Github user neykov commented on the pull request:
https://github.com/apache/incubator-brooklyn/pull/314#issuecomment-62967070
> NPE when changing standby state
Confirming it was my error (wrong parameter name). Could you add a
checkNotNull in the rest method to prevent future confusion.
Some weird messages when changing states
* Starting as master and switching to HOT_STANDBY results in
```
2014-11-13 22:45:54,255 WARN Management node pqvmrTbg detected master
change, from (?) to <none>
2014-11-13 22:45:54,592 INFO Rebind complete (HOT_STANDBY, iteration 0) in
462ms: 0 apps, 0 entities, 0 locations, 0 policies, 0 enrichers, 0 feeds, 9
catalog items
2014-11-13 22:45:55,987 INFO Rebinding from
JcloudsLocation[swift:https://ams01.objectstorage.softlayer.net/auth/v1.0:IBMOS321366-2:cloudsoft/swift:https://ams01.objectstorage.softlayer.net/auth/v1.0]:svet_test_201411...
2014-11-13 22:45:55,994 INFO Rebind complete (MASTER, iteration 1) in
491ms: 0 apps, 0 entities, 0 locations, 0 policies, 0 enrichers, 0 feeds, 9
catalog items
2014-11-13 22:45:56,744 INFO Back-up of persisted state created on
promotion, in
C:\Users\sam\.brooklyn\brooklyn-persisted-state\backups\2014-11-13-2245-pqvmrTbg-promotion-kU7a
2014-11-13 22:45:57,173 WARN This management node (pqvmrTbg) supposed to
be master but reportedly unhealthy? no-op as expect other node to fix:
self=BasicManagementNodeSyncRecord{brooklynVersion=0.7.0-SNAPSHOT,
nodeId=pqvmrTbg, uri=http://127.0.0.1:8081/, status=STANDBY, priority=0,
localTimestamp=1415911556744=2014-11-13 22:45:56.744,
remoteTimestamp=1415911559000=2014-11-13 22:45:59.000}
2014-11-13 22:45:57,173 INFO Management node pqvmrTbg running as HA
STANDBY (caller requested HOT_STANDBY, will attempt rebind for HOT_STANDBY
next); master pqvmrTbg
2014-11-13 22:45:57,366 WARN Unable to promote pqvmrTbg to HOT_STANDBY,
switching to FAILED: java.lang.IllegalStateException: Cannot start read-only
when already running with persistence
java.lang.IllegalStateException: Cannot start read-only when already
running with persistence
at
brooklyn.entity.rebind.RebindManagerImpl.startReadOnly(RebindManagerImpl.java:312)
~[brooklyn-core-0.7.0-SNAPSHOT.jar:0.7.0-SNAPSHOT]
2014-11-13 22:45:58,270 WARN Management node pqvmrTbg unable to promote to
HOT_STANDBY (currently FAILED); (see log for further details)
```
* MASTER -> STANDBY
```
2014-11-13 23:11:31,490 INFO Back-up of persisted state created on
demotion, in
C:\Users\sam\.brooklyn\brooklyn-persisted-state\backups\2014-11-13-2311-hOUE8a2g-demotion-foea
2014-11-13 23:11:33,639 INFO Rebinding from
JcloudsLocation[swift:https://ams01.objectstorage.softlayer.net/auth/v1.0:IBMOS321366-2:cloudsoft/swift:https://ams01.objectstorage.softlayer.net/auth/v1.0]:svet_test_201411...
2014-11-13 23:11:33,649 INFO Rebind complete (MASTER) in 522ms: 0 apps, 0
entities, 0 locations, 0 policies, 0 enrichers, 0 feeds, 9 catalog items
2014-11-13 23:11:34,332 INFO Back-up of persisted state created on
promotion, in
C:\Users\sam\.brooklyn\brooklyn-persisted-state\backups\2014-11-13-2311-hOUE8a2g-promotion-DG1P
2014-11-13 23:11:34,752 WARN This management node (hOUE8a2g) supposed to
be master but reportedly unhealthy? no-op as expect other node to fix:
self=BasicManagementNodeSyncRecord{brooklynVersion=0.7.0-SNAPSHOT,
nodeId=hOUE8a2g, uri=http://127.0.0.1:8081/, status=STANDBY, priority=0,
localTimestamp=1415913094333=2014-11-13 23:11:34.333,
remoteTimestamp=1415913096000=2014-11-13 23:11:36.000}
2014-11-13 23:11:34,753 INFO Management node hOUE8a2g running as HA
STANDBY (explicitly requested); master hOUE8a2g
2014-11-13 23:11:35,169 WARN This management node (hOUE8a2g) supposed to
be master but reportedly unhealthy? no-op as expect other node to fix:
self=BasicManagementNodeSyncRecord{brooklynVersion=0.7.0-SNAPSHOT,
nodeId=hOUE8a2g, uri=http://127.0.0.1:8081/, status=STANDBY, priority=0,
localTimestamp=1415913094754=2014-11-13 23:11:34.754,
remoteTimestamp=1415913097000=2014-11-13 23:11:37.000}
2014-11-13 23:11:36,576 WARN This management node (hOUE8a2g) supposed to
be master but reportedly unhealthy? no-op as expect other node to fix:
self=BasicManagementNodeSyncRecord{brooklynVersion=0.7.0-SNAPSHOT,
nodeId=hOUE8a2g, uri=http://127.0.0.1:8081/, status=STANDBY, priority=0,
localTimestamp=1415913096171=2014-11-13 23:11:36.171,
remoteTimestamp=1415913098000=2014-11-13 23:11:38.000}
(repeating indefinitely)
```
* MASTER -> HOT_BACKUP
```
...
2014-11-13 22:49:11,004 WARN Cannot request read-only mode for
brooklyn.entity.rebind.RebindManagerImpl@e64686[mgmt=ec8oxm5d] when already
running - Task[Periodic read-only rebind:TZ3dbgWD]; ignoring
...
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
