Re: [ovirt-users] Attaching storage to VM via JSON REST
I have no experience using JSON myself (I use XML when using REST), however, note that the storage_domain is a list of ids, inside of a storage_domains object. Maybe try to send a request formatted like that? Juan - any other insights? Gadi Ickowicz - Original Message - From: Andrew Morrison andy.wmorri...@gmail.com To: users@ovirt.org Sent: Wednesday, May 7, 2014 11:23:13 PM Subject: [ovirt-users] Attaching storage to VM via JSON REST Hi all, I'm exploring using ovirt's REST APIs to manage VM creation/deletion/etc. I can create the VM, attach an ISO and change boot order, but I'm unable to attach a storage disk to the VM. I send a POST request to /api/vms/vm uuid/disks with the structure: {storage_domain: {name:am_data, id:6fbee79d-5b71-48c1-b86c-ee9755a46013 }, size:4294967296, type:system, interface:ide, format:raw, sparse:false, bootable:true, wipe_after_delete:false } The response I receive is { reason : Operation Failed, detail : [Cannot add Virtual Machine Disk. Storage Domain doesn't exist.] } For the storage domain, I've tried just the name, just the id, and both with the same results. When I attach storage from the GUI, the disk is created/attached fine, and a GET request for the disks of the VM shows the storage domain as the same as I was attempting to use before. storage_domains : { storage_domain : [ { id : 6fbee79d-5b71-48c1-b86c-ee9755a46013 } ] }, Any idea what I might be doing wrong? Thanks, Andrew ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Storage Domain Not Found After upgrade from 3.3.2 to 3.4
Hi Paul, Great to hear that you were able to get 3.4 working. It makes sense that vdsm complained about the changed metadata - in general it is not a good idea to alter this yourself (and if you do, it is better to delete the checksum entry entirely and then let vdsm recalculate it next time it tries to read the metadata). It is still strange that 3.3.2 had no issue if this information was still in the metadata *before* the upgrade, but the solution would still be to remove the reference to it somehow, although the preferred method would be to tell vdsm cli commands to do it using (hope this never happens again, but listing it here for reference anyway ;) ): vdsClient -s 0 deactivateStorageDomain domain UUID pool UUID and vdsClient -s 0 detachStorageDomain domain UUID pool UUID Thanks, Gadi Ickowicz - Original Message - From: re...@mccleary.me.uk To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Wednesday, April 30, 2014 10:44:29 PM Subject: Re: [ovirt-users] Storage Domain Not Found After upgrade from 3.3.2 to 3.4 Hi Gadi, I thought I would have a look around for this missing storage domain. I searched all the storage areas and in the master domain directory structure I found a metadata file which had a reference to this storage domain ID under POOL_DOMAINS. I edited this file and removed the offending domain ID (all the other domain IDs matched those listed by vdsClient), whilst the host was in maintenance mode. s2data1_s2mgt_boot1/0a897f2e-1b01-4577-9f91-cd136ef4a978/dom_md/metadata: POOL_DOMAINS=7b083758-45f9-4896-913d-11fe02043e6e:Active,e637bb04-a8b7-4c77-809c-d58051494c52:Active,0a897f2e-1b01-4577-9f91-cd136ef4a978:Active,c2c4ade6-049e-4159-a294-a0c151f4983d:Active,9a4a80a1-5377-4a94-ade3-e58183e916ae:Active The storage domains failed to come online and this message was in vdsm.log: Thread-105::ERROR::2014-04-30 20:10:54,292::dispatcher::67::Storage.Dispatcher.Protect::(run) {'status': {'message': Meta Data seal is broken (checksum mismatch): 'cksum = e8b290eebc9a70d822b38a81d25d9f11eae6f282, computed_cksum = 83d6a7876b6a915f69818490610306b0287efe6f', 'code': 752}} I had seen a _SHA_CKSUM parameter in the metadata file, so I changed this to the computed value listed in the logfile and the storage domains activated fine. I then upgraded to 3.4 and when I took it out of maintenance mode, everything came online fine :) So it looks to me like 3.4 or the upgrade to it, has a problem with this erroneous storage domain ID being in the metadata file, that didn't cause a problem for 3.3.2. Either way both my servers are now running 3.4 :) Thanks for all your help with this. Cheers, Paul On 30/04/2014 06:48, Gadi Ickowicz wrote: Hi Paul, Looking at what you sent, it seems a bit strange - another domain was added to the data-center according to the logs you sent in the other mail (the ones after the upgrade). If you could send the engine logs for for the same time period and the engine upgrade log, there may be some information there, but currently the log you sent just starts from the fact that there is a missing storage domain, and it is already listed in the storage pool's metadata. In any case, it is (somewhat) good to hear that reverting to 3.3.2 at least returned the system to a sane state. Thanks, Gadi Ickowicz - Original Message - From: re...@mccleary.me.uk To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Tuesday, April 29, 2014 7:52:15 PM Subject: Re: [ovirt-users] Storage Domain Not Found After upgrade from 3.3.2 to 3.4 Hi Gadi, Thanks for the response. I've only just had chance to restore that node back, which completed fine and the DC and it's storage domains activated ok. So the ovirt-node is now fully functional again on the 3.3.2 version; the ovirt-engine server remains untouched on 3.4 and is working fine. I collected the requested info, but it doesn't list the ID you pulled out of the vdsm.log: [root@ovirt-node ~]# vdsClient -s 0 getStorageDomainsList e637bb04-a8b7-4c77-809c-d58051494c52 7b083758-45f9-4896-913d-11fe02043e6e c2c4ade6-049e-4159-a294-a0c151f4983d 0a897f2e-1b01-4577-9f91-cd136ef4a978 I tried anyway to list against that ID and I get the storage domain doesn't exist error, but maybe that's because it really doesn't exist!? [root@ovirt-node ~]# vdsClient -s 0 getStorageDomainInfo 9a4a80a1-5377-4a94-ade3-e58183e916ae Storage domain does not exist: ('9a4a80a1-5377-4a94-ade3-e58183e916ae',) I grabbed the info against all the Storage Domain IDs from the list without issue: [root@ovirt-node ~]# ## Storage Domain: e637bb04-a8b7-4c77-809c-d58051494c52 ## uuid = e637bb04-a8b7-4c77-809c-d58051494c52 pool = ['c713062f-300f-4256-9ac8-2d3fcfcdb002'] lver = -1 version = 3 role = Regular remotePath = /vdsm_store/s2data1_s2usr_boot1 spm_id = -1 type
Re: [ovirt-users] Could not connect host to Data Center after rebooting, how to resolve?
