a multi device filesystem with replication, and then while the - filesystem is in use take one device offline without any loss of - availability. + Mostly done, but not quite usable yet. We still need a stripe level + compaction path. - [[Encryption]] |