Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Paul Groeneweg | Pazion 
When upgrading the hosts
I did add another Cluster 3.6 to be able to keep my VMs running ( as I
couldn't connect el6 and el7 hosts in 1 cluster ).
I was afaraid to lose my hosted-engine during upgrade, so kept
hosted-engine as long as possible on old version.
I then switched old Cluster to 3.6 and moved all back as I had problems to
get the hosted-engine over to the new 3.6 Cluster.

So I guess I ran some steps in wrong order?
Which resulted in some missing configs/setup /storagePool...?


Op di 9 feb. 2016 om 11:51 schreef Paul Groeneweg | Pazion :

> For the record, my hosted-engine still runs el6.
> The hosts now run el7.
> This means I need to save backup and restore to a new instance?
> I do need to upgrade hosted-engine to el7?
>
> I don't get the overall picture, but what I find strange is, the spUUID in
> my hosted-engine.conf is not found. When I check the NFS hosted-engine
> storageDomain it reports this same pool ID. Which in my opinion is the
> reason for the failed ovf / vm.conf auto-import on shared storage is.
>
> So to get this correct I need to backup and reinstall hosted-engine to get
> the hosted engine working on my reinstalled hosts now runnign el7?
>
> Op di 9 feb. 2016 om 11:45 schreef Simone Tiraboschi  >:
>
>> On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>> The original (first host) was removed ( about a year a go I removed this
>>> one ) and it is also removed from hosted-engine --vm-status overview.
>>>
>>> I ran 'hosted-engine --deploy' again on all the hosts because I
>>> upgraded all hosts to EL7.
>>>
>>
>> Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
>> additional steps:
>> https://bugzilla.redhat.com/show_bug.cgi?id=1188679
>>
>> Adding Didi and Rafael who are working on that flow.
>>
>>
>>>
>>> ( I have als ran engine-setup for the upgrade )
>>>
>>> How can I fix this missing spUUID issue?
>>> Can I manually add this somehow or run somsetup again to fix this?
>>>
>>>
>>> Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion  wrote:

> I have added 2 logs from 2 hosts
>
>
 Both the hosts are just additional host:
 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
 human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
 dialog.__logString:219 DIALOG:SEND The specified storage
 location already contains a data domain. Is this an additional host setup
 (Yes, No)[Yes]?
 2016-02-07 20:32:37 INFO
 otopi.plugins.ovirt_hosted_engine_setup.storage.storage
 storage._handleHostId:249 Installing on additional host

 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
 human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
 dialog.__logString:219 DIALOG:SEND The specified storage
 location already contains a data domain. Is this an additional host setup
 (Yes, No)[Yes]?
 2016-02-05 23:47:07 INFO
 otopi.plugins.ovirt_hosted_engine_setup.storage.storage
 storage._handleHostId:123 Installing on additional host

 Did you redeployed also the engine VM on a fresh storage domain or di
 you just run 'hosted-engine --deploy' again on all the hosts?
 If it's just an upgrade you don't have to run 'hosted-engine --deploy'
 again.

 Please follow this guide:
 http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine



>
> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
> stira...@redhat.com>:
>
>> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
>> p...@pazion.nl> wrote:
>>
>>>
>>> I found vdsClient and ran a few commands.
>>>
>>> [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
>>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>>
>>> uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>>
>>> version = 3
>>>
>>> role = Master
>>>
>>> remotePath = hostedstorage.pazion.nl:/opt/hosted-engine
>>>
>>> type = NFS
>>>
>>> class = Data
>>>
>>> pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']
>>>
>>> name = hostedengine_nfs
>>>
>>>
>>>
>> Can you please attach hosted-engine-setup logs? at the end of the
>> setup procedure the hosted-engine setup shouldn't b attached to any
>> storagePool.
>>
>>
>>> [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
>>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>>
>>> Error using command: list index out of range
>>>
>>>
>>> ...
>>>
>>>
>>> [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
>>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>>
>>> Unknown pool id,

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Tue, Feb 9, 2016 at 12:10 PM, Simone Tiraboschi 
wrote:

>
>
> On Tue, Feb 9, 2016 at 11:57 AM, Paul Groeneweg | Pazion 
> wrote:
>
>>
>> When upgrading the hosts
>> I did add another Cluster 3.6 to be able to keep my VMs running ( as I
>> couldn't connect el6 and el7 hosts in 1 cluster ).
>>
>
> The approach was right since you cannot mix hosts with el6 and el7 in the
> same cluster.
> The issue is that in your hosted-engine answerfile there was the name of
> the original 3.5 cluster so hosted-engine-setup will try to add it again
> there.
> Did you manually edited the answerfile to to have it pointing to the new
> 3.6 cluster?
> How are you passing it to hosted-engine? I mean, are you downloading it
> from some other host or are you appending it with --config-append?
>

can you please share /etc/ovirt-hosted-engine/hosted-engine.conf and
ovirt-ha-agent logs from one of your upgraded hosts?


>
>
>> I was afaraid to lose my hosted-engine during upgrade, so kept
>> hosted-engine as long as possible on old version.
>> I then switched old Cluster to 3.6 and moved all back as I had problems
>> to get the hosted-engine over to the new 3.6 Cluster.
>>
>> So I guess I ran some steps in wrong order?
>> Which resulted in some missing configs/setup /storagePool...?
>>
>>
>> Op di 9 feb. 2016 om 11:51 schreef Paul Groeneweg | Pazion <
>> p...@pazion.nl>:
>>
>>> For the record, my hosted-engine still runs el6.
>>> The hosts now run el7.
>>> This means I need to save backup and restore to a new instance?
>>> I do need to upgrade hosted-engine to el7?
>>>
>>> I don't get the overall picture, but what I find strange is, the spUUID
>>> in my hosted-engine.conf is not found. When I check the NFS hosted-engine
>>> storageDomain it reports this same pool ID. Which in my opinion is the
>>> reason for the failed ovf / vm.conf auto-import on shared storage is.
>>>
>>> So to get this correct I need to backup and reinstall hosted-engine to
>>> get the hosted engine working on my reinstalled hosts now runnign el7?
>>>
>>> Op di 9 feb. 2016 om 11:45 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion <
 p...@pazion.nl> wrote:

> The original (first host) was removed ( about a year a go I removed
> this one ) and it is also removed from hosted-engine --vm-status overview.
>
> I ran 'hosted-engine --deploy' again on all the hosts because I
> upgraded all hosts to EL7.
>

 Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
 additional steps:
 https://bugzilla.redhat.com/show_bug.cgi?id=1188679

 Adding Didi and Rafael who are working on that flow.


>
> ( I have als ran engine-setup for the upgrade )
>
> How can I fix this missing spUUID issue?
> Can I manually add this somehow or run somsetup again to fix this?
>
>
> Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi <
> stira...@redhat.com>:
>
>> On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion <
>> p...@pazion.nl> wrote:
>>
>>> I have added 2 logs from 2 hosts
>>>
>>>
>> Both the hosts are just additional host:
>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>> dialog.__logString:219 DIALOG:SEND The specified storage
>> location already contains a data domain. Is this an additional host setup
>> (Yes, No)[Yes]?
>> 2016-02-07 20:32:37 INFO
>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>> storage._handleHostId:249 Installing on additional host
>>
>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>> dialog.__logString:219 DIALOG:SEND The specified storage
>> location already contains a data domain. Is this an additional host setup
>> (Yes, No)[Yes]?
>> 2016-02-05 23:47:07 INFO
>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>> storage._handleHostId:123 Installing on additional host
>>
>> Did you redeployed also the engine VM on a fresh storage domain or di
>> you just run 'hosted-engine --deploy' again on all the hosts?
>> If it's just an upgrade you don't have to run 'hosted-engine
>> --deploy' again.
>>
>> Please follow this guide:
>> http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine
>>
>>
>>
>>>
>>> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
 p...@pazion.nl> wrote:

>
> I found vdsClient and ran a few commands.

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion 
wrote:

> I have added 2 logs from 2 hosts
>
>
Both the hosts are just additional host:
2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
dialog.__logString:219 DIALOG:SEND The specified storage
location already contains a data domain. Is this an additional host setup
(Yes, No)[Yes]?
2016-02-07 20:32:37 INFO
otopi.plugins.ovirt_hosted_engine_setup.storage.storage
storage._handleHostId:249 Installing on additional host

2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
dialog.__logString:219 DIALOG:SEND The specified storage
location already contains a data domain. Is this an additional host setup
(Yes, No)[Yes]?
2016-02-05 23:47:07 INFO
otopi.plugins.ovirt_hosted_engine_setup.storage.storage
storage._handleHostId:123 Installing on additional host

Did you redeployed also the engine VM on a fresh storage domain or di you
just run 'hosted-engine --deploy' again on all the hosts?
If it's just an upgrade you don't have to run 'hosted-engine --deploy'
again.