Hi, Could you please attach the vdsm logs for the host for the time around the logs you pasted from the engine.log? Thanks, Gadi Ickowicz - Original Message - From: Boudewijn Ector ICT administra...@boduewijnector.nl To: users users@ovirt.org Sent: Wednesday, April 30, 2014 5:09:39 PM Subject: [ovirt-users] Could not connect host to Data Center after rebooting, how to resolve? Hi list, I had to do some fsck-related things last week on my ovirt box (centos, single node and NFS on localhost). Afterwards ovirt refused to start VMs according to the webinterface because it can't connect host to Data Center. On my OS it works fine (I just replaced the IP by 'IP'): [root@server ovirt-engine]# df Filesystem1K-blocks Used Available Use% Mounted on /dev/mapper/vg_leiden-lv_root 516061406872592 42112108 15% / tmpfs 3978120 03978120 0% /dev/shm /dev/sdb1495844 99542 370702 22% /boot /dev/mapper/vg_server-lv_home 629902602584272 57206196 5% /home /dev/sda15814366992 4629615452 1184751540 80% /raid IP:/raid/ovirt/data 5814367232 4629615616 1184751616 80% /rhev/data-center/mnt/IP:_raid_ovirt_data IP:/raid/ovirt/iso 5814367232 4629615616 1184751616 80% /rhev/data-center/mnt/IP:_raid_ovirt_iso in my event log in the webinterface I found correlation ID 7a735111 . So grepping my logs for that one: engine.log:2014-04-30 15:59:22,851 INFO [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand] (ajp--127.0.0.1-8702-8) [7a735111] Lock Acquired to object EngineLock [exclusiveLocks= key: 6bee0e2d-961c-453d-a266-e4623f91e162 value: STORAGE engine.log:2014-04-30 15:59:22,888 INFO [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] Running command: ActivateStorageDomainCommand internal: false. Entities affected : ID: 6bee0e2d-961c-453d-a266-e4623f91e162 Type: Storage engine.log:2014-04-30 15:59:22,894 INFO [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] Lock freed to object EngineLock [exclusiveLocks= key: 6bee0e2d-961c-453d-a266-e4623f91e162 value: STORAGE engine.log:2014-04-30 15:59:22,895 INFO [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] ActivateStorage Domain. Before Connect all hosts to pool. Time:4/30/14 3:59 PM engine.log:2014-04-30 15:59:22,945 INFO [org.ovirt.engine.core.vdsbroker.irsbroker.ActivateStorageDomainVDSCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] START, ActivateStorageDomainVDSCommand( storagePoolId = 0002-0002-0002-0002-00ec, ignoreFailoverLimit = false, storageDomainId = 6bee0e2d-961c-453d-a266-e4623f91e162), log id: 5fecb439 engine.log:2014-04-30 15:59:23,011 INFO [org.ovirt.engine.core.vdsbroker.irsbroker.IrsBrokerCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] hostFromVds::selectedVds - server, spmStatus Unknown_Pool, storage pool Default engine.log:2014-04-30 15:59:23,015 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.ConnectStoragePoolVDSCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] START, ConnectStoragePoolVDSCommand(HostName = server, HostId = ff23de79-f17c-439d-939e-d8f3d9672367, storagePoolId = 0002-0002-0002-0002-00ec, vds_spm_id = 1, masterDomainId = 6bee0e2d-961c-453d-a266-e4623f91e162, masterVersion = 1), log id: 15866a95 engine.log:2014-04-30 15:59:23,151 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.ConnectStoragePoolVDSCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] FINISH, ConnectStoragePoolVDSCommand, log id: 15866a95 engine.log:2014-04-30 15:59:23,152 ERROR [org.ovirt.engine.core.vdsbroker.irsbroker.IrsBrokerCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] IrsBroker::Failed::ActivateStorageDomainVDS due to: IRSNonOperationalException: IRSGenericException: IRSErrorException: IRSNonOperationalException: Could not connect host to Data Center(Storage issue) engine.log:2014-04-30 15:59:23,156 INFO [org.ovirt.engine.core.vdsbroker.irsbroker.ActivateStorageDomainVDSCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] FINISH, ActivateStorageDomainVDSCommand, log id: 5fecb439 engine.log:2014-04-30 15:59:23,157 ERROR [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand] (org.ovirt.thread.pool-6-thread-49) [7a735111] Command org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand throw Vdc Bll exception. With error message VdcBLLException: org.ovirt.engine.core.vdsbroker.irsbroker.IRSNonOperationalException: IRSGenericException: IRSErrorException: IRSNonOperationalException: Could not connect host to Data Center(Storage issue) (Failed with error ENGINE and code 5001) engine.log:2014-04-30 15:59:23,162 INFO [org.ovirt.engine.core.bll.storage.ActivateStorageDomainCommand
Re: [ovirt-users] Copying VM from one datacenter to another
Hi, AFAIK the way to copy a vm from one DC to another is: 1. Export vm to export domain 2. deactivate + detach export domain from source DC 3. attach and activate export domain to target dc 4. import the vm Gadi Ickowicz - Original Message - From: Gabi C gab...@gmail.com To: users@ovirt.org Sent: Tuesday, April 29, 2014 10:56:25 AM Subject: [ovirt-users] Copying VM from one datacenter to another Hello! De facto; I have two datacenter: - 1 with one host, configured with local storage, just added in the engine; - 1 with 3 host, 2 shared storage domain, 1 ISO and 1 Export For some backup reasons ( cold backup) I want to copy machines from second Datacenter to the Local-storage one. Is there some documented way of doing this? Any hints, thoughts, similar experience? Thanks! ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Copying VM from one datacenter to another
It shouldn't - that is the whole point of an export domain - do be able to move it between DCs Gadi Ickowicz - Original Message - From: Gabi C gab...@gmail.com To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Tuesday, April 29, 2014 11:23:55 AM Subject: Re: [ovirt-users] Copying VM from one datacenter to another Hello! I was thinking about this one to, but detaching export from source DC wouldn't bring DC down? Thanks! On Tue, Apr 29, 2014 at 11:17 AM, Gadi Ickowicz gicko...@redhat.com wrote: Hi, AFAIK the way to copy a vm from one DC to another is: 1. Export vm to export domain 2. deactivate + detach export domain from source DC 3. attach and activate export domain to target dc 4. import the vm Gadi Ickowicz - Original Message - From: Gabi C gab...@gmail.com To: users@ovirt.org Sent: Tuesday, April 29, 2014 10:56:25 AM Subject: [ovirt-users] Copying VM from one datacenter to another Hello! De facto; I have two datacenter: - 1 with one host, configured with local storage, just added in the engine; - 1 with 3 host, 2 shared storage domain, 1 ISO and 1 Export For some backup reasons ( cold backup) I want to copy machines from second Datacenter to the Local-storage one. Is there some documented way of doing this? Any hints, thoughts, similar experience? Thanks! ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] Storage Domain Not Found After upgrade from 3.3.2 to 3.4
