[MYRIAD-164] Myriad fails to start up again after Shutdown is called

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: Myriad 0.1.0
• Component/s: Scheduler
• Assignee: Swapnil Daingade
• Reporter: Aashreya Ravi Shankar
• Created: 03/Nov/15 01:24
• Updated: 05/Nov/15 18:58
• Resolved: 05/Nov/15 18:58

Description

Myriad does not start-up after shutdown is called.

Following is seen in the log :
I1029 12:07:08.327148 15886 group.cpp:313] Group process
(group(1)@10.10.101.118:33172) connected to ZooKeeper
I1029 12:07:08.327177 15886 group.cpp:787] Syncing group operations: queue size
(joins, cancels, datas) = (0, 0, 0)
I1029 12:07:08.327189 15886 group.cpp:385] Trying to create path '/mesos' in
ZooKeeper
I1029 12:07:08.328446 15909 detector.cpp:138] Detected a new leader: (id='0')
I1029 12:07:08.328730 15889 group.cpp:656] Trying to get
'/mesos/info_0000000000' in ZooKeeper
W1029 12:07:08.329212 15885 detector.cpp:444] Leading master
master@10.10.101.118:5050 is using a Protobuf binary format when registering
with ZooKeeper (info): this will be deprecated as of Mesos 0.24 (see
MESOS-2340)
I1029 12:07:08.329318 15885 detector.cpp:481] A new leading master
(UPID=master@10.10.101.118:5050) is detected
I1029 12:07:08.329453 15884 sched.cpp:254] New master detected at
master@10.10.101.118:5050
I1029 12:07:08.329644 15884 sched.cpp:264] No credentials provided. Attempting
to register without authentication
I1029 12:07:08.330493 15884 sched.cpp:819] Got error 'Completed framework
attempted to re-register'
I1029 12:07:08.330507 15884 sched.cpp:1625] Asked to abort the driver
I1029 12:07:08.331507 15884 sched.cpp:861] Aborting framework
'20151029-110656-1986333194-5050-10105-0001'

As the state-store is not being cleared up on showtown, when we start it again it is using the same framework ID which causes the above issue.