Please follow this guide:
http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine



>
> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi  >:
>
>> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>>
>>> I found vdsClient and ran a few commands.
>>>
>>> [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
>>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>>
>>> uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>>
>>> version = 3
>>>
>>> role = Master
>>>
>>> remotePath = hostedstorage.pazion.nl:/opt/hosted-engine
>>>
>>> type = NFS
>>>
>>> class = Data
>>>
>>> pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']
>>>
>>> name = hostedengine_nfs
>>>
>>>
>>>
>> Can you please attach hosted-engine-setup logs? at the end of the setup
>> procedure the hosted-engine setup shouldn't b attached to any storagePool.
>>
>>
>>> [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
>>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>>
>>> Error using command: list index out of range
>>>
>>>
>>> ...
>>>
>>>
>>> [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
>>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>>
>>> Unknown pool id, pool not connected:
>>> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)
>>>
>>>
>>> [root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList
>>>
>>> 0002-0002-0002-0002-0385
>>>
>>>
>>> Can I manually create the missing storage pool which is missing with:
>>>
>>> vdsClient  -s 0 createStoragePool 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>> hostedengine_nfs 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>>
>>> If this would be the solution, I still wonder why there is no existing
>>> storagePool for the storageDomain.
>>>
>>>
>>>
>>> Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion <
>>> p...@pazion.nl>:
>>>
 I still keep getting

 The Hosted Engine Storage Domain doesn't exist. It should be imported
 into the setup.

 I went digging a little further and found this in the engine.log

 2016-02-08 22:30:25,083 INFO
  
 [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
 (org.ovirt.thread.pool-8-thread-41) [] FINISH,
 HSMGetStorageDomainInfoVDSCommand, return:
 , log id: 56723427
 2016-02-08 22:30:25,083 INFO
  [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
 (org.ovirt.thread.pool-8-thread-41) [] FINISH,
 GetExistingStorageDomainListQuery, log id: 6ca437b1
 2016-02-08 22:30:25,084 INFO
  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
 (org.ovirt.thread.pool-8-thread-41) [364719ce] Lock Acquired to object
 'EngineLock:{exclusiveLocks='[]', sharedLocks='null'}'
 2016-02-08 22:30:25,102 WARN
  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
 (org.ovirt.thread.pool-8-thread-41) [364719ce] CanDoAction of action
 'ImportHostedEngineStorageDomain' failed for user SYSTEM. Reasons:
 VAR__ACTION__ADD,VAR__TYPE__STORAGE__DOMAIN,ACTION_TYPE_FAILED_STORAGE_DOMAIN_NOT_EXIST

 So it looks like it can't find the storage domain. When I check
 vdsm.log I found an error:
 Thread-711902::ERROR::2016-02-08
 21:17:01,240::dispatcher::76::Storage.Dispatcher::(wrapper) {'status':
 {'message': "Unknown pool id, pool not connected:
 ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)", 'code': 309}}

 It looks like 499b208c-9de9-4a2a-97de-30f410b4e6d4 is the id it is also
 trying to connect to in the engine.log.

 I

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Tue, Feb 9, 2016 at 11:57 AM, Paul Groeneweg | Pazion 
wrote:

>
> When upgrading the hosts
> I did add another Cluster 3.6 to be able to keep my VMs running ( as I
> couldn't connect el6 and el7 hosts in 1 cluster ).
>

The approach was right since you cannot mix hosts with el6 and el7 in the
same cluster.
The issue is that in your hosted-engine answerfile there was the name of
the original 3.5 cluster so hosted-engine-setup will try to add it again
there.
Did you manually edited the answerfile to to have it pointing to the new
3.6 cluster?
How are you passing it to hosted-engine? I mean, are you downloading it
from some other host or are you appending it with --config-append?


> I was afaraid to lose my hosted-engine during upgrade, so kept
> hosted-engine as long as possible on old version.
> I then switched old Cluster to 3.6 and moved all back as I had problems to
> get the hosted-engine over to the new 3.6 Cluster.
>
> So I guess I ran some steps in wrong order?
> Which resulted in some missing configs/setup /storagePool...?
>
>
> Op di 9 feb. 2016 om 11:51 schreef Paul Groeneweg | Pazion  >:
>
>> For the record, my hosted-engine still runs el6.
>> The hosts now run el7.
>> This means I need to save backup and restore to a new instance?
>> I do need to upgrade hosted-engine to el7?
>>
>> I don't get the overall picture, but what I find strange is, the spUUID
>> in my hosted-engine.conf is not found. When I check the NFS hosted-engine
>> storageDomain it reports this same pool ID. Which in my opinion is the
>> reason for the failed ovf / vm.conf auto-import on shared storage is.
>>
>> So to get this correct I need to backup and reinstall hosted-engine to
>> get the hosted engine working on my reinstalled hosts now runnign el7?
>>
>> Op di 9 feb. 2016 om 11:45 schreef Simone Tiraboschi > >:
>>
>>> On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion >> > wrote:
>>>
 The original (first host) was removed ( about a year a go I removed
 this one ) and it is also removed from hosted-engine --vm-status overview.

 I ran 'hosted-engine --deploy' again on all the hosts because I
 upgraded all hosts to EL7.

>>>
>>> Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
>>> additional steps:
>>> https://bugzilla.redhat.com/show_bug.cgi?id=1188679
>>>
>>> Adding Didi and Rafael who are working on that flow.
>>>
>>>

 ( I have als ran engine-setup for the upgrade )

 How can I fix this missing spUUID issue?
 Can I manually add this somehow or run somsetup again to fix this?


 Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi <
 stira...@redhat.com>:

> On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion <
> p...@pazion.nl> wrote:
>
>> I have added 2 logs from 2 hosts
>>
>>
> Both the hosts are just additional host:
> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
> dialog.__logString:219 DIALOG:SEND The specified storage
> location already contains a data domain. Is this an additional host setup
> (Yes, No)[Yes]?
> 2016-02-07 20:32:37 INFO
> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
> storage._handleHostId:249 Installing on additional host
>
> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
> dialog.__logString:219 DIALOG:SEND The specified storage
> location already contains a data domain. Is this an additional host setup
> (Yes, No)[Yes]?
> 2016-02-05 23:47:07 INFO
> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
> storage._handleHostId:123 Installing on additional host
>
> Did you redeployed also the engine VM on a fresh storage domain or di
> you just run 'hosted-engine --deploy' again on all the hosts?
> If it's just an upgrade you don't have to run 'hosted-engine --deploy'
> again.
>
> Please follow this guide:
> http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine
>
>
>
>>
>> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
>> stira...@redhat.com>:
>>
>>> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
>>> p...@pazion.nl> wrote:
>>>

 I found vdsClient and ran a few commands.

 [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

 uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

 version = 3

 role = Master

 remotePath = hostedstorage.pazion.nl:/opt/hosted-engine

 type = NFS

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion 
wrote:

> The original (first host) was removed ( about a year a go I removed this
> one ) and it is also removed from hosted-engine --vm-status overview.
>
> I ran 'hosted-engine --deploy' again on all the hosts because I upgraded
> all hosts to EL7.
>

Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
additional steps:
https://bugzilla.redhat.com/show_bug.cgi?id=1188679

Adding Didi and Rafael who are working on that flow.


>
> ( I have als ran engine-setup for the upgrade )
>
> How can I fix this missing spUUID issue?
> Can I manually add this somehow or run somsetup again to fix this?
>
>
> Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi  >:
>
>> On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>> I have added 2 logs from 2 hosts
>>>
>>>
>> Both the hosts are just additional host:
>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>> dialog.__logString:219 DIALOG:SEND The specified storage
>> location already contains a data domain. Is this an additional host setup
>> (Yes, No)[Yes]?
>> 2016-02-07 20:32:37 INFO
>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>> storage._handleHostId:249 Installing on additional host
>>
>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>> dialog.__logString:219 DIALOG:SEND The specified storage
>> location already contains a data domain. Is this an additional host setup
>> (Yes, No)[Yes]?
>> 2016-02-05 23:47:07 INFO
>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>> storage._handleHostId:123 Installing on additional host
>>
>> Did you redeployed also the engine VM on a fresh storage domain or di you
>> just run 'hosted-engine --deploy' again on all the hosts?
>> If it's just an upgrade you don't have to run 'hosted-engine --deploy'
>> again.
>>
>> Please follow this guide:
>> http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine
>>
>>
>>
>>>
>>> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
 p...@pazion.nl> wrote:

>
> I found vdsClient and ran a few commands.
>
> [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> version = 3
>
> role = Master
>
> remotePath = hostedstorage.pazion.nl:/opt/hosted-engine
>
> type = NFS
>
> class = Data
>
> pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']
>
> name = hostedengine_nfs
>
>
>
 Can you please attach hosted-engine-setup logs? at the end of the setup
 procedure the hosted-engine setup shouldn't b attached to any storagePool.