Hi Paul, Looking at what you sent, it seems a bit strange - another domain was added to the data-center according to the logs you sent in the other mail (the ones after the upgrade). If you could send the engine logs for for the same time period and the engine upgrade log, there may be some information there, but currently the log you sent just starts from the fact that there is a missing storage domain, and it is already listed in the storage pool's metadata. In any case, it is (somewhat) good to hear that reverting to 3.3.2 at least returned the system to a sane state. Thanks, Gadi Ickowicz - Original Message - From: re...@mccleary.me.uk To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Tuesday, April 29, 2014 7:52:15 PM Subject: Re: [ovirt-users] Storage Domain Not Found After upgrade from 3.3.2 to 3.4 Hi Gadi, Thanks for the response. I've only just had chance to restore that node back, which completed fine and the DC and it's storage domains activated ok. So the ovirt-node is now fully functional again on the 3.3.2 version; the ovirt-engine server remains untouched on 3.4 and is working fine. I collected the requested info, but it doesn't list the ID you pulled out of the vdsm.log: [root@ovirt-node ~]# vdsClient -s 0 getStorageDomainsList e637bb04-a8b7-4c77-809c-d58051494c52 7b083758-45f9-4896-913d-11fe02043e6e c2c4ade6-049e-4159-a294-a0c151f4983d 0a897f2e-1b01-4577-9f91-cd136ef4a978 I tried anyway to list against that ID and I get the storage domain doesn't exist error, but maybe that's because it really doesn't exist!? [root@ovirt-node ~]# vdsClient -s 0 getStorageDomainInfo 9a4a80a1-5377-4a94-ade3-e58183e916ae Storage domain does not exist: ('9a4a80a1-5377-4a94-ade3-e58183e916ae',) I grabbed the info against all the Storage Domain IDs from the list without issue: [root@ovirt-node ~]# ## Storage Domain: e637bb04-a8b7-4c77-809c-d58051494c52 ## uuid = e637bb04-a8b7-4c77-809c-d58051494c52 pool = ['c713062f-300f-4256-9ac8-2d3fcfcdb002'] lver = -1 version = 3 role = Regular remotePath = /vdsm_store/s2data1_s2usr_boot1 spm_id = -1 type = LOCALFS class = Data master_ver = 0 name = s2data1_s2usr_boot1 ## Storage Domain: 7b083758-45f9-4896-913d-11fe02043e6e ## uuid = 7b083758-45f9-4896-913d-11fe02043e6e pool = ['f027ec99-913f-4f00-ac95-ad484c9c6a4b', 'c713062f-300f-4256-9ac8-2d3fcfcdb002'] lver = -1 version = 0 role = Regular remotePath = ovirt-engine:/iso spm_id = -1 type = NFS class = Iso master_ver = 0 name = ISO1_ZFS ## Storage Domain: c2c4ade6-049e-4159-a294-a0c151f4983d ## uuid = c2c4ade6-049e-4159-a294-a0c151f4983d pool = ['c713062f-300f-4256-9ac8-2d3fcfcdb002'] lver = -1 version = 3 role = Regular remotePath = /vdsm_store/s2data1_s2mgt_app1 spm_id = -1 type = LOCALFS class = Data master_ver = 0 name = s2data1_s2mgt_app1 ## Storage Domain: 0a897f2e-1b01-4577-9f91-cd136ef4a978 ## uuid = 0a897f2e-1b01-4577-9f91-cd136ef4a978 pool = ['c713062f-300f-4256-9ac8-2d3fcfcdb002'] lver = 1 version = 3 role = Master remotePath = /vdsm_store/s2data1_s2mgt_boot1 spm_id = 1 type = LOCALFS class = Data master_ver = 1 name = s2data1_s2mgt_boot1 I checked the logfile for today and since the restore there are no errors about this storage domain not existing: [root@ovirt-node vdsm]# grep 04-29 vdsm.log|grep StorageDomainDoesNotExist [root@ovirt-node vdsm]# I also checked the IDs listed on the ovirt-engine server just on the off-chance, but the ID throwing the error doesn't exist on that either: [root@ovirt-engine ~]# vdsClient -s 0 getStorageDomainsList feb04d94-4ea8-471c-b759-3ed95943e9a3 b3c02266-2426-4285-b4dc-0acba75af530 7b083758-45f9-4896-913d-11fe02043e6e 79178b6b-8d98-45e4-93f2-3ce1d7a270a5 None of the storage domains exist on the root VG partitions, they are completely separate disks in separate VGs. I did delete some Storage Domains on both the ovirt-engine and the ovirt-node a number of weeks back, but if it was due to this I would have expected the issues on both servers, not just one. Have you found anything else that looks interesting in the log? Thanks, Paul On 28/04/2014 07:07, Gadi Ickowicz wrote: Hi Paul, I am still looking into this log, but from a quick first assessment, it looks like (for some reason I don't know yet...) there is a storage domain that is missing. This is visible in the following error traceback in the vdsm log: Thread-29::ERROR::2014-04-27 12:43:05,825::domainMonitor::239::Storage.DomainMonitorThread::(_monitorDomain) Error while collecting domain 9a4a80a1-5377-4a94-ade3-e58183e916ae m
Re: [ovirt-users] Hosted Engine started VM Multiple Times
Hi Andrew, Could you please send then engine and vdsm logs so we can take a look and try to understand how this happened? Gadi Ickowicz - Original Message - From: Andrew Lau and...@andrewklau.com To: users users@ovirt.org Sent: Tuesday, April 29, 2014 7:29:41 AM Subject: [ovirt-users] Hosted Engine started VM Multiple Times Hi, I added a new node to my hosted cluster today, only one other node was active in the cluster at that time. It was running the hosted-engine, however during the install it's agent died and when the second node came up. It started the hosted-engine, causing the hosted-engine to be run on both nodes. Has anyone had this happen? This seems very dangerous, as both VMs were running simultaneously.. data corruption alert! Andrew. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [ovirt-users] List disk images residing on specific lun
Hi Amedeo, Was the LUN that was lost a separate storage domain? if it was, you can list all the disks in the system and then filter the list for disks on that storage domain. for example, in python sdk: disks = api.disks.list() [disk for disk in disks if disk.get_storage_domains().get_storage_domain().get_id() == fc_storage_domain_id] Thanks, Gadi Ickowicz - Original Message - From: Amedeo Salvati ame...@oscert.net To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Sent: Sunday, April 13, 2014 3:05:37 PM Subject: Re: [ovirt-users] List disk images residing on specific lun Il 13/04/2014 12:37, Meital Bourvine ha scritto: Hi Amedeo, I don't think that it's possible. But if your storage domain contains only 1 FC lun, then listing the disks on this storage domain will provide you the info. - Original Message - From: Amedeo Salvati ame...@oscert.net To: users@ovirt.org Sent: Sunday, April 13, 2014 1:23:13 PM Subject: [ovirt-users] List disk images residing on specific lun Hello, does anyone know if it's possible, using ovirt-shell or api, to list every disk image residing on specific FC LUN? best regards Amedeo Salvati ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users Hi Meital and all, unfortunately we have several luns, but we've lost one of them (and it's not possible to recover it from the box), so we're looking to know what images are residing on that lun due to remove those images from db/engine gui... using lvm command on hypervisor node it's possible to match lv name still alive, with some id present on engine db or by ovirt-shell cmd like list disk --show-all ? thanks in advance a ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Clome thin provision to thin provision??
