At Sat, 31 May 2014 23:30:00 +0900, Hitoshi Mitake wrote: > > > Hi sheepdog users and developers, > > I released v0.8.2_rc0 of stable branch. You can download a source > archive from these URLs: > tar.gz: https://github.com/sheepdog/sheepdog/archive/v0.8.2_rc0.tar.gz > zip: https://github.com/sheepdog/sheepdog/archive/v0.8.2_rc0.zip > > Main topics of this release are: > - remove a serious bug in startup recovery of crashed cluster > - bugfix for majority voting algorithm of "dog vdi check" > - various bugfixes in command line processing of dog command > > If no one disagrees about this release candidate in 2 days, it will be > v0.8.2 officially.
I'll delay the release of v0.8.2. Because the bug: https://bugs.launchpad.net/sheepdog-project/+bug/1317755 is really serious. The recent posted patch (see the thread entitled [PATCH] sheep: don't recycle VDI ID) can fix the problem. After the patch is merged to master and backported to stable-0.8, I'll release v0.8.2_rc1. Thanks, Hitoshi > > Below is the summary of the commits: > > Hitoshi Mitake (2): > dog: forbid creating snapshot of hypervolume > dog: use nr_copies of inode object instead of command line parameter > > Robin Dong (3): > sheep/cluster: use sema_t to replace pthread_cont_t > sheep/recovery: allocate old vinfo by using sys->cinfo > sheep: add test case for recovery from failed cluster > > Ruoyu (5): > dog/vdi: bug fix for voting majority, add test case to test it > sheep/object_cache: fix potential crash in object_cache_remove > sheep/ops: fix potential type conversion error > lib/logger: fix potential type conversion error > dog/node: add raw output support for dog node md info > > Thanks, > Hitoshi -- sheepdog mailing list [email protected] http://lists.wpkg.org/mailman/listinfo/sheepdog