> [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>
> Error using command: list index out of range
>
>
> ...
>
>
> [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>
> Unknown pool id, pool not connected:
> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)
>
>
> [root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList
>
> 0002-0002-0002-0002-0385
>
>
> Can I manually create the missing storage pool which is missing with:
>
> vdsClient  -s 0 createStoragePool
> 499b208c-9de9-4a2a-97de-30f410b4e6d4 hostedengine_nfs
> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> If this would be the solution, I still wonder why there is no existing
> storagePool for the storageDomain.
>
>
>
> Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion <
> p...@pazion.nl>:
>
>> I still keep getting
>>
>> The Hosted Engine Storage Domain doesn't exist. It should be imported
>> into the setup.
>>
>> I went digging a little further and found this in the engine.log
>>
>> 2016-02-08 22:30:25,083 INFO
>>  
>> [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
>> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
>> HSMGetStorageDomainInfoVDSCommand, return:
>> > id='88b69eba-ef4f-4dbe-ba53-20dadd424d0e'},
>> 499b208c-9de9-4a2a-97de-30f410b4e6d4>, log id: 56723427
>> 2016-02-08 22:30:25,083 INFO
>>  [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
>> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
>> GetExistingStorageDomainListQuery, log id: 6ca437b1
>>

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Paul Groeneweg | Pazion 
The original (first host) was removed ( about a year a go I removed this
one ) and it is also removed from hosted-engine --vm-status overview.

I ran 'hosted-engine --deploy' again on all the hosts because I upgraded
all hosts to EL7.
( I have als ran engine-setup for the upgrade )

How can I fix this missing spUUID issue?
Can I manually add this somehow or run somsetup again to fix this?


Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi :

> On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion 
> wrote:
>
>> I have added 2 logs from 2 hosts
>>
>>
> Both the hosts are just additional host:
> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
> dialog.__logString:219 DIALOG:SEND The specified storage
> location already contains a data domain. Is this an additional host setup
> (Yes, No)[Yes]?
> 2016-02-07 20:32:37 INFO
> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
> storage._handleHostId:249 Installing on additional host
>
> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
> dialog.__logString:219 DIALOG:SEND The specified storage
> location already contains a data domain. Is this an additional host setup
> (Yes, No)[Yes]?
> 2016-02-05 23:47:07 INFO
> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
> storage._handleHostId:123 Installing on additional host
>
> Did you redeployed also the engine VM on a fresh storage domain or di you
> just run 'hosted-engine --deploy' again on all the hosts?
> If it's just an upgrade you don't have to run 'hosted-engine --deploy'
> again.
>
> Please follow this guide:
> http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine
>
>
>
>>
>> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi > >:
>>
>>> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion >> > wrote:
>>>

 I found vdsClient and ran a few commands.

 [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

 uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

 version = 3

 role = Master

 remotePath = hostedstorage.pazion.nl:/opt/hosted-engine

 type = NFS

 class = Data

 pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']

 name = hostedengine_nfs



>>> Can you please attach hosted-engine-setup logs? at the end of the setup
>>> procedure the hosted-engine setup shouldn't b attached to any storagePool.
>>>
>>>
 [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
 499b208c-9de9-4a2a-97de-30f410b4e6d4

 Error using command: list index out of range


 ...


 [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
 499b208c-9de9-4a2a-97de-30f410b4e6d4

 Unknown pool id, pool not connected:
 ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)


 [root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList

 0002-0002-0002-0002-0385


 Can I manually create the missing storage pool which is missing with:

 vdsClient  -s 0 createStoragePool 499b208c-9de9-4a2a-97de-30f410b4e6d4
 hostedengine_nfs 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

 If this would be the solution, I still wonder why there is no existing
 storagePool for the storageDomain.



 Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion <
 p...@pazion.nl>:

> I still keep getting
>
> The Hosted Engine Storage Domain doesn't exist. It should be imported
> into the setup.
>
> I went digging a little further and found this in the engine.log
>
> 2016-02-08 22:30:25,083 INFO
>  
> [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
> HSMGetStorageDomainInfoVDSCommand, return:
>  id='88b69eba-ef4f-4dbe-ba53-20dadd424d0e'},
> 499b208c-9de9-4a2a-97de-30f410b4e6d4>, log id: 56723427
> 2016-02-08 22:30:25,083 INFO
>  [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
> GetExistingStorageDomainListQuery, log id: 6ca437b1
> 2016-02-08 22:30:25,084 INFO
>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
> (org.ovirt.thread.pool-8-thread-41) [364719ce] Lock Acquired to object
> 'EngineLock:{exclusiveLocks='[]', sharedLocks='null'}'
> 2016-02-08 22:30:25,102 WARN
>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
> (org.ovirt.thread.pool-8-thread-41) [364719ce] CanDoAction of action
>

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Paul Groeneweg | Pazion 
For the record, my hosted-engine still runs el6.
The hosts now run el7.
This means I need to save backup and restore to a new instance?
I do need to upgrade hosted-engine to el7?

I don't get the overall picture, but what I find strange is, the spUUID in
my hosted-engine.conf is not found. When I check the NFS hosted-engine
storageDomain it reports this same pool ID. Which in my opinion is the
reason for the failed ovf / vm.conf auto-import on shared storage is.

So to get this correct I need to backup and reinstall hosted-engine to get
the hosted engine working on my reinstalled hosts now runnign el7?

Op di 9 feb. 2016 om 11:45 schreef Simone Tiraboschi :

> On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion 
> wrote:
>
>> The original (first host) was removed ( about a year a go I removed this
>> one ) and it is also removed from hosted-engine --vm-status overview.
>>
>> I ran 'hosted-engine --deploy' again on all the hosts because I upgraded
>> all hosts to EL7.
>>
>
> Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
> additional steps:
> https://bugzilla.redhat.com/show_bug.cgi?id=1188679
>
> Adding Didi and Rafael who are working on that flow.
>
>
>>
>> ( I have als ran engine-setup for the upgrade )
>>
>> How can I fix this missing spUUID issue?
>> Can I manually add this somehow or run somsetup again to fix this?
>>
>>
>> Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi > >:
>>
>>> On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion 
>>> wrote:
>>>
 I have added 2 logs from 2 hosts


>>> Both the hosts are just additional host:
>>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>>> 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
>>> dialog.__logString:219 DIALOG:SEND The specified storage
>>> location already contains a data domain. Is this an additional host setup
>>> (Yes, No)[Yes]?
>>> 2016-02-07 20:32:37 INFO
>>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>>> storage._handleHostId:249 Installing on additional host
>>>
>>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>>> human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
>>> 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
>>> dialog.__logString:219 DIALOG:SEND The specified storage
>>> location already contains a data domain. Is this an additional host setup
>>> (Yes, No)[Yes]?
>>> 2016-02-05 23:47:07 INFO
>>> otopi.plugins.ovirt_hosted_engine_setup.storage.storage
>>> storage._handleHostId:123 Installing on additional host
>>>
>>> Did you redeployed also the engine VM on a fresh storage domain or di
>>> you just run 'hosted-engine --deploy' again on all the hosts?
>>> If it's just an upgrade you don't have to run 'hosted-engine --deploy'
>>> again.
>>>
>>> Please follow this guide:
>>> http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine
>>>
>>>
>>>

 Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
 stira...@redhat.com>:

> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
> p...@pazion.nl> wrote:
>
>>
>> I found vdsClient and ran a few commands.
>>
>> [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>
>> uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>
>> version = 3
>>
>> role = Master
>>
>> remotePath = hostedstorage.pazion.nl:/opt/hosted-engine
>>
>> type = NFS
>>
>> class = Data
>>
>> pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']
>>
>> name = hostedengine_nfs
>>
>>
>>
> Can you please attach hosted-engine-setup logs? at the end of the
> setup procedure the hosted-engine setup shouldn't b attached to any
> storagePool.
>
>
>> [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>
>> Error using command: list index out of range
>>
>>
>> ...
>>
>>
>> [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
>> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>>
>> Unknown pool id, pool not connected:
>> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)
>>
>>
>> [root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList
>>
>> 0002-0002-0002-0002-0385
>>
>>
>> Can I manually create the missing storage pool which is missing with:
>>
>> vdsClient  -s 0 createStoragePool
>> 499b208c-9de9-4a2a-97de-30f410b4e6d4 hostedengine_nfs
>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>>
>> If this would be the solution, I still wonder why there is no
>> existing storagePool for the storageDomain.
>>
>>
>>
>> Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion <
>> p...@pazion.nl>:
>>
>>> I still keep

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion 
wrote:

>
> I found vdsClient and ran a few commands.
>
> [root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> version = 3
>
> role = Master
>
> remotePath = hostedstorage.pazion.nl:/opt/hosted-engine
>
> type = NFS
>
> class = Data
>
> pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']
>
> name = hostedengine_nfs
>
>
>
Can you please attach hosted-engine-setup logs? at the end of the setup
procedure the hosted-engine setup shouldn't b attached to any storagePool.


