Bug 1062159

Description of problem:  monitor got killed in intermediate master in cascaded setup while syncing meta data with "ValueError: signal only works in main thread". 

Backtrace in geo-rep intermediate master log-file

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
[2014-02-06 17:51:14.20453] I [master(/bricks/brick6):438:crawlwrap] _GMaster: crawl interval: 60 seconds
[2014-02-06 17:52:10.319305] E [syncdutils(monitor):240:log_raise_exception] <top>: FAIL: 
Traceback (most recent call last):
  File "/usr/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 266, in twrap
    tf(*aa)
  File "/usr/libexec/glusterfs/python/syncdaemon/monitor.py", line 190, in wmon
    cpid, _ = self.monitor(w, argv, cpids)
  File "/usr/libexec/glusterfs/python/syncdaemon/monitor.py", line 161, in monitor
    self.terminate()
  File "/usr/libexec/glusterfs/python/syncdaemon/monitor.py", line 89, in terminate
    set_term_handler(lambda *a: set_term_handler())
  File "/usr/libexec/glusterfs/python/syncdaemon/syncdutils.py", line 332, in set_term_handler
    signal(SIGTERM, hook)
ValueError: signal only works in main thread
[2014-02-06 17:52:10.341447] I [syncdutils(monitor):192:finalize] <top>: exiting.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Version-Release number of selected component (if applicable):glusterfs-3.4.0.59rhs-1


How reproducible: This happens often.


Steps to Reproduce:
1.create a geo-rep relationship in a cascaded setup. ( master->imaster->slave)
2.create data on master
3.start geo-rep relation ship between master and imaster.
4.wait for it to sync to imaster,
5.after it syncs completely to imaster, start geo-rep between imaster and slave
6.Check if the monitor gets killed, if not try the above steps again. 

Actual results: monitor got killed in intermediate master with "ValueError: signal only works in main thread"


Expected results: Monitor should never be killed. 


Additional info: