[Gluster-devel] Help needed to implement Pause and Resume feature in Geo-replication

[Vijay Bellur]

On 04/02/2014 02:42 PM, Aravinda wrote:
> Hi All,

>
> As of now the idea is to stop processing on receiving pause signal(entry
> ops and rsync will stop eventually since processing is stopped) but
> crawling and identifying changes will continue. Sent initial
> patch(http://review.gluster.org/#/c/7322/) for the same.
>
> /Plan:/
> gluster cli will send SIGUSR1 to geo-rep monitor process, then monitor
> will send SIGUSR1 to all the worker processes.
> Worker processes uses os.pipe() and select to handle the signal received
> from monitor.

If gluster cli sends a signal to the geo-replication process, how do all 
the geo-rep monitor processes in the cluster get notified?
>
> /Problem:/
> Signal handling is not working in monitor. (No error/traceback), looks
> like python's limitation(http://bugs.python.org/issue5315)

I am not very conversant with signal handling in python but can we not 
use an unix domain socket or alternate IPC mechanisms to have some form 
of co-ordination between glusterd/gluster and the geo-replication
monitor process?

Regards,
Vijay