[sheepdog-users] [ANNOUNCE] sheepdog stable release v0.7.6-rc0

[Liu Yuan]

On Tue, Dec 17, 2013 at 06:06:02PM +0900, Hitoshi Mitake wrote:
> 
> Hi sheepdog users and developers,
> 
> I released v0.7.6-rc0 of stable branch. You can download a source
> archive from these URLs:
> tar.gz: https://github.com/sheepdog/sheepdog/archive/v0.7.6-rc0.tar.gz
> zip: https://github.com/sheepdog/sheepdog/archive/v0.7.6-rc0.zip
> 
> This update contains many important bugfixes and import some features helpful
> for managing long running clusters. The features are:
> 
> 1. rotate log when sheep daemon receives SIGHUP
> 2. runtime loglevel changing with "dog node log level set"
> 3. specify a directory for storing log file with -L option of sheep
> 4. repair broken objects based on majority voting with "dog vdi check"
> 5. tracking object with object id by a new option "-o" of "dog vdi track"
> 
> If no one disagrees about this release in 2 days, it will be v0.7.6 officialy.
> 
> Below is the summary of commits this update contains:
> 
> Hitoshi Mitake (9):
>       sheepfs: add -lpthread to the linker option
>       sheep, dog: runtime loglevel changing
>       sheep: rotate log with SIGHUP
>       dog: permit two or more depth of subcommands
>       dog: let usage printing adopt nested subcommands
>       dog: repair objects based on majority vote
>       tests/functional: add a new test for lost and corrupted objects
>       dog: add an option "-o" to vdi track for tracking object with oid
>       sheep: let init_vdi_state() return correct value
> 
> Liu Yuan (2):
>       md: remove unnecessary sd_err
>       sheep: check EAGAIN for all the sd_lock helpers
> 
> MORITA Kazutaka (1):
>       logger: remove unnecessary lock for log rotation
> 
> Ryusuke Konishi (2):
>       sheep: fix memory leak at create_client()
>       sheep: fix memory leaks at start_deletion()
> 
> Teruaki Ishizaki (1):
>       sheep: add an option for selecting the output directory of sheep.log
> 

I think you should add

- sheep: remove sha1_from_buffer()

when it is merged by master.

This fixes a fatal bug that breaks the data integrity for cluster snapshot.

Thanks
Yuan