GlusterFS 3.7.20已经发布了。这是GlusterFS-3.7的一次常规发布,也是GlusterFS-3.7计划中的最后一次发布,GlusterFS-3.10计划在下个月发布,一旦该版本发布,GlusterFS-3.7就会进入EOL,停止支持。EOL时间表的任何改动都会通知社区。

GlusterFS-3.7.20的发布说明可以在github上找到

社区发布的tar包可以在https://www.gluster.org/download/找到。

CentOS平台上的Storage SIG包很快就会在centos-gluster37软件仓库中编译出来。

 

修复的bug

本次发布修复了一下bug

  • #1401032: OOM kill of nfs-ganesha on one node while fs-sanity test suite is executed.

  • #1411614: Fix the place where graph switch event is logged

  • #1411901: DHT doesn't evenly balance files on FreeBSD with ZFS

  • #1412890: Extra lookup/fstats are sent over the network when a brick is down.

  • #1415133: Improve output of "gluster volume status detail"

 

已知问题:

当然还有很多已知问题没有解决(后面确认这些问题基本都已经解决了,没有下面的这些已知问题,官方发布说明从3.7.18之后也没明确说明这些问题是否解决,有些遗憾,可能是有些问题可能是bug,但后续发现并没有这些问题,所以也就没有再提):

  • Commit b33f3c9, which introduces changes to improve IPv6 support in GlusterFS has been reverted as it exposed problems in network encryption, which could cause a GlusterFS cluster to fail operating correctly when management network encryption is used.

  • Network encryption has an issue which could sometimes prevent reconnections from correctly happening.

  • #1353856: Inode leak in fuse

  • #1379838: gluster missing gfid attribute, healing doesn't work

  • #1379178: split brain on file recreate during downed brick.

  • #1373618: remove-brick status shows 0 rebalanced files

  • #1373396: Misleading error messages on brick logs while creating directory (mkdir) on fuse mount

  • #1372729: Possible VSZ leak in glusterfsd brick process

 

总结说明:

       基本上来说,GlusterFS-3.7.20会是GlusterFS-3.7系列的最后一个发布版本,除非GlusterFS-3.10计划的二月份发布放鸽子(GlusterFS发布计划可见:)。显然,将GlusterFS-3.7.20当作是生产环境上的版本不是很妥,内存溢出等问题是大问题,特别是对线上环境影响很大,几乎不可接受。

      根据最新了解到的情况,3.7.20版本还是可以考虑线上部署使用的。如有其它问题,可以反馈至GlusterFS技术交流QQ群,大家共同交流进步。


参考原文:

https://kshlm.in/post/glusterfs-3-7-20/

https://github.com/gluster/glusterfs/blob/release-3.7/doc/release-notes/3.7.20.md