> [root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>
> Error using command: list index out of range
>
>
> ...
>
>
> [root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
> 499b208c-9de9-4a2a-97de-30f410b4e6d4
>
> Unknown pool id, pool not connected:
> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)
>
>
> [root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList
>
> 0002-0002-0002-0002-0385
>
>
> Can I manually create the missing storage pool which is missing with:
>
> vdsClient  -s 0 createStoragePool 499b208c-9de9-4a2a-97de-30f410b4e6d4
> hostedengine_nfs 88b69eba-ef4f-4dbe-ba53-20dadd424d0e
>
> If this would be the solution, I still wonder why there is no existing
> storagePool for the storageDomain.
>
>
>
> Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion  >:
>
>> I still keep getting
>>
>> The Hosted Engine Storage Domain doesn't exist. It should be imported
>> into the setup.
>>
>> I went digging a little further and found this in the engine.log
>>
>> 2016-02-08 22:30:25,083 INFO
>>  
>> [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
>> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
>> HSMGetStorageDomainInfoVDSCommand, return:
>> > id='88b69eba-ef4f-4dbe-ba53-20dadd424d0e'},
>> 499b208c-9de9-4a2a-97de-30f410b4e6d4>, log id: 56723427
>> 2016-02-08 22:30:25,083 INFO
>>  [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
>> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
>> GetExistingStorageDomainListQuery, log id: 6ca437b1
>> 2016-02-08 22:30:25,084 INFO
>>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
>> (org.ovirt.thread.pool-8-thread-41) [364719ce] Lock Acquired to object
>> 'EngineLock:{exclusiveLocks='[]', sharedLocks='null'}'
>> 2016-02-08 22:30:25,102 WARN
>>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
>> (org.ovirt.thread.pool-8-thread-41) [364719ce] CanDoAction of action
>> 'ImportHostedEngineStorageDomain' failed for user SYSTEM. Reasons:
>> VAR__ACTION__ADD,VAR__TYPE__STORAGE__DOMAIN,ACTION_TYPE_FAILED_STORAGE_DOMAIN_NOT_EXIST
>>
>> So it looks like it can't find the storage domain. When I check vdsm.log
>> I found an error:
>> Thread-711902::ERROR::2016-02-08
>> 21:17:01,240::dispatcher::76::Storage.Dispatcher::(wrapper) {'status':
>> {'message': "Unknown pool id, pool not connected:
>> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)", 'code': 309}}
>>
>> It looks like 499b208c-9de9-4a2a-97de-30f410b4e6d4 is the id it is also
>> trying to connect to in the engine.log.
>>
>> I did found https://bugzilla.redhat.com/show_bug.cgi?id=1269768  and
>> https://bugzilla.redhat.com/show_bug.cgi?id=1294457
>>
>> Can it be I have setup another DC with a master iscsi storage  which is
>> now offline?  ( 1 DC with Cluster,  VMs with hosted engine and master
>> storage FC all up and running.  )
>>
>> Not sure where to go from here. Can I somehow restart this storage domain
>> / do a new hosted-engine deploy with different settings?
>>
>>
>> Op ma 8 feb. 2016 om 12:05 schreef Paul Groeneweg | Pazion <
>> p...@pazion.nl>:
>>
>>> Op ma 8 feb. 2016 om 11:28 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Mon, Feb 8, 2016 at 10:55 AM, Paul Groeneweg | Pazion <
 p...@pazion.nl> wrote:

> Thanks for your response.
>
> I have reverted the commented entries and restarted ovirt-ha-agent.
> Now status on the daemon gives the error again:
> http://screencast.com/t/DCZo6QJM1
>
>
 Here I'm a bit confused: is this a fresh hosted-engine deployment with
 3.6 or an upgrade from an existing 3.5 hosted-engine deployment?

>>>
>>> I have  reinstalled host ( complete OS reinstall) and ran hosted-engine
>>> --deploy for redeployment on an existing oVirt ( hosted-engine ).
>>>
>>> As described in previous mail, during the deployment I got errors about
>>> a missing hostedengine.metadata symlink to /var/run/vdsd/storage/...
>>> I created these by hand and then I ran deployment again and it worked.
>>>
>>>

> The hosted engine domain is not shown in the web storage overview...
> When I start import it shows a warning:
> http://screencast.com/t/7HfJ9qWFwb2L
>
> Can I manually import the hostedengine

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-09 Thread Simone Tiraboschi 
On Tue, Feb 9, 2016 at 11:51 AM, Paul Groeneweg | Pazion 
wrote:

> For the record, my hosted-engine still runs el6.
>

You can run the 3.6 engine also on el6 so no issue there.


> The hosts now run el7.
>

To reach 3.6 cluster compatibility level you need to upgrade all the hosts
to el7.

Now I don't understand where the engine VM is running.
In order to be able to (re)deploy the host you upgraded to el7 you need a
running engine.
Is your engine VM running?
If so, on which host?


> This means I need to save backup and restore to a new instance?
> I do need to upgrade hosted-engine to el7?
>
> I don't get the overall picture, but what I find strange is, the spUUID in
> my hosted-engine.conf is not found. When I check the NFS hosted-engine
> storageDomain it reports this same pool ID. Which in my opinion is the
> reason for the failed ovf / vm.conf auto-import on shared storage is.
>

This is a bit tricky:
when you setup hosted-engine on 3.4 or 3.5, the hosted-engine storage
domain was on a dedicated (boostrap) storage pool that nobody was really
using.
In order to be able to manage and edit the engine VM from the engine, on
3.6 the hosted-engine storage domain should be attached to the storagePool
of the datacenter as seen by the engine but to do that the hosted-engine
storage domain should be detached from its bootstrap storage pool.

The flow is this:
1. put everything into maintenance
2. upgrade ovirt-ha-agent to 3.6
3. restart ovirt-ha-agent: the new agent will
- create a new volume with all the hosted-engine configuration on the
shared volume
- detach hosted-engine storage domain from its boostrap storage pool
4. when all the hosts are at 3.6, you can set 3.6 cluster compatibility
level into the engine; this will trigger the autoimport of the
hosted-engine storage domain from the engine and the hosted-engine storage
domain will be attached to the datacenter storage pool.

In your case is a bit different since your hosts was with el6 and so you
cannot upgrade ovirt-ha-agent to 3.6.


>
> So to get this correct I need to backup and reinstall hosted-engine to get
> the hosted engine working on my reinstalled hosts now runnign el7?
>
> Op di 9 feb. 2016 om 11:45 schreef Simone Tiraboschi  >:
>
>> On Tue, Feb 9, 2016 at 10:53 AM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>> The original (first host) was removed ( about a year a go I removed this
>>> one ) and it is also removed from hosted-engine --vm-status overview.
>>>
>>> I ran 'hosted-engine --deploy' again on all the hosts because I
>>> upgraded all hosts to EL7.
>>>
>>
>> Upgrading hosted-engine from 3.5 on el6 to 3.6 on el7 requires some
>> additional steps:
>> https://bugzilla.redhat.com/show_bug.cgi?id=1188679
>>
>> Adding Didi and Rafael who are working on that flow.
>>
>>
>>>
>>> ( I have als ran engine-setup for the upgrade )
>>>
>>> How can I fix this missing spUUID issue?
>>> Can I manually add this somehow or run somsetup again to fix this?
>>>
>>>
>>> Op di 9 feb. 2016 om 10:43 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>
 On Tue, Feb 9, 2016 at 9:45 AM, Paul Groeneweg | Pazion  wrote:

> I have added 2 logs from 2 hosts
>
>
 Both the hosts are just additional host:
 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
 human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
 2016-02-07 20:32:30 DEBUG otopi.plugins.otopi.dialog.human
 dialog.__logString:219 DIALOG:SEND The specified storage
 location already contains a data domain. Is this an additional host setup
 (Yes, No)[Yes]?
 2016-02-07 20:32:37 INFO
 otopi.plugins.ovirt_hosted_engine_setup.storage.storage
 storage._handleHostId:249 Installing on additional host

 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
 human.queryString:156 query OVEHOSTED_ADDITIONAL_HOST
 2016-02-05 23:47:05 DEBUG otopi.plugins.otopi.dialog.human
 dialog.__logString:219 DIALOG:SEND The specified storage
 location already contains a data domain. Is this an additional host setup
 (Yes, No)[Yes]?
 2016-02-05 23:47:07 INFO
 otopi.plugins.ovirt_hosted_engine_setup.storage.storage
 storage._handleHostId:123 Installing on additional host

 Did you redeployed also the engine VM on a fresh storage domain or di
 you just run 'hosted-engine --deploy' again on all the hosts?
 If it's just an upgrade you don't have to run 'hosted-engine --deploy'
 again.

