- Devel - oVirt List Archives

Project vdsm-jsonrpc-java backup maintainer
by Sandro Bonazzola 30 Jun '14

30 Jun '14

Hi, looks like only one person has +2 / merge rights on vdsm-jsonrpc-java. I think every project should have a "backup" maintainer. Since it seems pkliczewski is the only committer there, I would like to propose him as maintainer too. -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

3 2

Tune HA timing,is it possible
by mad Engineer 29 Jun '14

29 Jun '14

Hi, Is it possible to tune time required to reboot VM in case of host failures. in our test we powered off one of the Host. Manager was still showing both Guest and Host as UP for around 5 minutes and after that it took another 4-5 minutes to start VM. Is there any tunable parameter that can be modified to bring down the time required to recognize Host failures. Thanks

2 1

[QE][ACTION REQUIRED] postponing beta release due to basic sanity test failure
by Sandro Bonazzola 29 Jun '14

29 Jun '14

Hi, while performing basic sanity test on 3.5.0 beta candidate I've seen the following bugs: Bug 1113882 - [ ERROR ] Failed to execute stage 'Closing up': [ERROR]::oVirt sdk is disconnected from the server. Bug 1113891 - storage domains are listed twice Bug 1113898 - New Virtual Machine dialog drop down list are all empty Bug 1113898 make impossible to create a new VM, so basic sanity test failed. We need to postpone beta release until this one is fixed. repoclosure test failed on F19: http://lists.ovirt.org/pipermail/users/2014-June/025475.html Since candidate beta build failed sanity test, I'm going to rebuild engine from master and repeat sanity test on the new build. -- Sandro Bonazzola Better technology. Faster innovation. Powered by community collaboration. See how it works at redhat.com

2 1

Engine - Changes to various issues around commands and queries construction and internal execution - please read
by Yair Zaslavsky 28 Jun '14

28 Jun '14

Hi all, Thanks to the help of Alon, oved, Tal, Moti, Arik and others, the following changes were introduced: 1. Internal commands invocation - When invoking an internal command from a command, please use the following : Instead of Backend.getInstance().runInternalAction... Use runInternalAction - a new method that was introdced at CommandBase. This method has two variants - one that accepts command context, and the other that does not have a command contet - runInternalAction(VdcActionType,VdcActionParametersBase,CommandContext) and runInternalAction(VdcActionType,VdcActionParametersBase) If CommandContext is not passed the context at the calling command will be cloned and set at the child command. If a Command context is pased - it should be the responsibility of the caller to clone, however, this will give the caller some degree of freedom to determine whether various parts of the context will be cloned, or not. Examples: runInternalAction(VdcActionType.AddPermission, permissionParams) has the same effect as : runInternlAction(VdcActionType.AddPermissiosn, permissionParams, getContext().clone()) runInternlAction(VdcActionType.AddPermissiosn, permissionParams, getContext().clone().withoutCompensationContext()) - will cause the compensation context to be reset, and let the child command determine the value of compensation context (at handleTransactivity method). The complete list of "context alteration methods" are - withCompensationContext(CompensationContext) , withoutCompensationContext() withLock(EngineLock), withoutLock() withExecutionContext(ExecutionContext), withoutExecutionContext() - bare in mind that all these follow the chaining method "design pattern" [1] (I would like to thank Moti for the naming suggestions) two methods for running InternalAction with context for tasks were created: runInternalActionWithTasksContext(VdcActionType, VdcActionParametersBase) runInternalActionWithTasksContext(VdcActionType, VdcActionParametersBase, EngineLock) These methods use ExecutionHandler.createDefaultContextForTasks to create the relevant command context to be passed to the child command. runInternalMultipleActions was introduced to command base in a similar manner, with 3 versions: runInternalMultipleActions(VdcActionType, ArrayList<VdcActionParametersBase>) runInternalMultipleActions(VdcActionType, ArrayList<VdcActionParametersBase>, ExecutionContext) runInternalMultipleActions(VdcActionType, ArrayList<VdcActionParametersBase>, CommandContext) 2. Queries invocation - runInternalQuery(VdcQueryType, VdcQueryParametersBase) was introduced to command base. Basically it takes the engine context from the current command context, and runs the internal query with it. EngineContext is the context which should hold all the common attributes to our flows at engine - currently it holds the engineSessionId, working towards moving correlationId to it as well. 3. Commands & Queries coding Each internal query should have a ctor that takes the parameters, and also the engine context . As some of the queries are both internal and non internal you may have two ctors - one with parameters only, one with parameters and EngineContext for example public class GetUnregisteredDiskQuery<P extends GetUnregisteredDiskQueryParameters> extends QueriesCommandBase<P> { public GetUnregisteredDiskQuery(P parameters) { this(parameters, null); } public GetUnregisteredDiskQuery(P parameters, EngineContext context) { super(parameters, context); } Notice that the ctor without the context calls the one with the context. Same happens at Commands: public RemovePermissionCommand(T parameters) { this(parameters, null); } public RemovePermissionCommand(T parameters, CommandContext commandContext) { super(parameters, commandContext); } 4. runVdsCommand was introduced to CommandBase as well runVdsCommand(VDSCommandType, VdsCommandParameters) - currently this just runs the vds command on vds broker, working on propagating the engine context via vds broker as well. Please use the above in your code. If you see any issues , or places where its problematic to use, feel free to contact me. [1] http://en.wikipedia.org/wiki/Method_chaining

