[ovirt-devel] [ OST Failure Report ] [ oVirt Master (cockpit-ovirt) ] [ 15-03-2018 ] [ 098_ovirt_provider_ovn.use_ovn_provider ]

Dominik Holler dholler at redhat.com
Mon Mar 19 08:40:32 UTC 2018


On Sat, 17 Mar 2018 16:16:49 +0200
Dan Kenigsberg <danken at redhat.com> wrote:

> Thanks for filing those, but let us not keep them secret:
> 
> Bug 1557419 - Recommend that a call should be tried again
> sounds most reasonable. Providing more information to the client
> makes sense
> 
> Bug 1557424 - Automatically retry call failed because he failed to
> acquire a lock
> is more tricky. when would REST perform the retry? 

If the command returns a machine-readable hint that the conflict is
temporary.

> how often? We must make sure we never cause a livelock.
> 

I agree that it is a good idea to have an upper limit for the count of
retries to prevent a livelock.

> Can we alternatively take a normal blocking lock when we create an
> external network (instead of a trylock)?
> 

This might work, but currently no command is doing this.
(The lock is acquired in CommandBase.acquireLockInternal() which could
be overridden for all commands using an external network provider.)
But I would prefer a solution which solve the issue for other commands,
too.

> On Fri, Mar 16, 2018 at 4:52 PM, Dominik Holler <dholler at redhat.com>
> wrote:
> > I have created
> > https://bugzilla.redhat.com/show_bug.cgi?id=1557419 and
> > https://bugzilla.redhat.com/show_bug.cgi?id=1557424
> > to discuss how failing to acquire locks in engine should be handled
> > in engine's REST-API, in ovirt-sdk or in application (OST in this
> > case).
> >
> >
> > On Fri, 16 Mar 2018 09:32:47 +0000
> > Dafna Ron <dron at redhat.com> wrote:
> >  
> >> Thank you for the fast reply and help.
> >>
> >> On Thu, Mar 15, 2018 at 8:21 PM, Dominik Holler
> >> <dholler at redhat.com> wrote:
> >>  
> >> > On Thu, 15 Mar 2018 16:24:10 +0000
> >> > Dafna Ron <dron at redhat.com> wrote:
> >> >  
> >> > > Hi,
> >> > >
> >> > > We have a failure on master for test
> >> > > 098_ovirt_provider_ovn.use_ovn_provider in project
> >> > > cockpit-ovirt. This seems to be a race because object is
> >> > > locked. also, the actual failure is logged as WARN and not
> >> > > ERROR.
> >> > >
> >> > > I don't think the patch is actually related to the failure but
> >> > > I think the test should be fixed.
> >> > > can you please review to make sure we do not have an actual
> >> > > regression and let me know if we need to open a bz to fix the
> >> > > test?
> >> > >
> >> > >
> >> > > *Link and headline of suspected patches: *
> >> > > *https://gerrit.ovirt.org/#/c/89020/2
> >> > > <https://gerrit.ovirt.org/#/c/89020/2> - *
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > >
> >> > > *wizard: Enable scroll on start page for low-res screensLink to
> >> > > Job:http://jenkins.ovirt.org/job/ovirt-master_change-queue-tester/6374
> >> > > <http://jenkins.ovirt.org/job/ovirt-master_change-queue-tester/6374>Link
> >> > > to all
> >> > > logs:http://jenkins.ovirt.org/job/ovirt-master_change-queue-  
> >> > tester/6374/artifacts  
> >> > > <http://jenkins.ovirt.org/job/ovirt-master_change-queue-
> >> > tester/6374/artifacts>(Relevant)
> >> > > error snippet from the log: <error>2018-03-15 10:05:00,160-04
> >> > > DEBUG
> >> > > [org.ovirt.engine.core.sso.servlets.OAuthTokenInfoServlet]
> >> > > (default task-10) [] Sending json response2018-03-15
> >> > > 10:05:00,160-04 DEBUG
> >> > > [org.ovirt.engine.core.sso.utils.TokenCleanupUtility] (default
> >> > > task-10) [] Not cleaning up expired tokens2018-03-15
> >> > > 10:05:00,169-04 INFO
> >> > > [org.ovirt.engine.core.bll.provider.network.SyncNetworkProviderCommand]
> >> > > (EE-ManagedThreadFactory-engineScheduled-Thread-90) [789edb23]
> >> > > Lock Acquired to object
> >> > > 'EngineLock:{exclusiveLocks='[c38a67ec-0b48-4e6f-be85-  
