| 2015-08-26 01:05:04 |
clayg |
bug |
|
|
added bug |
| 2015-08-26 01:05:35 |
clayg |
description |
This is related to the fix for lp bug #1335581
I think eventually the FakeRing behavior got dropped in I8bfc388a04eff6491038991cdfd7686c9d961545
But the comment remains:
class FakeRing(Ring):
def __init__(self, replicas=3, max_more_nodes=0, part_power=0,
base_port=1000):
"""
:param part_power: make part calculation based on the path
If you set a part_power when you setup your FakeRing the parts you get
out of ring methods will actually be based on the path - otherwise we
exercise the real ring code, but ignore the result and return 1.
"""
self._base_port = base_port
self.max_more_nodes = max_more_nodes
self._part_shift = 32 - part_power
# 9 total nodes (6 more past the initial 3) is the cap, no matter if
# this is set higher, or R^2 for R replicas
self.set_replicas(replicas)
self._reload()
All that jazz about part power and doing something trixy in _get_parts should be pulled out. |
This is related to the fix for lp bug #1335581
I think eventually the FakeRing behavior got dropped in I8bfc388a04eff6491038991cdfd7686c9d961545
But the comment remains:
class FakeRing(Ring):
def __init__(self, replicas=3, max_more_nodes=0, part_power=0,
base_port=1000):
"""
:param part_power: make part calculation based on the path
If you set a part_power when you setup your FakeRing the parts you get
out of ring methods will actually be based on the path - otherwise we
exercise the real ring code, but ignore the result and return 1.
"""
self._base_port = base_port
self.max_more_nodes = max_more_nodes
self._part_shift = 32 - part_power
# 9 total nodes (6 more past the initial 3) is the cap, no matter if
# this is set higher, or R^2 for R replicas
self.set_replicas(replicas)
self._reload()
All that jazz about part power and doing something trixy in _get_parts should be pulled out. |
|
| 2015-08-26 01:10:10 |
clayg |
description |
This is related to the fix for lp bug #1335581
I think eventually the FakeRing behavior got dropped in I8bfc388a04eff6491038991cdfd7686c9d961545
But the comment remains:
class FakeRing(Ring):
def __init__(self, replicas=3, max_more_nodes=0, part_power=0,
base_port=1000):
"""
:param part_power: make part calculation based on the path
If you set a part_power when you setup your FakeRing the parts you get
out of ring methods will actually be based on the path - otherwise we
exercise the real ring code, but ignore the result and return 1.
"""
self._base_port = base_port
self.max_more_nodes = max_more_nodes
self._part_shift = 32 - part_power
# 9 total nodes (6 more past the initial 3) is the cap, no matter if
# this is set higher, or R^2 for R replicas
self.set_replicas(replicas)
self._reload()
All that jazz about part power and doing something trixy in _get_parts should be pulled out. |
This is related to the fix for lp bug #1335581
I think eventually the FakeRing behavior got dropped in I8bfc388a04eff6491038991cdfd7686c9d961545
But the comment remains:
{{{
class FakeRing(Ring):
def __init__(self, replicas=3, max_more_nodes=0, part_power=0,
base_port=1000):
"""
:param part_power: make part calculation based on the path
If you set a part_power when you setup your FakeRing the parts you get
out of ring methods will actually be based on the path - otherwise we
exercise the real ring code, but ignore the result and return 1.
"""
self._base_port = base_port
self.max_more_nodes = max_more_nodes
self._part_shift = 32 - part_power
# 9 total nodes (6 more past the initial 3) is the cap, no matter if
# this is set higher, or R^2 for R replicas
self.set_replicas(replicas)
self._reload()
}}}
All that jazz about part power and doing something trixy in _get_parts should be pulled out. |
|
| 2015-09-29 15:33:15 |
Aniruddha Singh Gautam |
swift: assignee |
|
Aniruddha Singh Gautam (aniruddha-gautam) |
|
| 2016-01-21 10:59:53 |
OpenStack Infra |
swift: status |
New |
Fix Released |
|
| 2016-01-26 14:15:48 |
OpenStack Infra |
tags |
low-hanging-fruit |
in-feature-crypto low-hanging-fruit |
|
| 2016-01-26 14:15:50 |
OpenStack Infra |
cve linked |
|
2016-0738 |
|
| 2016-03-08 23:31:53 |
OpenStack Infra |
tags |
in-feature-crypto low-hanging-fruit |
in-feature-crypto in-feature-hummingbird low-hanging-fruit |
|