 Please follow this guide:
 http://www.ovirt.org/Hosted_Engine_Howto#Upgrade_Hosted_Engine



>
> Op di 9 feb. 2016 om 09:07 schreef Simone Tiraboschi <
> stira...@redhat.com>:
>
>> On Mon, Feb 8, 2016 at 11:25 PM, Paul Groeneweg | Pazion <
>> p...@pazion.nl> wrote:
>>
>>>
>>> I found vdsClient and ran a few commands.
>>>
>>> [root@geisha-1 ~]#

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Paul Groeneweg | Pazion 
Thanks for your response.

I have reverted the commented entries and restarted ovirt-ha-agent.
Now status on the daemon gives the error again:
http://screencast.com/t/DCZo6QJM1

The hosted engine domain is not shown in the web storage overview...
When I start import it shows a warning: http://screencast.com/t/7HfJ9qWFwb2L

Can I manually import the hostedengine storage?
( Hosted engine is visible in web interface and I have running VMs on
another storage domain ).


Op ma 8 feb. 2016 om 10:36 schreef Simone Tiraboschi :

>
>
> On Mon, Feb 8, 2016 at 9:47 AM, Paul Groeneweg | Pazion 
> wrote:
>
>> Update:
>>
>
> The OVF_STORE volume will contain the engine VM description in a format
> that the engine can easily handle.
> It will be created by the engine when:
> - it imported the hosted-engine storage domain
> - the engineVM is visible
> - the datacenter is up (you need at least one additional regular storage
> domain)
>
> Please make sure that all these conditions are met.
>
> If the OVF_STORE is not there or it's not accessible, ovirt-ha-agent will
> fallback to the initial vm.conf from deployment time. You see a warning
> since we are failing back to an initial status and this can temporary
> override your latest customization to the engine VM configuration.
>
>
>> 1. I was able to remove the warnirngs about the not found ovf / vm.conf
>> by commenting the conf_volume_UUID, conf_image_UUID entries in the
>> hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
>> gone. ( I believe this is not a permanent solution ).
>>
>
> Please revert it.
>
>
>>
>> 2. This morning I noticed the warning in the event log.
>> "The Hosted Engine Storage Domain doesn't exist. It should be imported
>> into the setup."
>>
>
> It's just a side effect of your point 1, please revert it.
>
>
>>
>> Any help would be appreciated.
>>
>> Best Regards,
>>
>> Paul Groeneweg
>>
>> Op zo 7 feb. 2016 om 22:14 schreef Paul Groeneweg | Pazion <
>> p...@pazion.nl>:
>>
>>>
>>> I have been upgrading our ovirt hosts to el7 to be able to run version
>>> 3.6.
>>> When deploying a new host have problems with the shared OVF storage
>>> config part.
>>>
>>> 1. I get an error during deploy:
>>>
>>> [ INFO  ] Stage: Setup validation
>>> [ ERROR ] Failed to execute stage 'Setup validation': failed to read
>>> metadata: [Errno 2] No such file or directory:
>>> '/rhev/data-center/mnt/[NFS3_SHARE]/88b69eba-ef4f-4dbe-ba53-20dadd424d0e/ha_agent/hosted-engine.metadata'
>>> [ INFO  ] Stage: Clean up
>>>
>>> It seems the symlink of hosted-engine.metadata is missing.
>>> So I added manually the /var/run/vdsm/storage folder with subfolder
>>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e and symlinks.
>>> Then I reran the hosted-engine deploy and this time all went well.
>>>
>>> 2. Now when the host is installed I notice the following in the logs
>>>
>>> 
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Preparing
>>> images
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> INFO:ovirt_hosted_engine_ha.lib.image.Image:Preparing images
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Reloading
>>> vm.conf from the shared storage domain
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Trying
>>> to get a fresher copy of vm configuration from the OVF_STORE
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> WARNING:ovirt_hosted_engine_ha.lib.ovf.ovf_store.OVFStore:Unable to find
>>> OVF_STORE
>>> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
>>> ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config ERROR Unable
>>> to get vm.conf from OVF_STORE, falling back to initial vm.conf
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> ERROR:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Unable
>>> to get vm.conf from OVF_STORE, falling back to initial vm.conf
>>> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
>>> ovirt_hosted_engine_ha.agent.agent.Agent ERROR Error: 'Path to volume
>>> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
>>> trying to restart agent
>>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>>> ERROR:ovirt_hosted_engine_ha.agent.agent.Agent:Error: 'Path to volume
>>> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
>>> trying to restart agent
>>>
>>> So at this point I was a bit stuck.
>>> I removed complete ovirt36 release and installed ovirt35 for the hosted
>>> engine deploy.
>>>
>>> This time all went without any problem directly.
>>> From there I retried upgrade and same problems again.
>>>
>>> I run ovirt hosted-engine on nfs3. I did found
>>> https://bugzilla.redhat.com/show_bug.cgi?id=1159314
>>> I added no_subtree_check and tried deploy again, but again without luck.
>>>
>>> I believe the hosted engine configs are not being migrated to the

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Paul Groeneweg | Pazion 
Update:

1. I was able to remove the warnirngs about the not found ovf / vm.conf by
commenting the conf_volume_UUID, conf_image_UUID entries in the
hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
gone. ( I believe this is not a permanent solution ).

2. This morning I noticed the warning in the event log.
"The Hosted Engine Storage Domain doesn't exist. It should be imported into
the setup."

Any help would be appreciated.

Best Regards,

Paul Groeneweg

Op zo 7 feb. 2016 om 22:14 schreef Paul Groeneweg | Pazion :

>
> I have been upgrading our ovirt hosts to el7 to be able to run version 3.6.
> When deploying a new host have problems with the shared OVF storage config
> part.
>
> 1. I get an error during deploy:
>
> [ INFO  ] Stage: Setup validation
> [ ERROR ] Failed to execute stage 'Setup validation': failed to read
> metadata: [Errno 2] No such file or directory:
> '/rhev/data-center/mnt/[NFS3_SHARE]/88b69eba-ef4f-4dbe-ba53-20dadd424d0e/ha_agent/hosted-engine.metadata'
> [ INFO  ] Stage: Clean up
>
> It seems the symlink of hosted-engine.metadata is missing.
> So I added manually the /var/run/vdsm/storage folder with subfolder
> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e and symlinks.
> Then I reran the hosted-engine deploy and this time all went well.
>
> 2. Now when the host is installed I notice the following in the logs
>
> 
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Preparing
> images
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.lib.image.Image:Preparing images
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Reloading
> vm.conf from the shared storage domain
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Trying
> to get a fresher copy of vm configuration from the OVF_STORE
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> WARNING:ovirt_hosted_engine_ha.lib.ovf.ovf_store.OVFStore:Unable to find
> OVF_STORE
> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
> ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config ERROR Unable
> to get vm.conf from OVF_STORE, falling back to initial vm.conf
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> ERROR:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Unable
> to get vm.conf from OVF_STORE, falling back to initial vm.conf
> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
> ovirt_hosted_engine_ha.agent.agent.Agent ERROR Error: 'Path to volume
> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
> trying to restart agent
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> ERROR:ovirt_hosted_engine_ha.agent.agent.Agent:Error: 'Path to volume
> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
> trying to restart agent
>
> So at this point I was a bit stuck.
> I removed complete ovirt36 release and installed ovirt35 for the hosted
> engine deploy.
>
> This time all went without any problem directly.
> From there I retried upgrade and same problems again.
>
> I run ovirt hosted-engine on nfs3. I did found
> https://bugzilla.redhat.com/show_bug.cgi?id=1159314
> I added no_subtree_check and tried deploy again, but again without luck.
>
> I believe the hosted engine configs are not being migrated to the shared
> storage. And my problem has todo with
> http://www.ovirt.org/Features/Hosted_Engine_configuration_on_shared_storage
>
> And found  /etc/ovirt-hosted-engine/hosted-engine.conf,
> conf_volume_UUID, conf_image_UUID
>
>
> Hosts run el7
> Hosted engine runs el6
> Hosted engine storage is through NFS3
>
> Can I import the hosted engine storage in to engine ( in web interface )?
> Or how can I create or trigger manually the OVF store the logs complain
> about?
>
> Best Regards,
> Paul Groeneweg
>
> --
> Met vriendelijke groeten,
>
> Paul Groeneweg
> Pazion
> Webdevelopment  -  Hosting  -  Apps
>
> T +31 26 3020038
> M +31 614 277 577
> E  p...@pazion.nl
>
>  ***disclaimer***
> "This e-mail and any attachments thereto may contain information which is
> confidential and/or protected by intellectual property rights and are
> intended for the sole use of the recipient(s) named above. Any use of the
> information contained herein (including, but not limited to, total or
> partial reproduction, communication or distribution in any form) by persons
> other than the designated recipient(s) is prohibited. If you have received
> this e-mail in error, please notify the sender either by telephone or by
> e-mail and delete the material from any computer. Thank you for your
> cooperation."
>
-- 
Met vriendelijke groeten,

