Commit Graph

3297 Commits

Author SHA1 Message Date
Chiradeep Vittal 2c995c597b fix mocks 2011-09-07 14:07:50 -07:00
Chiradeep Vittal fbfb1b8f5b move interval to listener -- allows it to be configurable if needed 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 22c1e66a6c clean up tests for security group manager v2 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 3f9922ca01 Add more detail when unable to schedule cleanup 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 9b75fab343 Retry only a few times 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 3403ccdaf1 change some logs to debug level 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 13f416698a fix merge problem 2011-09-07 13:58:49 -07:00
Chiradeep Vittal 2027049fd6 if the xenserver host cannot do bridge firewalling do not attempt to retry the security rule updat 2011-09-07 13:58:49 -07:00
Chiradeep Vittal cdbfb5fb0b added a few more dangerous JMX operations 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 549749a2d7 Add ability to trigger ruleset updates from JMX 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 65414903b7 Add the ability to disable some vms from being scheduled 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 1ddda37e13 enable logging of the management server id 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 78ded9c78e jconsole doesn't understand Ternary 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 3c5becbd95 Enable monitoring over JMX 2011-09-07 13:58:48 -07:00
Chiradeep Vittal d531f3af29 Dequeue the vm with the largest id first -- this ensures that newly started vms get updated early 2011-09-07 13:58:48 -07:00
Chiradeep Vittal b0be029cb2 Retry 3 times on deadlock: if not updated, then throw a runtime exception to roll back the vm start 2011-09-07 13:58:48 -07:00
Chiradeep Vittal b7c1296534 make signature calculation consistent across management servers 2011-09-07 13:58:48 -07:00
Chiradeep Vittal 4b11e5e508 retry a transaction if it deadlocks 2011-09-07 13:58:48 -07:00
Chiradeep Vittal d532bec118 fix perf. issue in deploy vm. also remove dead code in dao 2011-09-07 13:58:47 -07:00
Chiradeep Vittal c504ce1aaf if true, this could lead to the dog chasing its own tail 2011-09-07 13:58:47 -07:00
Chiradeep Vittal 8006ed159b only update sequence number when asked to. full sync SHOULD NOT request sequence number updates since this will lead to the dog chasing its own tail 2011-09-07 13:58:47 -07:00
Chiradeep Vittal ed666cb01a Avoid going back to the nics table for information we already have 2011-09-07 13:58:47 -07:00
Chiradeep Vittal 2611d1d61f thread.yield() to attempt to induce deadlocks only 2011-09-07 13:58:47 -07:00
Chiradeep Vittal 46200d122e try using multi inserts to overcome deadlock during batch update 2011-09-07 13:58:47 -07:00
Chiradeep Vittal 3c45a7a195 Reduce batch size to 16 to avoid sql errors 2011-09-07 13:58:47 -07:00
Chiradeep Vittal d6448d9fb8 useful logging 2011-09-07 13:58:47 -07:00
Chiradeep Vittal c8712401ca unit test for SG mgr v2 2011-09-07 13:58:47 -07:00
Chiradeep Vittal c5fea7a03c use the new implementation of SG manager 2011-09-07 13:58:47 -07:00
Chiradeep Vittal 8c6fe3a9af Use hashset since it is 50% faster. we can use treeset when we support priority 2011-09-07 13:58:47 -07:00
Chiradeep Vittal ed867d0799 unnecessary statement 2011-09-07 13:58:46 -07:00
Chiradeep Vittal 68dcad8ce6 Mocks for all 2011-09-07 13:58:46 -07:00
Chiradeep Vittal 0cc3a1e3f8 WIP: implement a queue where inserting a job for an existing vm does not create a new job 2011-09-07 13:58:46 -07:00
Chiradeep Vittal 6465ccff1b WIP : memory based queueing 2011-09-07 13:58:46 -07:00
Chiradeep Vittal fc274739f0 bug 10997: unnecessary database operation 2011-09-07 13:58:46 -07:00
Chiradeep Vittal 5172272d99 bug 10997: No need to go further when we don't have a proper job to process 2011-09-07 13:58:46 -07:00
Alex Huang 3a087c2e62 schedule the work after transaction 2011-09-07 13:58:46 -07:00
Alex Huang 47957c85d9 missed one file 2011-09-07 13:58:46 -07:00
Alex Huang 0900b232fc one big transaction to see if it's any faster 2011-09-07 13:58:45 -07:00
alena 668276e22c bug 11326: don't try to transfer the agent if it's a forward agent
status 11326: resolved fixed