2 1

Node down,but ovirt still shows VM as up
by mad Engineer 28 Jun '14

28 Jun '14

Hi, I am using Cisco UCS C200 M2 as Host running Centos 6.5 and KVM, Power Management not working properly,hence even with Node down Ovirt shows VM as still up,with uptime of VM increasing(on the manager) if i continue and save the changes its causing problems: 1. HA is not working ( Node status changed to Non responsive but VM status still up!! 2. Restart of Host gives wrong information-it shows host as rebooting but actually nothing is happening to Host!! While configuring power management: On ovirtmanager edited Power management and chose cisco_ucs with proper authentication to CIMC,but when clicked on test,it shows *Test Failed, Failed: You have to enter plug number Please use '-h' for usage * What is this plug,i couldn't find anything in CIMC This can be the reason for failure of HA and restart Can someone please help me fix this Thanks

1 0

Stats for oVirt Downloads: Jan-May 2014
by Brian Proffitt 28 Jun '14

28 Jun '14

All: For some time, traffic statistics for the sites in the ovirt.org domain (lists, resources, gerrit, and linode01) have been collected and organized using awstats[1] at stats.ovirt.org[2]. From the data provided for resources, I have been able to put together what should be a fairly definitive set of statistics for software downloads within the oVirt Project. Methodology The statistics that were analyzed were for five parts of the project: * Engine * Live * Node * Engine Reports * Engine dwh Downloads of the Engine RPM file, it was determined, would be indicative of actual oVirt installs. The allinone RPMs were not analyzed at this time, since installing this RPM is a choice made during the installation process itself. Tracking the Engine Reports and dwh RPMs was done to determine the popularity of these two tools and to ensure their numbers were comparable. To count downloads, data was gathered each month that listed the total downloads of every file, a number derived from the total hits each file had, minus any 206 hits, which indicated incomplete downloads. Key files for Engine, Engine Reports, and Engine dwh were identified and the data filtered to include counts for each one of the files, in whatever version released. Tracking of Live and Node RPMs was already set up within the awstats reporting, and was taken directly from awstats in each month. If there is any part of this methodology that is in error, feedback is very much appreciated. Presentation of Data Data was aggregated using Pivot Tables in Google Docs, and the application of SUMIFS functions in the same spreadsheet document. This document is still being formatted into a more presentable form, but until then, I wanted to deliver some preliminary results to the community for the first five months of the year. It will be noted that in general, download numbers are on the rise, particularly around the time oVirt 3.4 was released. There are significant gaps in downloads of Engine Reports and Engine dwh. I am still investigating the cause of this lack of data. Engine Jan 5815 Feb 4251 Mar 1980 Apr 10509 May 12043 Engine Reports Jan 0 Feb 1209 Mar 0 Apr 10282 May 9647 Engine dwh Jan 0 Feb 603 Mar 0 Apr 9974 May 10311 Live Jan 845 Feb 757 Mar 739 Apr 1981 May 1187 Node Jan 95 Feb 757 Mar 739 Apr 1981 May 1187 Thanks for Alon Bar-Lev and Michael Scherer for their invaluable assistance with this data gathering and reporting. BKP [1] http://awstats.sourceforge.net/ [2] http://stats.ovirt.org/ -- Brian Proffitt oVirt Community Manager Project Atomic Community Lead Open Source and Standards, Red Hat - http://community.redhat.com Phone: +1 574 383 9BKP IRC: bkp @ OFTC

2 2

multipath configuration
by Yeela Kaplan 27 Jun '14

27 Jun '14

Hi, Currently multipath.conf is being rotated each time we reconfigure it. We'd like to change the behaviour for multipath.conf so that current configuration will be commented out and we'd stop rotating (in the same manner as libvirt conf works today). Does anybody have any comment for/ against? Thanks in advance, Yeela

7 14

Note on SearchableListModel.supportsServerSideSorting override
by Vojtech Szocs 27 Jun '14

27 Jun '14

Hi guys, I see that some UiCommon models, such as VmListModel, already have override for SearchableListModel.supportsServerSideSorting method. In general, any SearchableListModel subclass should contain: @Override public boolean supportsServerSideSorting() { return true; } if, and only if, its syncSearch method uses VdcQueryType.Search (Search query). This applies to all SearchableListModel subclasses, i.e. both main tab & sub tab list models. Due to current class inheritance where sub tab list models (like ClusterVmListModel) extend main tab list models (like VmListModel), please make sure to check following scenario: - main tab model uses Search query, supportsServerSideSorting=true - sub tab model does NOT use search query --> must provide override supportsServerSideSorting=false Thanks, Vojtech

1 0

Call for Papers in Two Weeks: Linux.conf.au
by Brian Proffitt 27 Jun '14

27 Jun '14

Conference: Linux.conf.au Information: Each year open source geeks from across the globe gather in Australia or New Zealand to meet their fellow technologists, share the latest ideas and innovations, and spend a week discussing and collaborating on open source projects. The conference is well known for the speakers and delegates depth of talent, and its focus on technical linux content. Possible topics: Virtualization, oVirt, KVM, libvirt, RDO, OpenStack, Foreman Date: January 12-15, 2015 Location: Auckland, New Zealand Website: http://lca2015.linux.org.au/ Call for Papers Deadline: July 13, 2014 Call for Papers URL: http://lca2015.linux.org.au/cfp Contact me for more information and assistance with presentations. -- Brian Proffitt oVirt Community Manager Project Atomic Community Lead Open Source and Standards, Red Hat - http://community.redhat.com Phone: +1 574 383 9BKP IRC: bkp @ OFTC

1 0

Host not restarting through manager
by mad Engineer 27 Jun '14

27 Jun '14

Hi, I have two UCS server,both are configured for power management, but when i put host to maintenance and restart it by selecting power management>restart on the OVirt Manager it shows as rebooting but the host is not really rebooting,i had the same issue with HP servers ILO2. What should i do to make it work Also under Host>power management>test is failing with "Test Failed, Failed: You have to enter plug number Please use '-h' for usage" Thanks for any help

1 0