Hi, This should be possible AFAIK - are you selecting thin provision for each disk? if you are, could you please attach engine logs and vdsm logs from the spm host for the time of creation of the vm from the snapshot? also - your vdsm version is quite a bit older than the engine version Thanks, Gadi Ickowicz - Original Message - From: Federico Alberto Sayd fs...@uncu.edu.ar To: users@oVirt.org users@ovirt.org Sent: Monday, March 31, 2014 6:45:13 PM Subject: [Users] Clome thin provision to thin provision?? Hello: Is it possible to clone a vm with thin provisioned disks to another vm with thin provisioned disks too? Currently when I clone a VM from a snapshot, although I choose thin provision in resource allocation-Storage Allocation from the Clone VM from Snapshot dialog, the VM is created with preallocated disk(s) Is this the default or there is a problem with my engine/nodes installation? Engine: Centos 6.5, ovirt-engine 3.4.0-1.el6 Nodes: Centos 6.5, vdsm-4.13.3-2.el6 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] VM Status Unknown
If you are sure that the host (node) running the vm has reset, and the vm is no longer running on that host you can select the host and manually Confirm host has been rebooted (from the webadmin, not sure exactly how to do it from ovirt-shell). Gadi Ickowicz - Original Message - From: Simon Barrett simon.barr...@tradingscreen.com To: Omer Frenkel ofren...@redhat.com, Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Sent: Wednesday, March 26, 2014 10:44:15 AM Subject: Re: [Users] VM Status Unknown Was this ever resolved? I have a similar situation where a node was reset while a VM was running on it. Now the VM is in an unknown state and I cannot stop/start etc. I also cannot put the node into maintenance mode as it thinks it still has running VM’s [oVirt shell (connected)]# list vms --query id=807593a5-8667-4265-a0ab-b3e37feb1fc1 --show-all | grep status status-state : unknown [oVirt shell (connected)]# action vm 807593a5-8667-4265-a0ab-b3e37feb1fc1 stop ERROR status: 409 reason: Conflict detail: Cannot stop VM. VM is not running. === [oVirt shell (connected)]# action host d3a555b1-b5dd-48d1-802c-03096bdf8b00 deactivate ERROR status: 409 reason: Conflict detail: Cannot switch Host to Maintenance mode. Host still has running VMs on it and is in Non Responsive state. === Thanks, From: users-boun...@ovirt.org [mailto:users-boun...@ovirt.org] On Behalf Of Omer Frenkel Sent: 12 January 2014 08:47 To: Meital Bourvine Cc: users@ovirt.org Subject: Re: [Users] VM Status Unknown From: Meital Bourvine mbour...@redhat.com To: Ryan Womer ryan.wo...@cytechservices.com Cc: users@ovirt.org Sent: Friday, January 10, 2014 11:54:51 AM Subject: Re: [Users] VM Status Unknown Can you please provide vdsm.log (of both hosts) and engine.log? also, what version are you using? we had some bugs around this solved for 3.3 From: Ryan Womer ryan.wo...@cytechservices.com To: users@ovirt.org Sent: Thursday, January 9, 2014 10:43:37 PM Subject: [Users] VM Status Unknown During a migration, the destination host lost connectivity to the san and crashed. Once the server came back up, 3 VMs that didn’t finish migrating have been stuck in status “Unknown.” Vdsclient doesn’t list any of the vms on either host. Qemu doesn’t have them listed as mounted on either host. Action vm start and stop result in “Status: 409”. The disks for all 3 VMs are listed as green in the WebAdmin. I’ve tried “action vm name start” “action vm name stop” “update vm name --status-state down” no joy. They remain in “unknown.” ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Ovirt 3.4rc + Qemu Live snapshot : Problem to delete a Snapshot
Hi Eric, Live merge is not yet supported (IIRC by qemu). From the Live_Snapshots wiki page: Live merge is not present in the upstream component yet (libvirt/qemu) Gadi Ickowicz - Original Message - From: Eric Bollengier e...@baculasystems.com To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Sent: Sunday, March 16, 2014 11:41:54 AM Subject: Re: [Users] Ovirt 3.4rc + Qemu Live snapshot : Problem to delete a Snapshot Hello Meital, Thanks for your answer, On 03/16/2014 10:04 AM, Meital Bourvine wrote: I upgraded to Ovirt 3.4rc allinone on centos6, and I compiled/installed the qemu-kvm-rhev to be able to take live snapshots. Taking the snapshot is OK, but I would like to delete them, and the button and the menu are disabled (grey) until I shutdown the VM. Once powered off, actions are available, and I'm able to delete snapshots. This is the expected behaviour. When looking the Snapshot Actions Matrix: http://www.ovirt.org/Live_Snapshots Is the Revert operation needed to delete a snapshot ? (Deleting a snapshot file implies to write changes from the snapshot file to the parent file, looks to be Commit for me). As far as I understand, revert means to go back to the point that you took the snapshot, while delete means to stay at the current point of time but deleting the option to go back. OK, so given the Snapshot Actions Matrix, it should be possible to delete the snapshot via a Live Merge operation. So, why the delete action is not available while the VM is running? How can I enable this feature? Or maybe this feature is in the roadmap? Thanks, Regards, Eric ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Reimporting storage domains after reinstalling ovirt
Hi, Regarding what to do in order to be able to add the domain, due to the bug you pointed out (https://bugzilla.redhat.com/show_bug.cgi?id=1059604), the workaround is to move the host to a datacenter. Check the host's cluster under the web interface - if it does not list a datacenter, you can move the host to maintenance, select the cluster, click 'edit' and then select a datacenter. The second error Error while executing action: Cannot add Storage Connection. Storage connection already exists. is also a known bug (https://bugzilla.redhat.com/show_bug.cgi?id=1014966). You need to find the storage connection and remove it manually from oVirt. This can be done with REST or SDK. For REST: 1. find the ID of the connection. should be at: fqdn/ip of engine/api/storageconnections . (look for the connection that lists 192.168.1.44:/raid/ovirt-old/data and note it's ID) 2. Send a DELETE request to fqdn/ip of engine/api/storageconnections/ID of storage connection Thanks, Gadi Ickowicz - Original Message - From: Boudewijn Ector boudew...@boudewijnector.nl To: users@ovirt.org Sent: Monday, March 10, 2014 1:00:35 AM Subject: Re: [Users] Reimporting storage domains after reinstalling ovirt (there's another storage doman which has to be converted too but that's not that relevant imo). Now I log in onto the webinterface, select storage domain - import domain : Domain function set to export/NFS export path: 192.168.1.44:/raid/ovirt-old/data/ (192.168.1.44 is the server's IP) Now this error pops up: Error while executing action: Cannot add Storage. Storage format V3 is not supported on the selected host version. I just had a look for this one and found this bug (in which you replied... such a small world :) ) : https://bugzilla.redhat.com/show_bug.cgi?id=1059604 But there's no obvious solution over there. Despite this the domains are available in the storage domain list but they're unattached so no use to me. Am I still missing something? Cheers, Boudewijn ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users Interesting: i just tried readding it again (maybe rereading the storage domain takes a lot of time) but despite it not showing up in the storage domain list, ovirt's webadmin tells met the repo has already been added: Error while executing action: Cannot add Storage Connection. Storage connection already exists. Strange. Cheers, Boudewijn ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Snapshots
Hi, Were you taking a live snapshot? That process is actually composed of 2 steps: 1) Taking the snapshot (creating the a new volume that is part of the image) 2) Configuring the vm to use the new volume A failure in step 2 would result in the new volume being created, but the VM still writing to the old volume, and that warning could be what you saw. could you please attach the engine logs, and if possible, the vdsm logs for the SPM at the time you took the snapshot. Thanks, Gadi Ickowicz - Original Message - From: Maurice James midnightst...@msn.com To: users@ovirt.org Sent: Tuesday, March 4, 2014 10:45:39 PM Subject: [Users] Snapshots I attempted to create a snapshot and an alert came up saying that it failed, but when I look at the snapshots tab for that specific VM, it says that the status is OK. Which should I believe? Ver 3.3.3-2.el6 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Migration Failed
Hi, Unfortunately it seems the vdsm logs cycled - these vdsm logs do not match the times for the engine log snippet you pasted - they start at around 10:00 AM and the engine points to 8:33... Gadi Ickowicz - Original Message - From: Koen Vanoppen vanoppen.k...@gmail.com To: Meital Bourvine mbour...@redhat.com, users@ovirt.org Sent: Thursday, February 27, 2014 11:04:11 AM Subject: Re: [Users] Migration Failed In attachment... Thanx! 2014-02-27 9:21 GMT+01:00 Meital Bourvine mbour...@redhat.com : Hi Koen, Can you please attach the relevant vdsm logs? From: Koen Vanoppen vanoppen.k...@gmail.com To: users@ovirt.org Sent: Thursday, February 27, 2014 9:38:46 AM Subject: [Users] Migration Failed Dear all, I added a new host to our ovirt. Everything went good, exept in the beginnen there was a problem with the firmware of the FibreCard but that is solved (maybe relevant to the issue coming up ;-) ), host is green en up now. But when I tried to migrate a machine for testing purpose to see if everythin was ok, I get the following error in the engine.log and the migration fails: 2014-02-27 08:33:08,082 INFO [org.ovirt.engine.core.bll.MigrateVmCommand] (pool-6-thread-49) [f1a68d8] Running command: MigrateVmCommand internal: false. Entities affected : ID: 3444fc9d-0395-4cbb-9a11-28a42802560c Type: VM 2014-02-27 08:33:08,362 INFO [org.ovirt.engine.core.vdsbroker.MigrateVDSCommand] (pool-6-thread-49) [f1a68d8] START, MigrateVDSCommand(HostName = soyuz, HostId = 6dfa2f9c-85c6-4fb3-b65f-c84620115a1a, vmId=3444fc9d-0395-4cbb-9a11-28a42802560c, srcHost= soyuz.brusselsairport.aero , dstVdsId=6707fa40-753a-4c95-9304-e47198477e4d, dstHost= buran.brusselsairport.aero:54321 , migrationMethod=ONLINE, tunnelMigration=false), log id: 50cd7284 2014-02-27 08:33:08,371 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateBrokerVDSCommand] (pool-6-thread-49) [f1a68d8] VdsBroker::migrate::Entered (vm_guid=3444fc9d-0395-4cbb-9a11-28a42802560c, srcHost= soyuz.brusselsairport.aero , dstHost= buran.brusselsairport.aero:54321 , method=online 2014-02-27 08:33:08,405 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateBrokerVDSCommand] (pool-6-thread-49) [f1a68d8] START, MigrateBrokerVDSCommand(HostName = soyuz, HostId = 6dfa2f9c-85c6-4fb3-b65f-c84620115a1a, vmId=3444fc9d-0395-4cbb-9a11-28a42802560c, srcHost= soyuz.brusselsairport.aero , dstVdsId=6707fa40-753a-4c95-9304-e47198477e4d, dstHost= buran.brusselsairport.aero:54321 , migrationMethod=ONLINE, tunnelMigration=false), log id: 20806b79 2014-02-27 08:33:08,441 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateBrokerVDSCommand] (pool-6-thread-49) [f1a68d8] FINISH, MigrateBrokerVDSCommand, log id: 20806b79 2014-02-27 08:33:08,451 INFO [org.ovirt.engine.core.vdsbroker.MigrateVDSCommand] (pool-6-thread-49) [f1a68d8] FINISH, MigrateVDSCommand, return: MigratingFrom, log id: 50cd7284 2014-02-27 08:33:08,491 INFO [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (pool-6-thread-49) [f1a68d8] Correlation ID: f1a68d8, Job ID: c3642418-3f05-41eb-8b1d-07fe04867742, Call Stack: null, Custom Event ID: -1, Message: Migration started (VM: ADW-DevSplunk, Source: soyuz, Destination: buran, User: admin@internal). 2014-02-27 08:33:20,036 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-82) VM ADW-DevSplunk 3444fc9d-0395-4cbb-9a11-28a42802560c moved from MigratingFrom -- Up 2014-02-27 08:33:20,042 INFO [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-82) Adding VM 3444fc9d-0395-4cbb-9a11-28a42802560c to re-run list 2014-02-27 08:33:20,051 ERROR [org.ovirt.engine.core.vdsbroker.VdsUpdateRunTimeInfo] (DefaultQuartzScheduler_Worker-82) Rerun vm 3444fc9d-0395-4cbb-9a11-28a42802560c. Called from vds soyuz 2014-02-27 08:33:20,107 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-6-thread-50) START, MigrateStatusVDSCommand(HostName = soyuz, HostId = 6dfa2f9c-85c6-4fb3-b65f-c84620115a1a, vmId=3444fc9d-0395-4cbb-9a11-28a42802560c), log id: 75ac0a46 2014-02-27 08:33:20,124 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-6-thread-50) Failed in MigrateStatusVDS method 2014-02-27 08:33:20,130 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-6-thread-50) Error code noConPeer and error message VDSGenericException: VDSErrorException: Failed to MigrateStatusVDS, error = Could not connect to peer VDS 2014-02-27 08:33:20,136 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-6-thread-50) Command org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand return value StatusOnlyReturnForXmlRpc [mStatus=StatusForXmlRpc [mCode=10, mMessage=Could not connect to peer VDS]] 2014-02-27 08:33:20,139 INFO [org.ovirt.engine.core.vdsbroker.vdsbroker.MigrateStatusVDSCommand] (pool-6-thread-50
Re: [Users] How do you move an host with local-storage into regular Data Center?