Also added more logging to the agent rebalance code.
2011-09-07 12:49:02 -07:00
anthony 9842a9aed3 bug 10078:
1. introduce migratewait in global configuration, the default value is 1 hour
 2. use async xapi VM migration API

status 10078: resolved fixed
2011-09-07 12:40:30 -07:00
kishan 3a6c226bc9 bug 11345: Compare current and previous network stats before updating. Ignore stats if the current stats are not same as the prev stats. Set NetworkUsageAnswer log level to debug 2011-09-07 13:16:11 +05:30
alena 03958369e0 bug 11331: allow extending vnet range as a part of updateZone call
status 11331: resolved fixed
2011-09-06 17:41:38 -07:00
alena 654bcdeb5e bug 10751: don't remove already removed templates during the account cleanup
status 10751: resolved fixed
2011-09-02 16:10:00 -07:00
alena 7cea149dc2 bug 11002: userVmDao - create searchBuilder as a part of dao init process (otherwise it's not threadsafe)
status 11002: resolved fixed
2011-09-02 15:51:24 -07:00
Kelven Yang a7ac75f920 bug 11304: restore host status after initialization failure 2011-09-02 15:17:57 -07:00
anthony 57e731b60e set timeout for CheckOnHostCommand to 50 s 2011-09-02 15:01:06 -07:00
anthony 238ffd0a9b put getConnection into try 2011-09-02 15:00:58 -07:00
alena a357dcc728 bug 11259: ROOT admin can extract template/iso even when it's not mark as extractable
status 11259: resolved fixed
2011-09-02 12:33:30 -07:00
alena 0242dd51ab bug 11332: don't try to log vm's information when the vm is already Expunged
status 11332: resolved fixed

Conflicts:

