etan39
/
cloudstack
mirror of https://github.com/apache/cloudstack.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
6,677 Commits 323 Branches 325 Tags 976 MiB
Commit Graph

59 Commits

Author SHA1 Message Date
Alex Huang 66713a490d fixed problem with the standalone connection has open transaction 2011-07-22 11:25:43 -07:00
Kelven Yang 95b48ba222 bug 10791: add data integrity check upon management server startup 2011-07-21 17:07:21 -07:00
alena b1d5b5a51d 2 fixes for Agent Load Balancer: 
* when management server dies and notifies other management servers about this, the running management server has to cleanup host_transfer records belonging to the died management server
* issue agent load balancing task only when agent load (number of connected agents in the system) exceeds "agent.load.threshold" - 70% by default
 2011-07-21 15:43:52 -07:00
Kelven Yang 281201be8e Make cluster request timeout to be configurable 2011-07-14 18:23:23 -07:00
Kelven Yang 6d8eb45bad add logging to clustering notification 2011-07-11 17:16:23 -07:00
Kelven Yang e4bd357c53 bug 10677: add client side time out to cluster servlet interface 2011-07-11 13:37:33 -07:00
Alex Huang ea48d40e5f connection conierge to deal with connections that are not returned to pools 2011-07-08 18:04:57 -07:00
Alex Huang f28a2b40a3 bug 10501: This is really Kelven's bug but I'll fix it for him anyways. DAO code already have a way to extract the DB connection from a transaction that is stored in the TLS. There's no real reason for the DAO code to add special semantics to use a different DB connection. That can be done by simply switching the transaction before it even reached the dao code. Think about it. Why would anyone want to call one dao function, switch transaction, and then switch back. The right thing is for the caller to switch transaction, call a series of dao codes, and switch it back. That's the semantics I changed to. By doing this, it also eliminates the number of debug messages in this bug. 2011-07-01 11:03:37 -07:00
Alex Huang 7a4af21f72 bug 10322: Transaction is using the new Merovingian 2011-06-29 15:34:53 -07:00
Alex Huang c8c84af109 bug 10322: integrated into cluster management 2011-06-29 12:03:36 -07:00
Alex Huang fefc798197 bug 10322: better mannered merovingians 2011-06-29 12:03:22 -07:00
Alex Huang 687e9f6e67 bug 10322: arrival of the next generation Merovingians 2011-06-29 12:03:05 -07:00
Alex Huang 27809e651e bug 10322: arrival of the next generation Merovingians 2011-06-29 12:02:53 -07:00
Alex Huang 89269069f9 Added JMX to Transaction 2011-06-28 01:17:46 -07:00
alena 18dc85c765 Pass isForRebalance parameter to processConnect method of all the listeners - some listeners don't have to be notified when connection happens as a a part of Agent Rebalance process (VirtualMachineManagerImpl listener for instance) 2011-06-27 10:37:00 -07:00
alena 826f5eedf1 bug 10447: don't notify VirtualMachineManager listener when do host rebalance - vm sync is not needed in this case. 
status 10447: resolved fixed
 2011-06-27 10:37:00 -07:00
Kelven Yang a295e381a7 bug 10449: handle the case when cluster servlet may be called during management server initialization phase 2011-06-24 18:12:25 -07:00
Kelven Yang 1db2667c54 bug 10029: update mshost state instead of removing it 2011-06-24 11:18:28 -07:00
Kelven Yang 8490fc2bf5 bug 10176: add cluster listener registration back 2011-06-23 14:50:58 -07:00
alena 4335b8b179 bug 10416: fixed the bug in agentLoadBalancerPlanner 
status 10416: resolved fixed
 2011-06-22 17:14:14 -07:00
alena 00a35314bb Some fixes to agent lb code: 
* Schedule agent LB as a TimerTask as it can take time for the management server to start; and it can accept rebalance requests only when it's up and running
* Removed Starting state from mshost as it's not being used anywhere
* Fixed the bug where requests weren't routed properly from the old host owner to the new one.
 2011-06-13 22:28:22 -07:00
alena 158807c877 Fixed unittest 2011-06-06 16:52:11 -07:00
alena 14cdc7de14 bug 9127: covered failure scenarios for agent LB. 
status 9127: resolved fixed