Actually to remove the the last domain in a DC the following steps will do it in and clean up everything without resorting to manual steps: 1. Deactivate and then detach all domains in the DC except last one 2. Deactivate last domain 3. Remove the datacenter (this takes a while since actually the host has to contend for SPM again, like reactivating the domain, in order to remove the DC) 4. DC should be removed and domain should remain unattached - now domain can be removed 5. Move host to maintenance and switch it's cluster to proper dc (should now be possible since it's DC has been removed and is now blank) or move the host directly to proper cluster One thing to note when using force remove for DC is that it does *not* clean up the actual storage - only references to it from the DB (and that is why it disappears from the UI). You have to manually go to the host and clean the storage itself to free up that space Gadi Ickowicz - Original Message - From: Giorgio Bersano giorgio.bers...@gmail.com To: d...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Thursday, February 27, 2014 3:04:25 PM Subject: Re: [Users] How do you move an host with local-storage into regular Data Center? 2014-02-26 17:35 GMT+01:00 Dafna Ron d...@redhat.com: you did not remove the storage before moving the host. Well, I tried but then there was always something impossible to do, i.e. the Local Storage Domain was the only SD so, beeng it the Master SD, it wasn't possible to remove it... something like the chicken or the egg dilemma. If you select the force remove DC option it should clean all object under that DC (just make sure you are selecting the one you want to remove ;)) That worked. I was just trying to avoid that forced option, but you confirmed me it was the only one. Thank you, Giorgio. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Disk Migration
IIUC, since disk migrations are, as Dafna said, essentially a storage operation, the 'vm migration network' interface is not used for it. However, if the storage is iSCSI, on 3.4 you can set up 'iSCSI Bond' interfaces and define which interface is used to access which target. This would essentially allow you to create a bond that is used to access only the storage, and disk migrations should use that. Gadi Ickowicz - Original Message - From: Dafna Ron d...@redhat.com To: Maurice James midnightst...@msn.com Cc: Ofer Blaut obl...@redhat.com, users@ovirt.org Sent: Wednesday, February 26, 2014 7:34:11 PM Subject: Re: [Users] Disk Migration I don't think that you can configure interface for disk migration. Disk migration is actually copy of information from the original disk to a new disk created on a new domain + delete of the original disk once that is done. it's not actually a migration and so I am not sure you can actually configure an interface for that. adding ofer - perhpas he has a solution or it's possible and I am not aware of it. Dafna On 02/26/2014 05:24 PM, Maurice James wrote: I have a specific interface set up for migrations. Why do disk migrations not use the interface that I have set for migrations? Is that by design? Shouldnt it use the interfaces that I have set aside for migrations? VM migrations work as they should but not disk migrations ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- Dafna Ron ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Old storage domain ID left behind in master storage domain's metadata
Hi, How exactly did you delete the storage domain and dc? From the steps you are describing, since the host was not part of the DC when you removed it you had to have used 'force remove' on the DC, which does indeed remove the DC and the domain from the UI, but that is *all* it does. It only removes references to those objects from engine's DB, it does not remove the actual storage domain (VG). Thanks, Gadi Ickowicz - Original Message - From: Boyan Tabakov bl...@alslayer.net To: users@ovirt.org Sent: Tuesday, February 25, 2014 12:59:09 PM Subject: [Users] Old storage domain ID left behind in master storage domain's metadata Hello, I started with an all-in-one installation of ovirt 3.3.2 on a FC19 host. After the initial setup was done, I created a new datacenter, cluster and storage domain in the engine and moved the single host in the engine to the new cluster. After that I deleted the automatically created storage domain, local datacenter, etc. It seems, however, that the new master storage domain still contains some references to the now removed old storage domains in its metadata. Here's the metadata, extracted from the tags of the VG. I got them using vgs: MDT_DESCRIPTION=StorageDomainX MDT_LOCKPOLICY= MDT_PV0=pv:36090a098103b1821532d057a5a0120d444uuid:xreVSW-NmjD-MUqk-jucQ-Hrbs-KRe8-P93tn344pestart:044pecount:79844mapoffset:0 MDT_TYPE=ISCSI MDT_LOGBLKSIZE=512 MDT_VGUUID=npukh2-ulUm-MEbj-T0TZ-0kDX-pUkp-2hJEKo MDT_LEASERETRIES=3 MDT_IOOPTIMEOUTSEC=10 MDT_LOCKRENEWALINTERVALSEC=5 MDT_SDUUID=3307f6fa-dd58-43db-ab23-b1fb299006c7 MDT_PHYBLKSIZE=512 MDT_CLASS=Data MDT_VERSION=3 RHAT_storage_domain MDT_POOL_UUID=61f15cc0-8bba-482d-8a81-cd636a581b58 MDT_ROLE=Master MDT_MASTER_VERSION=44 MDT_POOL_DESCRIPTION=DataCenterX MDT_POOL_DOMAINS=ec4ba43a-e334-4ce2-90d1-86e5a6da071d:Active44a7af75fe-b96a-4ebe-8903-5743a8176311:Active44ff33856e-e00f-4be8-9456-8e47c592f82f:Active443307f6fa-dd58-43db-ab23-b1fb299006c7:Active MDT_PV1=pv:36090a09810bb99fefb2da57b9433202744uuid:UMUs49-GvEF-IOgZ-p6jD-yP1r-dpnd-syawJz44pestart:044pecount:79844mapoffset:798 MDT_POOL_SPM_ID=1 MDT_POOL_SPM_LVER=30 MDT__SHA_CKSUM=ad43ad855d386b103a672a0801a932a9ee115ed7 Notably, see MDT_POOL_DOMAINS. Domains ec4ba43a-e334-4ce2-90d1-86e5a6da071d and a7af75fe-b96a-4ebe-8903-5743a8176311 no longer exist in the engine (were removed from admin UI). Their presence in the metadata causes vdsmd to repeatedly try to access those domains, resulting in constant error messages in the log that the domains don't exist. Any advice on how to properly clean up the old domains from the metadata? Thanks in advance! Boyan Tabakov ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Old storage domain ID left behind in master storage domain's metadata
Regarding the way to remove the old DC and move from local to iSCSI storage, you can: 1) deactivate the last domain in the DC (After detaching all other domains in the DC) 2) remove the DC (while there is still a host in the DC) 3) remove the domain regularly (which should clean the storage itself 4) move the host to the new DC Regarding the fact that the other domain was removed cleanly and is still visible in the metadata that is very strange. Do you happen to have logs from the creation/deletion of that domain? Gadi Ickowicz - Original Message - From: Boyan Tabakov bl...@alslayer.net To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Tuesday, February 25, 2014 2:48:30 PM Subject: Re: [Users] Old storage domain ID left behind in master storage domain's metadata On 25.2.2014, 14:42, Boyan Tabakov wrote: Hi, On 25.2.2014, 14:00, Gadi Ickowicz wrote: Hi, How exactly did you delete the storage domain and dc? From the steps you are describing, since the host was not part of the DC when you removed it you had to have used 'force remove' on the DC, which does indeed remove the DC and the domain from the UI, but that is *all* it does. It only removes references to those objects from engine's DB, it does not remove the actual storage domain (VG). If I recall correctly (that was couple of months ago), I might have really forced the removal. The issue was, that I wanted to get rid of the default created local storage domain and datacenter, so that I can create a proper iSCSI storage domain and later add other nodes (convert the all-in-one to multinode setup). The idea was to convert the all-in-one installation to one having multiple nodes and one of those running the engine. Since in the beginning there was only one host, the all-in-one one, it had to be moved to a new DC. That left the default DC without any host. Was there a better way to achieve this? Is it possible to manually update the metadata or tell vdsmd to do so somehow? Thanks, Boyan Additionally, one of the left out domains was a test iSCSI SD created in the new SD, which was removed cleanly, afaik. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Old storage domain ID left behind in master storage domain's metadata
I am not really sure what the recommended way is (adding Nir Soffer, hopefully he can help or point us to the right person). Maybe you can try manually deactivating and detaching the domains from the pool using vdsClient on the SPM host (for each of the domains): deactivateStorageDomain domainID poolID detachStorageDomain domainID poolID if you have vdsm configured for secure connection (default) then you can ssh to the host and run the following command: vdsClient -s 0 [commands from above]. Otherwise you don't need the -s. I guess you already have the domainID (VG name) and the poolID (from the previous mails you sent). Not sure if this is the correct way to solve this problem - Nir? Gadi Ickowicz - Original Message - From: Boyan Tabakov bl...@alslayer.net To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Sent: Tuesday, February 25, 2014 3:50:42 PM Subject: Re: [Users] Old storage domain ID left behind in master storage domain's metadata On 25.2.2014, 15:26, Gadi Ickowicz wrote: Regarding the way to remove the old DC and move from local to iSCSI storage, you can: 1) deactivate the last domain in the DC (After detaching all other domains in the DC) 2) remove the DC (while there is still a host in the DC) 3) remove the domain regularly (which should clean the storage itself 4) move the host to the new DC Thanks! Unfortunately I have some production loads already and starting over would be tough. Any pointers on how the metadata can be fixed? Regarding the fact that the other domain was removed cleanly and is still visible in the metadata that is very strange. Do you happen to have logs from the creation/deletion of that domain? Looks like those logs are rotated away already. ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] actual disk size of vm via api