	server/src/com/cloud/vm/dao/VMInstanceDaoImpl.java
2011-09-02 12:08:16 -07:00
Sheng Yang 96cf40972c bug 11305: Remove router stop logic from disconnect host.
Now the logic is: if we can only connect to one of the two redundant routers, we
would stop the one that can't be connected. If we fail to program both routers,
just let it go.
2011-08-31 19:27:17 -07:00
Sheng Yang 5e450e9852 bug 11305: Stop disconnected redundant virtual router if we can process with the connected one
When we apply rules or start new VM, we may encounter some running routers that
we can't program. That can due to network issue or host is down or vCenter is
disconnected, etc. To keep the synchronization, we would stop them, but only
when there is the other router we've successfully updated. If both routers are
unable to communicate with, we simply give up and report it user.
2011-08-31 19:15:38 -07:00
Kelven Yang 1216aebbc0 bug 10765: add japanese keyboard support for linux OSes 2011-08-31 18:17:38 -07:00
alena 78b2df8549 Do account permission checks for snapshots the way it's done for all other ControlledEntities 2011-08-31 17:53:48 -07:00
frank 996d6302eb Bug 8966 - Oracle VM (OVM) support
configure hostname of host for OCFS2
small fixes
2011-08-31 15:37:16 -07:00
kishan a6d196e96c bug 11316: In 2.2.y, removed new template upgrade changes. Added upgrade path from 2210 to 2211. Moved existing changes to 2211to2212
status 11316: resolved fixed
2011-08-31 15:11:36 +05:30
kishan a7b760a482 bug 11316: Removed new template upgrade changes. Added upgrade path from 2210 to 2211. Existing changes moved to 2211to2212 upgrade
status 1316: resolved fixed
2011-08-31 15:01:39 +05:30
alena 4f6914d56b bug 11298: give detailed error message when failed to authenticate a user
status 11298: resolved fixed
2011-08-30 16:52:52 -07:00
alena aaf46356bb Mark account as cleanup_required=0 when cleanup was finished successfully 2011-08-30 14:10:07 -07:00
alena 4dd4b41344 Don't remove the domain when it has accounts that are removed, but require cleanup 2011-08-30 14:10:00 -07:00
alena 60ce12b892 Fixed multipe problems in account cleanup process:
* don't try to delete the template when it's already removed
* no need to perform permission check when deleteFirewallRule is called by System (as a part of cleanupAccount process for instance)
2011-08-30 11:22:56 -07:00
alena 5e28e5e3f1 Fixed the bug when networkRate for the domR's guest network wasn't taken from the service offering 2011-08-30 10:40:54 -07:00
Murali Reddy f0fa75f793 bug 11179: template url and format are not validated
added check to verify url ends with extension same as the template format specified
2011-08-30 16:23:06 +05:30
Edison Su c90397bd0e bug 11217: only work on the snapshot whose state is "backedup" and is not removed
status 11217: resolved fixed
2011-08-29 15:29:34 -07:00
Edison Su e09e4e5bed bug 11217: add db upgrade script, if sechost_id is missing of a snapshot, add one got from presnapshot_id
status 11217: resovled fixed
2011-08-29 15:29:14 -07:00
alena 19ee92a528 bug 11196: fixed the code when gson serialization was done twice
status 11196: resolved fixed
2011-08-29 14:48:09 -07:00
alena 4a4e832fdb bug 11290: insert VPN config values as a part of 229-2210 upgrade
status 11290: resolved fixed
2011-08-29 11:20:57 -07:00
alena 6ea450d362 Reverted wrong commit for logging in HostDaoImpl 2011-08-26 17:48:10 -07:00
Kelven Yang bf814aeadd bug 11263: only update host Id when VM is in running state when detected external VM migration 2011-08-26 16:27:42 -07:00
alena 686f9406e1 Fixed the bug in extractTemplate - do ignoreCase comparison when verify the download mode. 2011-08-26 10:11:18 -07:00
Nitin 56c59bb5d5 bug 11244: Thread updating "op_host_capacity" table shouldn't do updates for removed pods/zones.
status 11244: resolved fixed
2011-08-26 18:56:37 +05:30
Naredula Janardhana Reddy 701ddbfc1d bug 11253:
Added New value "link-local"  to  global config network.loadbalancer.haproxy.stats.visibility . With this change it can take new parameter "link-local" value apart from the existing 3 values global,guest-network,disabled.
       global -  stats visible from public network
       guest-network - stats visible only to guestnetwork.
       link-local - stats visible only to link local network
       disabled - stats disabled.
2011-08-26 15:25:45 +05:30
alena 2a0e596f15 Set firewall.rule.ui.enabled to false by default 2011-08-25 16:11:03 -07:00
Kelven Yang 80f5b9683f bug 11162: merge fix from 2.2.10 premium 2011-08-25 15:37:26 -07:00
Kelven Yang c74ae5d61f bug 11162: support memory overprovisioning in VMware 2011-08-25 15:37:26 -07:00
prachi e11d2f4093 Bug 9921 - template tags
-Added missing import statement.
2011-08-25 15:18:21 -07:00
prachi 916177b691 Bug 9921 - template tags
-Upgrade path from 2.2.10 to 2.2.11

Conflicts:

	server/src/com/cloud/upgrade/PremiumDatabaseUpgradeChecker.java
2011-08-25 15:18:19 -07:00
prachi 089b23f7a6 Bug 9921 - template tags
Changes:
- CreateTemplate and RegisterTemplate now support adding a template tag. It is a string value. This is root-admin only action - only admin can add template tags.
- ListTemplates will return the template tag in response.
- HostAllocator changed to use template tag along with the existing tag on service offering. If both tags are present, allocator now finds hosts satisfying both tags. If no hosts have both tags, allocation will fail.
- DB changes to add new column to vm_template table.
- DB upgrade changes for upgrade from 2.2.10 to 2.2.11

Conflicts:

	server/src/com/cloud/api/ApiResponseHelper.java
	server/src/com/cloud/template/TemplateAdapterBase.java
	server/src/com/cloud/vm/UserVmManagerImpl.java
2011-08-25 15:18:18 -07:00
alena 1ea383e96f bug 11260: Take networkRate for domR's guest and public interface from the corresponding system service offering 2011-08-25 14:13:23 -07:00
prachi 1baf90b18b Bug 11261 - Allow to update host tags
Changes:
- Changes to updateHostCmd to accepts hosttags parameter
- Changes to wipe out existing tags and save new ones in host_tags DB.
- UpdateHost is Admin only operation - so only root admin can update host tags
2011-08-25 12:58:37 -07:00
alena e902599847 bug 11185: support multiple CIDR on overlapping port ranges for firewall rules
status 11185: resolved fixed

