Nadav Har'El cc4117d6c1 repair: do not use an atomic integer ...

Avi asked not to use an atomic integer to produce ids for repair
operations. The existing code had another bug: It could return some
id immediately, but because our start_repair() hasn't started running
code on cpu 0 yet, the new id was not yet registered and if we were to
call repair_get_status() for this id too quickly, it could fail.

The solution for both issues is that start_repair() should return not
an int, but a future<int>: the integer id is incremented on cpu 0 (so
no atomics are needed), and then returned and the future is fulfilled.

Note that the future returned by start_repair() does not wait for the
repair to be over - just for its index to be registered and be usable
to a call to repair_get_status().

Signed-off-by: Nadav Har'El <nyh@cloudius-systems.com>

2015-08-31 09:31:19 +03:00

..

api-doc

API: storage_service Add the swagger definition for ownership

2015-08-25 19:39:13 +03:00

api.cc

Revert "Merge "Move the API configuration from command line to configuration" from Amnon"

2015-08-30 12:09:00 +03:00

api.hh

Revert "Merge "Move the API configuration from command line to configuration" from Amnon"

2015-08-30 12:09:00 +03:00

cache_service.cc

API: Add row related method to cache_service API

2015-08-17 20:09:07 +03:00

cache_service.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

collectd.cc

adding the collectd implementation

2015-06-28 17:37:59 +03:00

collectd.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

column_family.cc

Merge "test for compaction metadata's ancestors" from Raphael

2015-08-27 11:07:53 +03:00

column_family.hh

API: Expose the map_reduce_cf from column_family

2015-08-17 19:42:23 +03:00

commitlog.cc

Adding the commit log metric stub implementation

2015-07-06 12:38:22 +03:00

commitlog.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

compaction_manager.cc

api: implementing some ops of compaction manager api

2015-08-12 15:10:27 -03:00

compaction_manager.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

endpoint_snitch.cc

API: Add the get_snitch command

2015-08-16 12:15:38 +03:00

endpoint_snitch.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

failure_detector.cc

Adding the Failure detector API implementation

2015-06-03 19:13:03 +03:00

failure_detector.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

gossiper.cc

Cleaning the gossiper API

2015-07-15 14:20:28 +03:00

gossiper.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

hinted_handoff.cc

Cleaning the hinted_handoff API

2015-07-15 14:20:27 +03:00

hinted_handoff.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

lsa.cc

api: Fix return type for /lsa/compact

2015-08-25 15:29:10 +03:00

lsa.hh

api: Introduce RESTful API for LSA

2015-08-06 16:50:15 +02:00

messaging_service.cc

api/messaging_service: Use get_stats

2015-07-16 17:23:26 +08:00

messaging_service.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

storage_proxy.cc

API: Adding the histogram implementation to storage_proxy

2015-07-26 11:03:31 +03:00

storage_proxy.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00

storage_service.cc

repair: do not use an atomic integer

2015-08-31 09:31:19 +03:00

storage_service.hh

api: Switch to "#pragma once" as include guard

2015-07-22 18:03:42 +03:00