> >> > 70c700df5483=PROVIDER]',  
> >> > > sharedLocks=''}'2018-03-15 10:05:00,184-04 INFO
> >> > > [org.ovirt.engine.core.bll.provider.network.SyncNetworkProviderCommand]
> >> > > (EE-ManagedThreadFactory-engineScheduled-Thread-90) [789edb23]
> >> > > Running command: SyncNetworkProviderCommand internal:
> >> > > true.2018-03-15 10:05:00,228-04 DEBUG
> >> > > [org.ovirt.engine.core.dal.dbbroker.PostgresDbEngineDialect$  
> >> > PostgresSimpleJdbcCall]  
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > Compiled stored procedure. Call string is [{call
> >> > > getdcidbyexternalnetworkid(?)}]2018-03-15 10:05:00,228-04 DEBUG
> >> > > [org.ovirt.engine.core.dal.dbbroker.PostgresDbEngineDialect$  
> >> > PostgresSimpleJdbcCall]  
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > SqlCall for procedure [GetDcIdByExternalNetworkId]
> >> > > compiled2018-03-15 10:05:00,229-04 DEBUG
> >> > > [org.ovirt.engine.core.common.di.interceptor.DebugLoggingInterceptor]
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > method: runQuery, params: [GetAllExternalNetworksOnProvider,
> >> > > IdQueryParameters:{refresh='false', filtered='false'}],
> >> > > timeElapsed: 353ms2018-03-15 10:05:00,239-04 INFO
> >> > > [org.ovirt.engine.core.bll.network.dc.AddNetworkCommand]
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > Failed to Acquire Lock to object
> >> > > 'EngineLock:{exclusiveLocks='[network_1=NETWORK,
> >> > > c38a67ec-0b48-4e6f-be85-70c700df5483=PROVIDER]',
> >> > > sharedLocks=''}'2018-03-15 10:05:00,239-04 WARN
> >> > > [org.ovirt.engine.core.bll.network.dc.AddNetworkCommand]
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > Validation of action 'AddNetwork' failed for user
> >> > > admin at internal-authz. Reasons:
> >> > > VAR__TYPE__NETWORK,VAR__ACTION__ADD,ACTION_TYPE_FAILED_PROVIDER_LOCKED,$  
> >> > providerId  
> >> > > c38a67ec-0b48-4e6f-be85-70c700df54832018-03-15 10:05:00,240-04
> >> > > DEBUG
> >> > > [org.ovirt.engine.core.common.di.interceptor.DebugLoggingInterceptor]
> >> > > (default task-13) [e1328379-17b7-49f8-beb2-cf8331784828]
> >> > > method: runAction, params: [AddNetwork,
> >> > > AddNetworkStoragePoolParameters:{commandId='61b365ec-27c1-  
> >> > 49af-ad72-f907df8befcd',  
> >> > > user='null', commandType='Unknown'}], timeElapsed:
> >> > > 10ms2018-03-15 10:05:00,250-04 ERROR
> >> > > [org.ovirt.engine.api.restapi.resource.AbstractBackendResource]
> >> > > (default task-13) [] Operation Failed: [Cannot add Network.
> >> > > Related operation on provider with the id
> >> > > c38a67ec-0b48-4e6f-be85-70c700df5483 is currently in progress.
> >> > > Please try again later.]2018-03-15 10:05:00,254-04 DEBUG
> >> > > [org.ovirt.engine.core.utils.servlet.LocaleFilter] (default
> >> > > task-14) [] Incoming locale 'en-US'. Filter determined locale
> >> > > to be 'en-US'2018-03-15 10:05:00,254-04 DEBUG
> >> > > [org.ovirt.engine.core.sso.servlets.OAuthTokenServlet] (default
> >> > > task-14) [] Entered OAuthTokenServlet Query String: null,
> >> > > Parameters : password = ***, grant_type = password, scope =
> >> > > ovirt-app-api ovirt-ext=token-info:validate, username =
> >> > > admin at internal, </error>*  
> >> >
> >> > I will care about this.
> >> > The problem is that SyncNetworkProviderCommand is running in the
> >> > background and locking the provider, which blocks the lock for
> >> > the tested AddNetworkCommand.
> >> > The related changes are
> >> > core: Add locking for Add and RemoveNetworkCommand
> >> > https://gerrit.ovirt.org/#/c/85480/
> >> > and
> >> > core: Add SyncNetworkProviderCommand
> >> > https://gerrit.ovirt.org/#/c/85134/
> >> >
> >> >
> >> >  
> >
> > _______________________________________________
> > Devel mailing list
> > Devel at ovirt.org
> > http://lists.ovirt.org/mailman/listinfo/devel
> >
> >  



More information about the Devel mailing list