Conflicts:

	server/src/com/cloud/network/firewall/FirewallManagerImpl.java
2011-08-25 12:06:28 -07:00
Abhinandan Prateek 0d14faec34 bug 11181: taking care of a boundary condition with 0 user VM 2011-08-25 15:36:08 +05:30
Edison Su 7b034a68c0 TESTClient: more features: all the api will go through 8080 with apikey/securitykey, both keys are generated dynamically when you deploy your setup, add list/map type for api, and bug fix 2011-08-24 18:45:11 -07:00
alena f5dacdb473 bug 11242: check if db object is already removed before generating the Debug log. It can happen in case when deleteHost command is called many times in parallel for the same host
status 11242: resolved fixed
2011-08-24 15:44:58 -07:00
Kelven Yang c3b6fd561c bug 11230, 11229: merge fix from 2.2.10 branch, fix Eclipse project settings for premium-move 2011-08-24 15:16:59 -07:00
Kelven Yang 65744569c3 bug 11219: use local storage flag from service offering when it is ROOT disk 2011-08-24 15:03:39 -07:00
alena d32dbcbf7b Method signature change 2011-08-24 15:00:41 -07:00
alena 488b995299 bug 11236: domainAdmin/regularUser can edit/delete/copy/extract Public template/iso only if it was created by them. They still can use/see public template/iso when execute list/deploy/attachIso commands. Root admin can operate with other people templates w/o any restrictions.
status 11236: resolved fixed
2011-08-24 14:48:22 -07:00
alena 0ce47f5cdf bug 11237: 1) Remove capacity records when corresponding pod/zone is removed 2) On api layer check if zone exists before listing capacity record
status 11237: resolved fixed
2011-08-24 13:19:27 -07:00
Edison Su be249bb10a bug 11217: add secondary storage id into snapshot, if it's an empty snapshot
status 11217: resolved fixed
2011-08-24 11:05:19 -07:00
alena 524ed0b15c bug 11060: use lockTable instead of rowLock when createDomR/createDhcp
status 11060: resolved fixed
2011-08-24 11:01:02 -07:00
Edison Su 3786d42853 bug 11217: add secondary storage id into snapshot, if it's an empty snapshot
status 11217: resolved fixed
2011-08-24 10:47:47 -07:00
Nitin f4eb81f9b8 bug 10972: Improve logging - put in the hostid and hostname on which the vm launches.
status 10972: resolved fixed
2011-08-24 18:00:09 +05:30
root eda107daae removing spurious debug statement 2011-08-24 16:48:52 +05:30
Abhinandan Prateek a6ab4ac3ad bug 11181: batching the select statements 2011-08-24 16:24:07 +05:30
Naredula Janardhana Reddy 0e4de58d4e bug 11191:
Added global config to enable/disable rp_filter for domR.
   previous commit: d966906374d4a0cb8fa57326a1f7625c871f64fd

   Test Case-1 :
      1) Set network.disable.rpfilter  global config to true
      2) Restart the domR
      3) check the settings reflected in proc filesystem
           - for public interface like eth2,eth3 : /proc/sys/net/ipv4/conf/eth2/rp_filter should have 0 , and rest other interfaces should have value of 1
   Test Case-2 :
      1) set network.disable.rpfilter global config to false
      2) Restart the domR
      3) check the settings reflected in proc filesystem
           - for public interface like eth2,eth3 : /proc/sys/net/ipv4/conf/eth2/rp_filter should have 1 , and rest other interfaces should also have value of 1
2011-08-24 13:36:43 +05:30
Naredula Janardhana Reddy b78f66f597 bug 11191: Added global config to disable/enable rp_filter for domR(public interfaces) 2011-08-24 13:36:16 +05:30
frank ca7fb65ca2 full opensource 2011-08-23 19:23:49 -07:00
Sheng Yang f5e939b4a5 bug 11214: Don't try to start the running router again
It's very like caused by StartRouterCmd sent to the running router. I can
reproduce it by issue a StartRouterCmd to a running redundant router. And this
patch should the following exception:

Exception: com.cloud.exception.ResourceUnavailableException: Resource
[VirtualNetworkApplianceManagerImpl$$EnhancerByCGLIB$$565b4d45:0] is
unreachable: There are already two redundant routers with IP 10.91.32.126, they
are r-5-VM(5) and r-4-VM(4)

status 11214: resolved fixed
2011-08-23 18:20:28 -07:00
Sheng Yang 4bdd56a63e Add more debug info 2011-08-23 17:59:38 -07:00
Kelven Yang 98212299b5 Move VMware/Premium Secondary storage VM to opensource 2011-08-23 14:33:05 -07:00