Hi, There are 2 different ways to create a vm from a template: * thin * clone If you create a vm that is a thin-copy of a template, it works like a snapshot, creating a new volume on top of the template's disk. If you create a vm that is a full clone then you clone the whole disk image. did you created a thin-copy of the template? Gadi Ickowicz - Original Message - From: Sven Kieske s.kie...@mittwald.de To: Users@ovirt.org List Users@ovirt.org Sent: Wednesday, February 12, 2014 5:57:28 PM Subject: [Users] actual disk size of vm via api Hi, I got a question regarding the reported actual disk size of a vm. you can obtain this value fine via api: https://your_engine/api/vms/{id}/disks/ however, if the vm is cloned from a template this disk size does not seem to show the actual size of the whole disk but instead the difference from the actual template size. is this a bug or a feature? I would like to obtain the real size. e.g. I have a template with 50GB (thin provisioned) the actual size of this template is 1,5 GB I clone a vm from this template, the actual size which gets reported via api is about 8 MB (stuff that changes from the template). So I can not obtain the complete used size of the vm. This was tested under ovirt 3.3.2 -- Mit freundlichen Grüßen / Regards Sven Kieske Systemadministrator Mittwald CM Service GmbH Co. KG Königsberger Straße 6 32339 Espelkamp T: +49-5772-293-100 F: +49-5772-293-333 https://www.mittwald.de Geschäftsführer: Robert Meyer St.Nr.: 331/5721/1033, USt-IdNr.: DE814773217, HRA 6640, AG Bad Oeynhausen Komplementärin: Robert Meyer Verwaltungs GmbH, HRB 13260, AG Bad Oeynhausen ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Disk error
Hi, It means that when the host tried to access the disk for that VM, the host could not find the disk on the storage domain. How did this happen? - Can you please attach engine logs and vdsm logs for the host that tried to run the VM? - What type of storage domain is it (NFS/Gluster/iSCSI...)? Thanks, Gadi Ickowicz - Original Message - From: Maurice James midnightst...@msn.com To: users@ovirt.org Sent: Wednesday, February 12, 2014 4:06:50 AM Subject: [Users] Disk error ImageDoesNotExistInSD: Image does not exist in domain: 'image=d09418b1-2854-40e3-b4be-b4b5062f51d9, domain=21619c8e-99ea-4813-be02-d708971e5393' Does anyone know how to fix this? I have a windows server using that disk and its having issues booting up because of it. Look like I will have to fix it by hand ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] node can not access disks
Hi, It looks like the node cannot find the vms' images (lvs) although it can see the domain, which is very strange. Is this an iSCSI or FCP domain? More logs would help understand exactly what is going on there. Also, could you please paste the output of: # lvs ce77262f-8346-42e5-823a-bd321f0814e7 on the problematic node and also from another node in the same cluster (if possible)? After getting the output of the lvs command, you can try to set the host to maintenance and then activate it again from oVirt Thanks, Gadi Ickowicz - Original Message - From: Jakub Bittner j.bitt...@nbu.cz To: users@ovirt.org Sent: Monday, February 10, 2014 6:55:40 PM Subject: [Users] node can not access disks Hi, after node restart oVirt can not start any virtual machine due to unknown storage error. Logs are here http://fpaste.org/75881/. It looks like it can not read LVM. Is there any way how to fix that? Thanks ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Constant lopping log of task activity
Hi Matt, Are the permissions on the NFS share set vdsm:kvm (36:36)? They have to be in order to be used by ovirt. Also, can you please attach the full vdsm log for the problematic host? I am probably missing something from the logs you pasted, but I cannot see any errors, and it is easier to work with the full logs. Thanks, Gadi Ickowicz - Original Message - From: Matt Warren mwar...@hnw.com To: users@ovirt.org Sent: Monday, February 3, 2014 4:41:51 AM Subject: [Users] Constant lopping log of task activity I'm seeing a constant loop of task issues in the vdsm.log Host is a CentOS 6.5 server otherwise setup as a db lab server. Engine is a CentOS 6.5 vm running on vmware esx host. The host has local storage only. Not sure if related, but I've been trying to attach a Win2k8-hosted NFS share for ISO's and failing. I can manually mount it just fine from the host os, but the ovirt-engine complains of privileges errors. I also had trouble getting the host recognized in ovirt dues to a messed up sudoers file. With that sorted, it added and report fine. Any troubleshooting tips? Thread-84::DEBUG::2014-02-02 16:08:55,459::BindingXMLRPC::167::vds::(wrapper) client [10.1.9.11] Thread-84::DEBUG::2014-02-02 16:08:55,459::task::579::TaskManager.Task::(_updateState) Task=`cc9c6dac-ed6e-4cc9-a3e4-b741c5137caf`::moving from state init - state preparing Thread-84::INFO::2014-02-02 16:08:55,460::logUtils::44::dispatcher::(wrapper) Run and protect: getSpmStatus(spUUID='b780d909-b6a0-4a38-82e2-3d7fd3a2b745', options=None) Thread-84::INFO::2014-02-02 16:08:55,460::logUtils::47::dispatcher::(wrapper) Run and protect: getSpmStatus, Return response: {'spm_st': {'spmId': 1, 'spmStatus': 'SPM', 'spmLver': 1}} Thread-84::DEBUG::2014-02-02 16:08:55,460::task::1168::TaskManager.Task::(prepare) Task=`cc9c6dac-ed6e-4cc9-a3e4-b741c5137caf`::finished: {'spm_st': {'spmId': 1, 'spmStatus': 'SPM', 'spmLver': 1}} Thread-84::DEBUG::2014-02-02 16:08:55,460::task::579::TaskManager.Task::(_updateState) Task=`cc9c6dac-ed6e-4cc9-a3e4-b741c5137caf`::moving from state preparing - state finished Thread-84::DEBUG::2014-02-02 16:08:55,460::resourceManager::939::ResourceManager.Owner::(releaseAll) Owner.releaseAll requests {} resources {} Thread-84::DEBUG::2014-02-02 16:08:55,461::resourceManager::976::ResourceManager.Owner::(cancelAll) Owner.cancelAll requests {} Thread-84::DEBUG::2014-02-02 16:08:55,461::task::974::TaskManager.Task::(_decref) Task=`cc9c6dac-ed6e-4cc9-a3e4-b741c5137caf`::ref 0 aborting False Thread-85::DEBUG::2014-02-02 16:08:55,470::BindingXMLRPC::167::vds::(wrapper) client [10.1.9.11] Thread-85::DEBUG::2014-02-02 16:08:55,470::task::579::TaskManager.Task::(_updateState) Task=`59aa1ec2-64b9-4c6d-bdb8-66154f73915d`::moving from state init - state preparing Thread-85::INFO::2014-02-02 16:08:55,470::logUtils::44::dispatcher::(wrapper) Run and protect: getStoragePoolInfo(spUUID='b780d909-b6a0-4a38-82e2-3d7fd3a2b745', options=None) Thread-85::DEBUG::2014-02-02 16:08:55,471::resourceManager::197::ResourceManager.Request::(__init__) ResName=`Storage.b780d909-b6a0-4a38-82e2-3d7fd3a2b745`ReqID=`7bbb6abd-006b- 4b7b-b540-322dfdda7644`::Request was made in '/usr/share/vdsm/storage/hsm.py' line '2553' at 'getStoragePoolInfo' Thread-85::DEBUG::2014-02-02 16:08:55,471::resourceManager::541::ResourceManager::(registerResource) Trying to register resource 'Storage.b780d909-b6a0-4a38-82e2-3d7fd3a2b745' for lock type 'shared' Thread-85::DEBUG::2014-02-02 16:08:55,471::resourceManager::600::ResourceManager::(registerResource) Resource 'Storage.b780d909-b6a0-4a38-82e2-3d7fd3a2b745' is free. Now locking as 'shared' (1 active user) Thread-85::DEBUG::2014-02-02 16:08:55,471::resourceManager::237::ResourceManager.Request::(grant) ResName=`Storage.b780d909-b6a0-4a38-82e2-3d7fd3a2b745`ReqID=`7bbb6abd-006b- 4b7b-b540-322dfdda7644`::Granted request Thread-85::DEBUG::2014-02-02 16:08:55,472::task::811::TaskManager.Task::(resourceAcquired) Task=`59aa1ec2-64b9-4c6d-bdb8-66154f73915d`::_resourcesAcquired: Storage.b780d909-b6a0-4a38-82e2-3d7fd3a2b745 (shared) Thread-85::DEBUG::2014-02-02 16:08:55,472::task::974::TaskManager.Task::(_decref) Task=`59aa1ec2-64b9-4c6d-bdb8-66154f73915d`::ref 1 aborting False Thread-85::INFO::2014-02-02 16:08:55,472::logUtils::47::dispatcher::(wrapper) Run and protect: getStoragePoolInfo, Return response: {'info': {'spm_id': 1, 'master_uuid': 'd9f70b92-3c38-4503-ac69-b48325ead406', 'name': 'hnwdb05-Local', 'version': '3', 'domains': 'd9f70b92-3c38-4503-ac69-b48325ead406:Active', 'pool_status': 'connected', 'isoprefix': '', 'type': 'LOCALFS', 'master_ver': 1, 'lver': 1}, 'dominfo': {'d9f70b92-3c38-4503-ac69-b48325ead406': {'status': 'Active', 'diskfree': '409127440384', 'alerts': [], 'version': 3, 'disktotal': '945068838912'}}} Thread-85::DEBUG::2014-02-02 16:08:55,473::task::1168::TaskManager.Task::(prepare) Task=`59aa1ec2-64b9-4c6d-bdb8-66154f73915d`::finished: {'info
Re: [Users] online storage domain resize
Hi, If you want to enlarge an existing storage domain the easiest way is to extend the storage domain. You can edit the domain and add another LUN to be a part of that domain and oVirt will take care of extending the domain - the only thing you need is another LUN to use. Thanks, Gadi Ickowicz - Original Message - From: Jiří Sléžka jiri.sle...@slu.cz To: users@ovirt.org Sent: Monday, January 20, 2014 4:57:34 PM Subject: [Users] online storage domain resize Hello, I'm just curious and I didn't try it already. I'm using FC storage (Dell MD3620f) with some logical disks on it. I should be able online increase virtual disk capacity using storage management (I have some free capacity on disk group). Is there any way to on-line extend volume group used for vm's images storage and don't break anything? I just found this hint by Eduardo from list 1. Shutdown all VMs 2. Manually connect iscsi on the SPM host 3. Run pvresize on the LUN 4. Put the domain in maintenance 5. Activate the domains Is it possible to do this on-line without shutting down all vms? If not, it could be really nice feature for oncoming releases. Thanks in advance Jiri ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Engine installation issue