The feature is completed; please file separate bugs if any issue arises during the testing.
Wiki link describing how agentLB works: http://intranet.lab.vmops.com/engineering/release-2.2-features/agent-load-balancing
 2011-06-05 17:35:30 -07:00
alena 635a1075de bug 9127: intermediate checkin for agent load balancer 2011-06-02 21:47:54 -07:00
alena 52aab0e441 Use correct field name when search for management server by State 2011-06-02 10:18:30 -07:00
alena cd6d933e48 Better logging for cluster agent rebalancing 2011-05-25 18:59:26 -07:00
alena 388aea7be3 bug 9127: intermediate checkin for agent load balancer 2011-05-24 19:04:50 -07:00
alena 55692fd7cf bug 9127: intermediate checkin for agent load balancer 2011-05-24 18:45:00 -07:00
Alex Huang 2b650acc17 more gson crap 2011-05-18 11:01:46 -07:00
Chiradeep Vittal 303e2a7481 Start of zonesfeature / mycloud/cloudkit 2011-05-03 10:40:17 -07:00
Kelven Yang 824f0586f7 Add manager context to CleanupMaid to allow management service access in cleanup() method 2011-05-02 18:28:38 -07:00
Frank 92155522f2 Add license header to files 2011-04-14 11:23:14 -07:00
Alex Huang 6936a666fd removed some references to name 2011-04-14 03:54:34 -07:00
Kelven Yang 1b9cbd9166 bug 9223, 9224: persist runid to form cluster session, based on cluster session and DB condition to issue isolation notification for self-fencing 2011-04-13 15:13:54 -07:00
Kelven Yang 2a45327a89 Miss out one place to use queue instead, this is for decoupling notification from heartbeat to avoid being potentially blocked in callbacks 2011-04-12 15:06:08 -07:00
Kelven Yang 03afe54084 Regression fixes for adding openNew() to Transaction.java 2011-04-11 17:44:31 -07:00
Kelven Yang 149eb0fade Use seperate DB connection to handle cluster management 2011-04-11 16:37:58 -07:00
Alex Huang 5d488b106d bring changes to master from 2.2.4 2011-04-05 11:01:36 -07:00
Alex Huang 2230c24ab3 check in changes from 2.2.4 2011-04-05 11:01:35 -07:00
Alex Huang 075fba5899 stackmaid is now taskmanager 2011-04-05 10:17:22 -07:00
Kelven Yang fa0dd8a228 Allow cluster management confliction detection to handle quick manageemnt server restarts 2011-03-28 15:14:51 -07:00
Kelven Yang 359dc18992 Fix possible duplicated JMX registration issue 2011-03-25 15:16:41 -07:00
Alex Huang 82f7504832 checking in the upgrade stuff for now 2011-03-19 09:27:10 -07:00
Kelven Yang 65d4cc98be Allow template re-deployment once template is deleted from hypervisor and CloudStack still holds out-dated status 2011-03-17 17:59:51 -07:00
Kelven Yang 14c0eb74a0 Use the correct one to find ManagementServerHostVO 2011-03-10 15:32:58 -08:00
Kelven Yang 6b8f073f99 Fix issues for management server ping support in cluster management 2011-03-10 10:32:17 -08:00
Kelven Yang 6203ba6bfe bug 8558: improve cluster management 2011-03-09 17:18:48 -08:00
Kelven Yang 33b3333d65 Add JMX support to async job management, active job details will be checked in jConsole 2011-03-08 16:34:17 -08:00
Kelven Yang 13c5a4ebcf forget to add new files in 2011-03-07 19:43:37 -08:00
Kelven Yang d0892a964d Add JMX monitoring to cluster management 2011-03-07 18:40:33 -08:00