Paul Groeneweg
Pazion
Webdevelopment  -  Hosting  -  Apps

T +31 26 3020038
M +31 614 277 577
E  p...@pazion.nl

 ***disclaimer***
"This e-mail and any attachments thereto may contain information which is
confidential and/or protected by intellectual

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Simone Tiraboschi 
On Mon, Feb 8, 2016 at 9:47 AM, Paul Groeneweg | Pazion 
wrote:

> Update:
>

The OVF_STORE volume will contain the engine VM description in a format
that the engine can easily handle.
It will be created by the engine when:
- it imported the hosted-engine storage domain
- the engineVM is visible
- the datacenter is up (you need at least one additional regular storage
domain)

Please make sure that all these conditions are met.

If the OVF_STORE is not there or it's not accessible, ovirt-ha-agent will
fallback to the initial vm.conf from deployment time. You see a warning
since we are failing back to an initial status and this can temporary
override your latest customization to the engine VM configuration.


> 1. I was able to remove the warnirngs about the not found ovf / vm.conf by
> commenting the conf_volume_UUID, conf_image_UUID entries in the
> hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
> gone. ( I believe this is not a permanent solution ).
>

Please revert it.


>
> 2. This morning I noticed the warning in the event log.
> "The Hosted Engine Storage Domain doesn't exist. It should be imported
> into the setup."
>

It's just a side effect of your point 1, please revert it.


>
> Any help would be appreciated.
>
> Best Regards,
>
> Paul Groeneweg
>
> Op zo 7 feb. 2016 om 22:14 schreef Paul Groeneweg | Pazion  >:
>
>>
>> I have been upgrading our ovirt hosts to el7 to be able to run version
>> 3.6.
>> When deploying a new host have problems with the shared OVF storage
>> config part.
>>
>> 1. I get an error during deploy:
>>
>> [ INFO  ] Stage: Setup validation
>> [ ERROR ] Failed to execute stage 'Setup validation': failed to read
>> metadata: [Errno 2] No such file or directory:
>> '/rhev/data-center/mnt/[NFS3_SHARE]/88b69eba-ef4f-4dbe-ba53-20dadd424d0e/ha_agent/hosted-engine.metadata'
>> [ INFO  ] Stage: Clean up
>>
>> It seems the symlink of hosted-engine.metadata is missing.
>> So I added manually the /var/run/vdsm/storage folder with subfolder
>> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e and symlinks.
>> Then I reran the hosted-engine deploy and this time all went well.
>>
>> 2. Now when the host is installed I notice the following in the logs
>>
>> 
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Preparing
>> images
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> INFO:ovirt_hosted_engine_ha.lib.image.Image:Preparing images
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Reloading
>> vm.conf from the shared storage domain
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Trying
>> to get a fresher copy of vm configuration from the OVF_STORE
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> WARNING:ovirt_hosted_engine_ha.lib.ovf.ovf_store.OVFStore:Unable to find
>> OVF_STORE
>> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
>> ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config ERROR Unable
>> to get vm.conf from OVF_STORE, falling back to initial vm.conf
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> ERROR:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Unable
>> to get vm.conf from OVF_STORE, falling back to initial vm.conf
>> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
>> ovirt_hosted_engine_ha.agent.agent.Agent ERROR Error: 'Path to volume
>> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
>> trying to restart agent
>> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>> ERROR:ovirt_hosted_engine_ha.agent.agent.Agent:Error: 'Path to volume
>> 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
>> trying to restart agent
>>
>> So at this point I was a bit stuck.
>> I removed complete ovirt36 release and installed ovirt35 for the hosted
>> engine deploy.
>>
>> This time all went without any problem directly.
>> From there I retried upgrade and same problems again.
>>
>> I run ovirt hosted-engine on nfs3. I did found
>> https://bugzilla.redhat.com/show_bug.cgi?id=1159314
>> I added no_subtree_check and tried deploy again, but again without luck.
>>
>> I believe the hosted engine configs are not being migrated to the shared
>> storage. And my problem has todo with
>> http://www.ovirt.org/Features/Hosted_Engine_configuration_on_shared_storage
>>
>> And found  /etc/ovirt-hosted-engine/hosted-engine.conf,
>> conf_volume_UUID, conf_image_UUID
>>
>>
>> Hosts run el7
>> Hosted engine runs el6
>> Hosted engine storage is through NFS3
>>
>> Can I import the hosted engine storage in to engine ( in web interface )?
>> Or how can I create or trigger manually the OVF store the logs complain
>> about?
>>
>> Best Regards,
>> Paul Groeneweg
>>
>> --
>> Met vriendelijke groeten,
>>
>> Paul Groeneweg
>> Pazion
>> Webdevelopment  -  Hosting  -  Apps
>>
>> T +31 26 3020038
>> M +31 614 277 577
>> E

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Paul Groeneweg | Pazion 
Op ma 8 feb. 2016 om 11:28 schreef Simone Tiraboschi :

> On Mon, Feb 8, 2016 at 10:55 AM, Paul Groeneweg | Pazion 
> wrote:
>
>> Thanks for your response.
>>
>> I have reverted the commented entries and restarted ovirt-ha-agent.
>> Now status on the daemon gives the error again:
>> http://screencast.com/t/DCZo6QJM1
>>
>>
> Here I'm a bit confused: is this a fresh hosted-engine deployment with 3.6
> or an upgrade from an existing 3.5 hosted-engine deployment?
>

I have  reinstalled host ( complete OS reinstall) and ran hosted-engine
--deploy for redeployment on an existing oVirt ( hosted-engine ).

As described in previous mail, during the deployment I got errors about a
missing hostedengine.metadata symlink to /var/run/vdsd/storage/...
I created these by hand and then I ran deployment again and it worked.


>
>> The hosted engine domain is not shown in the web storage overview...
>> When I start import it shows a warning:
>> http://screencast.com/t/7HfJ9qWFwb2L
>>
>> Can I manually import the hostedengine storage?
>> ( Hosted engine is visible in web interface and I have running VMs on
>> another storage domain ).
>>
>
> It should auto-import it.
> Adding Roy here.
>
>
I believe this auto-import is failing...
Where can I find logs about this auto-import?




>
>>
>> Op ma 8 feb. 2016 om 10:36 schreef Simone Tiraboschi > >:
>>
>>>
>>>
>>> On Mon, Feb 8, 2016 at 9:47 AM, Paul Groeneweg | Pazion 
>>> wrote:
>>>
 Update:

>>>
>>> The OVF_STORE volume will contain the engine VM description in a format
>>> that the engine can easily handle.
>>> It will be created by the engine when:
>>> - it imported the hosted-engine storage domain
>>> - the engineVM is visible
>>> - the datacenter is up (you need at least one additional regular storage
>>> domain)
>>>
>>> Please make sure that all these conditions are met.
>>>
>>> If the OVF_STORE is not there or it's not accessible, ovirt-ha-agent
>>> will fallback to the initial vm.conf from deployment time. You see a
>>> warning since we are failing back to an initial status and this can
>>> temporary override your latest customization to the engine VM configuration.
>>>
>>>
 1. I was able to remove the warnirngs about the not found ovf / vm.conf
 by commenting the conf_volume_UUID, conf_image_UUID entries in the
 hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
 gone. ( I believe this is not a permanent solution ).

>>>
>>> Please revert it.
>>>
>>>

 2. This morning I noticed the warning in the event log.
 "The Hosted Engine Storage Domain doesn't exist. It should be imported
 into the setup."