If you are trying to access REST you have to specify full usernam@domain so try: admin@internal Gadi Ickowicz - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Yedidyah Bar David d...@redhat.com Cc: users@ovirt.org Sent: Thursday, January 16, 2014 9:26:37 AM Subject: Re: [Users] Engine installation issue For some reason, the password I put during installation is not working. I've tried the following combinations admin password-set-during-install root password-set-during-install Engine password-set-during-install ENGINE password-set-during-install engine password-set-during-install None of these work. On Thu, Jan 16, 2014 at 11:54 AM, Yedidyah Bar David d...@redhat.com wrote: That's two different questions... The engine's database, if created locally by setup, has a random password, and you can see all the credentials in /etc/ovirt-engine/engine.conf.d/10-setup-database.conf . The REST API is accessible using an engine's user/password, not the database's ones. You supplied the password for the user 'admin' during setup. -- Didi From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Alon Bar-Lev alo...@redhat.com Cc: users@ovirt.org Sent: Thursday, January 16, 2014 8:35:47 AM Subject: Re: [Users] Engine installation issue Thank you very much. That worked perfectly. One more question. What's the default username and password for the engine database for accessing through REST? On Wed, Jan 15, 2014 at 3:37 PM, Alon Bar-Lev alo...@redhat.com wrote: Job for postgresql.service failed. See 'systemctl status postgresql.service' and 'journalctl -xn' for details. I guess this is related to[1]. Please try: # engine-setup --otopi-environment=OVESETUP_SYSTEM/shmmax=int:68719476736 Regards, Alon Bar-Lev. [1] https://bugzilla.redhat.com/show_bug.cgi?id=1039616 - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Alon Bar-Lev alo...@redhat.com Cc: users@ovirt.org Sent: Wednesday, January 15, 2014 12:32:33 PM Subject: Re: [Users] Engine installation issue There you go. On Wed, Jan 15, 2014 at 2:47 PM, Alon Bar-Lev alo...@redhat.com wrote: Hi, Can you please attach[1]? Thanks! [1] /var/log/ovirt-engine/setup/ovirt-engine-setup-20140115032633.log - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: users@ovirt.org Sent: Wednesday, January 15, 2014 11:44:41 AM Subject: [Users] Engine installation issue Hello all, I can't seem to run engine setup on my machine. I get the following error. Googled around, can't find a solution [ INFO ] Creating PostgreSQL database [ INFO ] Configuring PostgreSQL [ ERROR ] Failed to execute stage 'Misc configuration': Command '/bin/systemctl' failed to execute [ INFO ] Yum Performing yum transaction rollback [ INFO ] Stage: Clean up Log file is located at /var/log/ovirt-engine/setup/ovirt-engine-setup-20140115032633.log [ INFO ] Stage: Pre-termination [ INFO ] Stage: Termination [ ERROR ] Execution of setup failed It happened to me on two machines which kinda makes me think someone else also must have come across this. A little help would be great. Regards Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- Didi -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Cannot start vm
Adding Nauman - he had a similar problem starting a vm: libvirtError: Child quit during startup handshake: Input/output error and we were unable to solve it. Nauman - Please try the solution Dafna suggested below: /usr/lib/systemd/systemd-vdsmd reconfigure if that doesn't work try: /usr/lib/systemd/systemd-vdsmd reconfigure force I am not sure that you have the same issue, but it might be worth it to try. Also you can check in the qemu log of the vm if you see something like this: libvirt: Lock Driver error : unsupported configuration: Read/write, exclusive access, disks were present, but no leases specified 2014-01-03 13:52:11.306+: shutting down Thanks, Gadi Ickowicz - Original Message - From: Dafna Ron d...@redhat.com To: Oliver Albl oliver.a...@fabasoft.com Cc: users@ovirt.org Sent: Friday, January 3, 2014 5:10:18 PM Subject: Re: [Users] Host cannot access storage domains awesome! Thanks for reporting this and sticking out with me :) Dafna On 01/03/2014 03:08 PM, Albl, Oliver wrote: Dafna, /usr/lib/systemd/systemd-vdsmd reconfigure force worked! VMs start and can be migrated. Thanks a lot for your help - and I'll stay with the node iso image :) All the best, Oliver -Ursprüngliche Nachricht- Von: Alon Bar-Lev [mailto:alo...@redhat.com] Gesendet: Freitag, 03. Jänner 2014 16:00 An: Albl, Oliver Cc: d...@redhat.com; users@ovirt.org Betreff: Re: [Users] Host cannot access storage domains - Original Message - From: Oliver Albl oliver.a...@fabasoft.com To: d...@redhat.com Cc: users@ovirt.org Sent: Friday, January 3, 2014 4:56:33 PM Subject: Re: [Users] Host cannot access storage domains Redirecting to /bin/systemctl reconfigure vdsmd.service Unknown operation 'reconfigure'. /usr/lib/systemd/systemd-vdsmd reconfigure force ... seems to me, I should get rid of the ovirt-node iso installation and move to a rpm based install? Thanks, Oliver -Ursprüngliche Nachricht- Von: Dafna Ron [mailto:d...@redhat.com] Gesendet: Freitag, 03. Jänner 2014 15:51 An: Albl, Oliver Cc: users@ovirt.org Betreff: Re: AW: AW: AW: AW: AW: [Users] Host cannot access storage domains can you run: service vdsmd reconfigure on the second host? On 01/03/2014 02:43 PM, Albl, Oliver wrote: Dafna, yes, the VM starts on the first node, the issues are on the second node only. /etc/libvirt/qemu-sanlock.conf is identical on on both nodes: auto_disk_leases=0 require_lease_for_disks=0 yum updates reports Using yum is not supported... Thanks, Oliver -Ursprüngliche Nachricht- Von: Dafna Ron [mailto:d...@redhat.com] Gesendet: Freitag, 03. Jänner 2014 15:39 An: Albl, Oliver Cc: users@ovirt.org Betreff: Re: AW: AW: AW: AW: [Users] Host cannot access storage domains ok, let's try to zoom in on the issue... can you run vm's on the first host or do you have issues only on the second host you added? can you run on both hosts? # egrep -v ^# /etc/libvirt/qemu-sanlock.conf can you run yum update on one of the hosts and see if there are newer packages? Thanks, Dafna On 01/03/2014 02:30 PM, Albl, Oliver wrote: I installed both hosts using the oVirt Node ISO image: OS Version: oVirt Node - 3.0.3 - 1.1.fc19 Kernel Version: 3.11.9 - 200.fc19.x86_64 KVM Version: 1.6.1 - 2.fc19 LIBVIRT Version: libvirt-1.1.3.1-2.fc19 VDSM Version: vdsm-4.13.0-11.fc19 Thanks, Oliver -Ursprüngliche Nachricht- Von: Dafna Ron [mailto:d...@redhat.com] Gesendet: Freitag, 03. Jänner 2014 15:24 An: Albl, Oliver Cc: users@ovirt.org Betreff: Re: AW: AW: AW: [Users] Host cannot access storage domains ignore the link :) so searching for this error I hit an old bug and it seemed to be an issue between libvirt/sanlock. https://bugzilla.redhat.com/show_bug.cgi?id=828633 are you using latest packages? On 01/03/2014 02:15 PM, Albl, Oliver wrote: Dafna, Libvirtd.log shows no errors, but VM log shows the following: 2014-01-03 13:52:11.296+: starting up LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin QEMU_AUDIO_DRV=spice /usr/bin/qemu-kvm -name OATEST2 -S -machine pc-1.0,accel=kvm,usb=off -cpu SandyBridge -m 1024 -realtime mlock=off -smp 1,sockets=1,cores=1,threads=1 -uuid d2bddcdb-a2c8-4c77-b0cf-b83fa3c2a0b6 -smbios type=1,manufacturer=oVirt,product=oVirt Node,version=3.0.3-1.1.fc19,serial=30313436-3631-5A43-4A33-3332304 C3 8 4 C,uuid=d2bddcdb-a2c8-4c77-b0cf-b83fa3c2a0b6 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/OATEST2.monitor,s er v e r,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=2014-01-03T13:52:11,driftfix=slew -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -device virtio-scsi-pci,id=scsi0,bus=pci.0,addr=0x4 -device virtio-serial-pci,id=virtio-serial0,bus=pci.0,addr=0x3 -drive if=none,id=drive-ide0-1-0,readonly=on,format=raw,serial= -device ide-cd,bus=ide.1,unit=0,drive=drive
Re: [Users] VM not starting up
Hi, Maybe you can send just the relevant part of the vdsm logs that contain the attempt to launch the vm itself - it should start with vmCreate. Try to find ::call vmCreate with (... and either find the matching return vmCreate with or an maybe there is an ERROR or Traceback in the vdsm log following the vmCreate call. Gadi Ickowicz - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 11:47:12 AM Subject: Re: [Users] VM not starting up Vdsm is running fine. Also checked the clocks. You were right, the Engine clock was 5 hours ahead of Node clock. I've corrected that but still the same issue. On Tue, Dec 31, 2013 at 12:44 PM, Meital Bourvine mbour...@redhat.com wrote: Are you sure that vdsm is running? The error happened at 2013-12-31 11:32:56,999 (engine.log) but the last line in vdsm.log is from 2013-12-31 07:20:50,807 Also make sure that your clocked are synced. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 9:39:09 AM Subject: Re: [Users] VM not starting up Sending files one by one. Vdsm.log On Tue, Dec 31, 2013 at 11:48 AM, Meital Bourvine mbour...@redhat.com wrote: Engine log can be found on the engine (while vdsm.log is on the host). The full path is - /var/log/ovirt-engine/engine.log audit.log - /var/log/audit/audit.log About the vdsm log - I don't see any error there, I think that you cut too much from it. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 8:42:43 AM Subject: Re: [Users] VM not starting up Hello Meital SELinux is off. The cluster doesn't work if SELinux is on, my hosts and storage is showing up as okay but VM is not starting up. I've managed to get vdsm.log but cannot find engine.log and audit.log. Sorry for being a noob, if you can tell me the path to these files I'll send them right away. On Tue, Dec 31, 2013 at 11:17 AM, Meital Bourvine mbour...@redhat.com wrote: Hi Nauman, Maybe it's a selinux issue? Please try running `setenforce 0` before starting the vm. Also please attach engine.log, vdsm.log, and audit.log From: Nauman Abbas nauman.ab...@seecs.edu.pk To: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 7:13:01 AM Subject: [Users] VM not starting up Hello all, First of all thank you for your help, I've finally managed to get my Cluster configured properly but now that I create VMs I'm facing another problem. Whenever I start my VM I get this error. VM TestVm is down. Exit message: Child quit during startup handshake: Input/output error. I've googled around but haven't found any satisfactory solution. If anyone can help out it would be great. Regards Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] VM not starting up
It would be helpful to get a bit more information from the logs - after the vmCreate there are several other things that occur before the vm process is actually launched (prepareImage and following that the actual vm's should be visible in the log). Try to search for an error in the vdsm log, and if you cannot find one attach logs containing the vmCreate and the other operations I mentioned. Errors in the vdsm log are usually logged in lines containing ERROR, but sometimes if the error comes from libvirt it may not contain the uppercase ERROR and be logged as libvirtError or something similar. you should also have the qemu log for the vm you are trying to run. usually under /var/log/libvirt/qemu/name_of_the_vm.log - maybe that will give more information regarding the failure Gadi Ickowicz - Original Message - From: Joop jvdw...@xs4all.nl To: Nauman Abbas nauman.ab...@seecs.edu.pk Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 12:51:10 PM Subject: Re: [Users] VM not starting up Nauman Abbas nauman.ab...@seecs.edu.pk wrote: I've attached the portions with vmcreate. I found 5 instances of pairs of call vmcreate and return vmcreate. I've separated them with lines. See if that helps. On Tue, Dec 31, 2013 at 2:58 PM, Gadi Ickowicz gicko...@redhat.com wrote: Hi, Maybe you can send just the relevant part of the vdsm logs that contain the attempt to launch the vm itself - it should start with vmCreate. Try to find ::call vmCreate with (... and either find the matching return vmCreate with or an maybe there is an ERROR or Traceback in the vdsm log following the vmCreate call. Gadi Ickowicz - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 11:47:12 AM Subject: Re: [Users] VM not starting up Vdsm is running fine. Also checked the clocks. You were right, the Engine clock was 5 hours ahead of Node clock. I've corrected that but still the same issue. On Tue, Dec 31, 2013 at 12:44 PM, Meital Bourvine mbour...@redhat.com wrote: Are you sure that vdsm is running? The error happened at 2013-12-31 11:32:56,999 (engine.log) but the last line in vdsm.log is from 2013-12-31 07:20:50,807 Also make sure that your clocked are synced. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 9:39:09 AM Subject: Re: [Users] VM not starting up Sending files one by one. Vdsm.log On Tue, Dec 31, 2013 at 11:48 AM, Meital Bourvine mbour...@redhat.com wrote: Engine log can be found on the engine (while vdsm.log is on the host). The full path is - /var/log/ovirt-engine/engine.log audit.log - /var/log/audit/audit.log About the vdsm log - I don't see any error there, I think that you cut too much from it. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 8:42:43 AM Subject: Re: [Users] VM not starting up Hello Meital SELinux is off. The cluster doesn't work if SELinux is on, my hosts and storage is showing up as okay but VM is not starting up. I've managed to get vdsm.log but cannot find engine.log and audit.log. Sorry for being a noob, if you can tell me the path to these files I'll send them right away. On Tue, Dec 31, 2013 at 11:17 AM, Meital Bourvine mbour...@redhat.com wrote: Hi Nauman, Maybe it's a selinux issue? Please try running `setenforce 0` before starting the vm. Also please attach engine.log, vdsm.log, and audit.log From: Nauman Abbas nauman.ab...@seecs.edu.pk To: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 7:13:01 AM Subject: [Users] VM not starting up Hello all, First of all thank you for your help, I've finally managed to get my Cluster configured properly but now that I create VMs I'm facing another problem. Whenever I start my VM I get this error. VM TestVm is down. Exit message: Child quit during startup handshake: Input/output error. I've googled around but haven't found any satisfactory solution. If anyone can help out it would be great. Regards Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 -- Nauman Abbas Assistant System Administrator (LMS), Room No. A-207, SEECS, National University of Sciences Technology, + 92 321 5359946 -- Nauman Abbas Assistant System Administrator
Re: [Users] VM not starting up
the lines you added that are right before the Fpaste link are the actual lines messages from libvirt that the vm dies. This message from libvirt is then shown later as the ERROR in vdsm in the Fpaste link. The log should contain, very close to the lines you paste, but a bit before, an xml. it should look something like what I am pasting below. Also, the line right before the xml should contain a prepareVolumePath entry: Thread-50::DEBUG::2013-12-25 14:30:45,282::vm::3031::vm.Vm::(_run) vmId=`fdfaec39-b981-4bd7-9852-6d5ac6ce2b37`::?xml version=1.0 encoding=utf-8? ...the xml continues a bit with all the vm properties also - what versions of vdsm, libvirt, sanlock and qemu are you using? Gadi Ickowicz - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Gadi Ickowicz gicko...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Wednesday, January 1, 2014 7:30:24 AM Subject: Re: [Users] VM not starting up Also I believe these lines right before what I pasted with Fpaste link would be relevant. Thread-622::DEBUG::2013-12-31 05:06:02,957::libvirtconnection::108::libvirtconnection::(wrapper) Unknown libvirterror: ecode: 38 edom: 0 level: 2 message: Child quit during startup handshake: Input/output error Thread-622::DEBUG::2013-12-31 05:06:02,957::vm::2106::vm.Vm::(_startUnderlyingVm) vmId=`147a1deb-3795-4153-a820-ceabd8db15f3`::_ongoingCreations released On Wed, Jan 1, 2014 at 10:23 AM, Nauman Abbas nauman.ab...@seecs.edu.pkwrote: I'm going through the log (which is quite big) and I've found out this part which is like 2 seconds after the vmcreate part. Have a look maybe this is what you're looking for. It contains the ERROR and Traceback in it. http://ur1.ca/gakhc On Tue, Dec 31, 2013 at 4:36 PM, Gadi Ickowicz gicko...@redhat.comwrote: It would be helpful to get a bit more information from the logs - after the vmCreate there are several other things that occur before the vm process is actually launched (prepareImage and following that the actual vm's should be visible in the log). Try to search for an error in the vdsm log, and if you cannot find one attach logs containing the vmCreate and the other operations I mentioned. Errors in the vdsm log are usually logged in lines containing ERROR, but sometimes if the error comes from libvirt it may not contain the uppercase ERROR and be logged as libvirtError or something similar. you should also have the qemu log for the vm you are trying to run. usually under /var/log/libvirt/qemu/name_of_the_vm.log - maybe that will give more information regarding the failure Gadi Ickowicz - Original Message - From: Joop jvdw...@xs4all.nl To: Nauman Abbas nauman.ab...@seecs.edu.pk Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 12:51:10 PM Subject: Re: [Users] VM not starting up Nauman Abbas nauman.ab...@seecs.edu.pk wrote: I've attached the portions with vmcreate. I found 5 instances of pairs of call vmcreate and return vmcreate. I've separated them with lines. See if that helps. On Tue, Dec 31, 2013 at 2:58 PM, Gadi Ickowicz gicko...@redhat.com wrote: Hi, Maybe you can send just the relevant part of the vdsm logs that contain the attempt to launch the vm itself - it should start with vmCreate. Try to find ::call vmCreate with (... and either find the matching return vmCreate with or an maybe there is an ERROR or Traceback in the vdsm log following the vmCreate call. Gadi Ickowicz - Original Message - From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 11:47:12 AM Subject: Re: [Users] VM not starting up Vdsm is running fine. Also checked the clocks. You were right, the Engine clock was 5 hours ahead of Node clock. I've corrected that but still the same issue. On Tue, Dec 31, 2013 at 12:44 PM, Meital Bourvine mbour...@redhat.com wrote: Are you sure that vdsm is running? The error happened at 2013-12-31 11:32:56,999 (engine.log) but the last line in vdsm.log is from 2013-12-31 07:20:50,807 Also make sure that your clocked are synced. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 9:39:09 AM Subject: Re: [Users] VM not starting up Sending files one by one. Vdsm.log On Tue, Dec 31, 2013 at 11:48 AM, Meital Bourvine mbour...@redhat.com wrote: Engine log can be found on the engine (while vdsm.log is on the host). The full path is - /var/log/ovirt-engine/engine.log audit.log - /var/log/audit/audit.log About the vdsm log - I don't see any error there, I think that you cut too much from it. From: Nauman Abbas nauman.ab...@seecs.edu.pk To: Meital Bourvine mbour...@redhat.com Cc: users@ovirt.org Users@ovirt.org Sent: Tuesday, December 31, 2013 8:42:43 AM Subject: Re