>>>
>>> It's just a side effect of your point 1, please revert it.
>>>
>>>

 Any help would be appreciated.

 Best Regards,

 Paul Groeneweg

 Op zo 7 feb. 2016 om 22:14 schreef Paul Groeneweg | Pazion <
 p...@pazion.nl>:

>
> I have been upgrading our ovirt hosts to el7 to be able to run version
> 3.6.
> When deploying a new host have problems with the shared OVF storage
> config part.
>
> 1. I get an error during deploy:
>
> [ INFO  ] Stage: Setup validation
> [ ERROR ] Failed to execute stage 'Setup validation': failed to read
> metadata: [Errno 2] No such file or directory:
> '/rhev/data-center/mnt/[NFS3_SHARE]/88b69eba-ef4f-4dbe-ba53-20dadd424d0e/ha_agent/hosted-engine.metadata'
> [ INFO  ] Stage: Clean up
>
> It seems the symlink of hosted-engine.metadata is missing.
> So I added manually the /var/run/vdsm/storage folder with subfolder
> 88b69eba-ef4f-4dbe-ba53-20dadd424d0e and symlinks.
> Then I reran the hosted-engine deploy and this time all went well.
>
> 2. Now when the host is installed I notice the following in the logs
>
> 
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Preparing
> images
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.lib.image.Image:Preparing images
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Reloading
> vm.conf from the shared storage domain
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Trying
> to get a fresher copy of vm configuration from the OVF_STORE
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
> WARNING:ovirt_hosted_engine_ha.lib.ovf.ovf_store.OVFStore:Unable to find
> OVF_STORE
> Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
> ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config ERROR 
> Unable
> to get vm.conf from OVF_STORE, falling back to initial vm.conf
> Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
>

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Simone Tiraboschi 
On Mon, Feb 8, 2016 at 10:55 AM, Paul Groeneweg | Pazion 
wrote:

> Thanks for your response.
>
> I have reverted the commented entries and restarted ovirt-ha-agent.
> Now status on the daemon gives the error again:
> http://screencast.com/t/DCZo6QJM1
>
>
Here I'm a bit confused: is this a fresh hosted-engine deployment with 3.6
or an upgrade from an existing 3.5 hosted-engine deployment?


> The hosted engine domain is not shown in the web storage overview...
> When I start import it shows a warning:
> http://screencast.com/t/7HfJ9qWFwb2L
>
> Can I manually import the hostedengine storage?
> ( Hosted engine is visible in web interface and I have running VMs on
> another storage domain ).
>

It should auto-import it.
Adding Roy here.


>
>
> Op ma 8 feb. 2016 om 10:36 schreef Simone Tiraboschi  >:
>
>>
>>
>> On Mon, Feb 8, 2016 at 9:47 AM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>> Update:
>>>
>>
>> The OVF_STORE volume will contain the engine VM description in a format
>> that the engine can easily handle.
>> It will be created by the engine when:
>> - it imported the hosted-engine storage domain
>> - the engineVM is visible
>> - the datacenter is up (you need at least one additional regular storage
>> domain)
>>
>> Please make sure that all these conditions are met.
>>
>> If the OVF_STORE is not there or it's not accessible, ovirt-ha-agent will
>> fallback to the initial vm.conf from deployment time. You see a warning
>> since we are failing back to an initial status and this can temporary
>> override your latest customization to the engine VM configuration.
>>
>>
>>> 1. I was able to remove the warnirngs about the not found ovf / vm.conf
>>> by commenting the conf_volume_UUID, conf_image_UUID entries in the
>>> hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
>>> gone. ( I believe this is not a permanent solution ).
>>>
>>
>> Please revert it.
>>
>>
>>>
>>> 2. This morning I noticed the warning in the event log.
>>> "The Hosted Engine Storage Domain doesn't exist. It should be imported
>>> into the setup."
>>>
>>
>> It's just a side effect of your point 1, please revert it.
>>
>>
>>>
>>> Any help would be appreciated.
>>>
>>> Best Regards,
>>>
>>> Paul Groeneweg
>>>
>>> Op zo 7 feb. 2016 om 22:14 schreef Paul Groeneweg | Pazion <
>>> p...@pazion.nl>:
>>>

 I have been upgrading our ovirt hosts to el7 to be able to run version
 3.6.
 When deploying a new host have problems with the shared OVF storage
 config part.

 1. I get an error during deploy:

 [ INFO  ] Stage: Setup validation
 [ ERROR ] Failed to execute stage 'Setup validation': failed to read
 metadata: [Errno 2] No such file or directory:
 '/rhev/data-center/mnt/[NFS3_SHARE]/88b69eba-ef4f-4dbe-ba53-20dadd424d0e/ha_agent/hosted-engine.metadata'
 [ INFO  ] Stage: Clean up

 It seems the symlink of hosted-engine.metadata is missing.
 So I added manually the /var/run/vdsm/storage folder with subfolder
 88b69eba-ef4f-4dbe-ba53-20dadd424d0e and symlinks.
 Then I reran the hosted-engine deploy and this time all went well.

 2. Now when the host is installed I notice the following in the logs

 
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Preparing
 images
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 INFO:ovirt_hosted_engine_ha.lib.image.Image:Preparing images
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine:Reloading
 vm.conf from the shared storage domain
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 INFO:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Trying
 to get a fresher copy of vm configuration from the OVF_STORE
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 WARNING:ovirt_hosted_engine_ha.lib.ovf.ovf_store.OVFStore:Unable to find
 OVF_STORE
 Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
 ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config ERROR Unable
 to get vm.conf from OVF_STORE, falling back to initial vm.conf
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 ERROR:ovirt_hosted_engine_ha.agent.hosted_engine.HostedEngine.config:Unable
 to get vm.conf from OVF_STORE, falling back to initial vm.conf
 Feb  5 22:33:08 geisha-2 journal: ovirt-ha-agent
 ovirt_hosted_engine_ha.agent.agent.Agent ERROR Error: 'Path to volume
 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
 trying to restart agent
 Feb  5 22:33:08 geisha-2 ovirt-ha-agent:
 ERROR:ovirt_hosted_engine_ha.agent.agent.Agent:Error: 'Path to volume
 6a5b55c3-3f03-47cc-b322-d6d7562371be not found in /rhev/data-center/mnt' -
 trying to restart agent

 So at this point I was a bit stuck.
 I removed complete ovirt36 release and installed ovirt35 for

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Paul Groeneweg | Pazion 
I found vdsClient and ran a few commands.

[root@geisha-1 ~]# vdsClient -s 0 getStorageDomainInfo
88b69eba-ef4f-4dbe-ba53-20dadd424d0e

uuid = 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

version = 3

role = Master

remotePath = hostedstorage.pazion.nl:/opt/hosted-engine

type = NFS

class = Data

pool = ['499b208c-9de9-4a2a-97de-30f410b4e6d4']

name = hostedengine_nfs


[root@geisha-1 ~]# vdsClient -s 0 connectStoragePool
499b208c-9de9-4a2a-97de-30f410b4e6d4

Error using command: list index out of range


...


[root@geisha-1 ~]# vdsClient -s 0 getStoragePoolInfo
499b208c-9de9-4a2a-97de-30f410b4e6d4

Unknown pool id, pool not connected:
('499b208c-9de9-4a2a-97de-30f410b4e6d4',)


[root@geisha-1 ~]# vdsClient -s 0 getConnectedStoragePoolsList

0002-0002-0002-0002-0385


Can I manually create the missing storage pool which is missing with:

vdsClient  -s 0 createStoragePool 499b208c-9de9-4a2a-97de-30f410b4e6d4
hostedengine_nfs 88b69eba-ef4f-4dbe-ba53-20dadd424d0e

If this would be the solution, I still wonder why there is no existing
storagePool for the storageDomain.



Op ma 8 feb. 2016 om 22:54 schreef Paul Groeneweg | Pazion :

> I still keep getting
>
> The Hosted Engine Storage Domain doesn't exist. It should be imported into
> the setup.
>
> I went digging a little further and found this in the engine.log
>
> 2016-02-08 22:30:25,083 INFO
>  [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
> HSMGetStorageDomainInfoVDSCommand, return:
>  id='88b69eba-ef4f-4dbe-ba53-20dadd424d0e'},
> 499b208c-9de9-4a2a-97de-30f410b4e6d4>, log id: 56723427
> 2016-02-08 22:30:25,083 INFO
>  [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
> (org.ovirt.thread.pool-8-thread-41) [] FINISH,
> GetExistingStorageDomainListQuery, log id: 6ca437b1
> 2016-02-08 22:30:25,084 INFO
>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
> (org.ovirt.thread.pool-8-thread-41) [364719ce] Lock Acquired to object
> 'EngineLock:{exclusiveLocks='[]', sharedLocks='null'}'
> 2016-02-08 22:30:25,102 WARN
>  [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
> (org.ovirt.thread.pool-8-thread-41) [364719ce] CanDoAction of action
> 'ImportHostedEngineStorageDomain' failed for user SYSTEM. Reasons:
> VAR__ACTION__ADD,VAR__TYPE__STORAGE__DOMAIN,ACTION_TYPE_FAILED_STORAGE_DOMAIN_NOT_EXIST
>
> So it looks like it can't find the storage domain. When I check vdsm.log I
> found an error:
> Thread-711902::ERROR::2016-02-08
> 21:17:01,240::dispatcher::76::Storage.Dispatcher::(wrapper) {'status':
> {'message': "Unknown pool id, pool not connected:
> ('499b208c-9de9-4a2a-97de-30f410b4e6d4',)", 'code': 309}}
>
> It looks like 499b208c-9de9-4a2a-97de-30f410b4e6d4 is the id it is also
> trying to connect to in the engine.log.
>
> I did found https://bugzilla.redhat.com/show_bug.cgi?id=1269768  and
> https://bugzilla.redhat.com/show_bug.cgi?id=1294457
>
> Can it be I have setup another DC with a master iscsi storage  which is
> now offline?  ( 1 DC with Cluster,  VMs with hosted engine and master
> storage FC all up and running.  )
>
> Not sure where to go from here. Can I somehow restart this storage domain
> / do a new hosted-engine deploy with different settings?
>
>
> Op ma 8 feb. 2016 om 12:05 schreef Paul Groeneweg | Pazion  >:
>
>> Op ma 8 feb. 2016 om 11:28 schreef Simone Tiraboschi > >:
>>
>>> On Mon, Feb 8, 2016 at 10:55 AM, Paul Groeneweg | Pazion >> > wrote:
>>>
 Thanks for your response.

 I have reverted the commented entries and restarted ovirt-ha-agent.
 Now status on the daemon gives the error again:
 http://screencast.com/t/DCZo6QJM1


>>> Here I'm a bit confused: is this a fresh hosted-engine deployment with
>>> 3.6 or an upgrade from an existing 3.5 hosted-engine deployment?
>>>
>>
>> I have  reinstalled host ( complete OS reinstall) and ran hosted-engine
>> --deploy for redeployment on an existing oVirt ( hosted-engine ).
>>
>> As described in previous mail, during the deployment I got errors about a
>> missing hostedengine.metadata symlink to /var/run/vdsd/storage/...
>> I created these by hand and then I ran deployment again and it worked.
>>
>>
>>>
 The hosted engine domain is not shown in the web storage overview...
 When I start import it shows a warning:
 http://screencast.com/t/7HfJ9qWFwb2L

 Can I manually import the hostedengine storage?
 ( Hosted engine is visible in web interface and I have running VMs on
 another storage domain ).

>>>
>>> It should auto-import it.
>>> Adding Roy here.
>>>
>>>
>> I believe this auto-import is failing...
>> Where can I find logs about this auto-import?
>>
>>
>>
>>
>>>

 Op ma 8 feb. 2016 om 10:36 schreef Simone Tiraboschi <
 stira...@redhat.com>:

>
>
> On

Re: [ovirt-users] ovirt 3.6 hostedengine.metadata, missing OVF_STORE

2016-02-08 Thread Paul Groeneweg | Pazion 
I still keep getting

The Hosted Engine Storage Domain doesn't exist. It should be imported into
the setup.

I went digging a little further and found this in the engine.log

2016-02-08 22:30:25,083 INFO
 [org.ovirt.engine.core.vdsbroker.vdsbroker.HSMGetStorageDomainInfoVDSCommand]
(org.ovirt.thread.pool-8-thread-41) [] FINISH,
HSMGetStorageDomainInfoVDSCommand, return:
, log id: 56723427
2016-02-08 22:30:25,083 INFO
 [org.ovirt.engine.core.bll.storage.GetExistingStorageDomainListQuery]
(org.ovirt.thread.pool-8-thread-41) [] FINISH,
GetExistingStorageDomainListQuery, log id: 6ca437b1
2016-02-08 22:30:25,084 INFO
 [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
(org.ovirt.thread.pool-8-thread-41) [364719ce] Lock Acquired to object
'EngineLock:{exclusiveLocks='[]', sharedLocks='null'}'
2016-02-08 22:30:25,102 WARN
 [org.ovirt.engine.core.bll.ImportHostedEngineStorageDomainCommand]
(org.ovirt.thread.pool-8-thread-41) [364719ce] CanDoAction of action
'ImportHostedEngineStorageDomain' failed for user SYSTEM. Reasons:
VAR__ACTION__ADD,VAR__TYPE__STORAGE__DOMAIN,ACTION_TYPE_FAILED_STORAGE_DOMAIN_NOT_EXIST

So it looks like it can't find the storage domain. When I check vdsm.log I
found an error:
Thread-711902::ERROR::2016-02-08
21:17:01,240::dispatcher::76::Storage.Dispatcher::(wrapper) {'status':
{'message': "Unknown pool id, pool not connected:
('499b208c-9de9-4a2a-97de-30f410b4e6d4',)", 'code': 309}}

It looks like 499b208c-9de9-4a2a-97de-30f410b4e6d4 is the id it is also
trying to connect to in the engine.log.

I did found https://bugzilla.redhat.com/show_bug.cgi?id=1269768  and
https://bugzilla.redhat.com/show_bug.cgi?id=1294457

Can it be I have setup another DC with a master iscsi storage  which is now
offline?  ( 1 DC with Cluster,  VMs with hosted engine and master storage
FC all up and running.  )

Not sure where to go from here. Can I somehow restart this storage domain /
do a new hosted-engine deploy with different settings?


Op ma 8 feb. 2016 om 12:05 schreef Paul Groeneweg | Pazion :

> Op ma 8 feb. 2016 om 11:28 schreef Simone Tiraboschi  >:
>
>> On Mon, Feb 8, 2016 at 10:55 AM, Paul Groeneweg | Pazion 
>> wrote:
>>
>>> Thanks for your response.
>>>
>>> I have reverted the commented entries and restarted ovirt-ha-agent.
>>> Now status on the daemon gives the error again:
>>> http://screencast.com/t/DCZo6QJM1
>>>
>>>
>> Here I'm a bit confused: is this a fresh hosted-engine deployment with
>> 3.6 or an upgrade from an existing 3.5 hosted-engine deployment?
>>
>
> I have  reinstalled host ( complete OS reinstall) and ran hosted-engine
> --deploy for redeployment on an existing oVirt ( hosted-engine ).
>
> As described in previous mail, during the deployment I got errors about a
> missing hostedengine.metadata symlink to /var/run/vdsd/storage/...
> I created these by hand and then I ran deployment again and it worked.
>
>
>>
>>> The hosted engine domain is not shown in the web storage overview...
>>> When I start import it shows a warning:
>>> http://screencast.com/t/7HfJ9qWFwb2L
>>>
>>> Can I manually import the hostedengine storage?
>>> ( Hosted engine is visible in web interface and I have running VMs on
>>> another storage domain ).
>>>
>>
>> It should auto-import it.
>> Adding Roy here.
>>
>>
> I believe this auto-import is failing...
> Where can I find logs about this auto-import?
>
>
>
>
>>
>>>
>>> Op ma 8 feb. 2016 om 10:36 schreef Simone Tiraboschi <
>>> stira...@redhat.com>:
>>>


 On Mon, Feb 8, 2016 at 9:47 AM, Paul Groeneweg | Pazion  wrote:

> Update:
>

 The OVF_STORE volume will contain the engine VM description in a format
 that the engine can easily handle.
 It will be created by the engine when:
 - it imported the hosted-engine storage domain
 - the engineVM is visible
 - the datacenter is up (you need at least one additional regular
 storage domain)

 Please make sure that all these conditions are met.

 If the OVF_STORE is not there or it's not accessible, ovirt-ha-agent
 will fallback to the initial vm.conf from deployment time. You see a
 warning since we are failing back to an initial status and this can
 temporary override your latest customization to the engine VM 
 configuration.


> 1. I was able to remove the warnirngs about the not found ovf /
> vm.conf by commenting the conf_volume_UUID, conf_image_UUID entries in the
> hosted-engine.conf. After a restart of the ovirt-ha-agent the errors were
> gone. ( I believe this is not a permanent solution ).
>

 Please revert it.


>
> 2. This morning I noticed the warning in the event log.
> "The Hosted Engine Storage Domain doesn't exist. It should be imported
>