Re: [Users] Testday aftermath
Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? Thanks, Kanagaraj On 02/01/2013 03:23 PM, Joop wrote: Yesterday was testday but not much fun :-( I had a reasonable working setup but for testday I decided to start from scratch and that ended rather soon. Installing and configuring engine was not a problem but I want a setup where I have two gluster hosts and two hosts as vmhosts. I added a second cluster using the webinterface set it to gluster storage and added two minimal installed Fedora 18 hosts where I setup static networking and verified that it worked. Adding the two hosts went OK but adding a Volume gives the following error on engine: 2013-02-01 09:32:39,084 INFO [org.ovirt.engine.core.bll.gluster.CreateGlusterVolumeCommand] (ajp--127.0.0.1-8702-4) [5ea886d] Running command: CreateGlusterVolumeCommand internal: false. Entities affected : ID: 8720debc-a184-4b61-9fa8-0fdf4d339b9a Type: VdsGroups 2013-02-01 09:32:39,117 INFO [org.ovirt.engine.core.vdsbroker.gluster.CreateGlusterVolumeVDSCommand] (ajp--127.0.0.1-8702-4) [5ea886d] START, CreateGlusterVolumeVDSCommand(HostName = st02, HostId = e7b74172-2f95-43cb-83ff-11705ae24265), log id: 4270f4ef 2013-02-01 09:32:39,246 WARN [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (ajp--127.0.0.1-8702-4) [5ea886d] Weird return value: StatusForXmlRpc [mCode=4106, mMessage=XML error error: 2 st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data tcp2GlusterData2 ] 2013-02-01 09:32:39,248 WARN [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (ajp--127.0.0.1-8702-4) [5ea886d] Weird return value: StatusForXmlRpc [mCode=4106, mMessage=XML error error: 2 st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data tcp2GlusterData2 ] 2013-02-01 09:32:39,249 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (ajp--127.0.0.1-8702-4) [5ea886d] Failed in CreateGlusterVolumeVDS method 2013-02-01 09:32:39,250 ERROR [org.ovirt.engine.core.vdsbroker.vdsbroker.BrokerCommandBase] (ajp--127.0.0.1-8702-4) [5ea886d] Error code unexpected and error message VDSGenericException: VDSErrorException: Failed to CreateGlusterVolumeVDS, error = XML error error: 2 st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data tcp2GlusterData2 2013-02-01 09:32:39,254 ERROR [org.ovirt.engine.core.vdsbroker.VDSCommandBase] (ajp--127.0.0.1-8702-4) [5ea886d] Command CreateGlusterVolumeVDS execution failed. Exception: VDSErrorException: VDSGenericException: VDSErrorException: Failed to CreateGlusterVolumeVDS, error = XML error error: 2 st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data tcp2GlusterData2 2013-02-01 09:32:39,255 INFO [org.ovirt.engine.core.vdsbroker.gluster.CreateGlusterVolumeVDSCommand] (ajp--127.0.0.1-8702-4) [5ea886d] FINISH, CreateGlusterVolumeVDSCommand, log id: 4270f4ef 2013-02-01 09:32:39,256 ERROR [org.ovirt.engine.core.bll.gluster.CreateGlusterVolumeCommand] (ajp--127.0.0.1-8702-4) [5ea886d] Command org.ovirt.engine.core.bll.gluster.CreateGlusterVolumeCommand throw Vdc Bll exception. With error message VdcBLLException: org.ovirt.engine.core.vdsbroker.vdsbroker.VDSErrorException: VDSGenericException: VDSErrorException: Failed to CreateGlusterVolumeVDS, error = XML error error: 2 st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data tcp2GlusterData2 2013-02-01 09:32:39,268 INFO [org.ovirt.engine.core.bll.gluster.CreateGlusterVolumeCommand] (ajp--127.0.0.1-8702-4) [5ea886d] Lock freed to object EngineLock [exclusiveLocks= key: 8720debc-a184-4b61-9fa8-0fdf4d339b9a value: GLUSTER , sharedLocks= ] 2013-02-01 09:32:40,902 INFO [org.ovirt.engine.core.vdsbroker.gluster.GlusterVolumesListVDSCommand] (QuartzScheduler_Worker-85) START, GlusterVolumesListVDSCommand(HostName = st02, HostId = e7b74172-2f95-43cb-83ff-11705ae24265), log id: 61cafb32 And on ST01 the following in vdsm.log Thread-644::DEBUG::2013-02-01 10:24:06,378::BindingXMLRPC::913::vds::(wrapper) client [192.168.216.150]::call volumeCreate with ('GlusterData', ['st01.nieuwland.nl:/home/gluster-data', 'st02.nieuwland.nl:/home/gluster-data'], 2, 0, ['TCP']) {} MainProcess|Thread-644::DEBUG::2013-02-01 10:24:06,381::misc::84::Storage.Misc.excCmd::() '/usr/sbin/gluster --mode=script volume create GlusterData replica 2 transport TCP st01.nieuwland.nl:/home/gluster-data st02.nieuwland.nl:/home/gluster-data --xml' (cwd None) MainProcess|Thread-644::DEBUG::2013-02-01 10:24:06,639::misc::84::Storage.Misc.excCmd::() SUCCESS: = ''; = 0 MainProcess|Thread-644::ERROR::2013-02-01 10:24:06,640::supervdsmServer::80::SuperVdsm.ServerCallback::(wrapper) Error in wrapper Traceback (most recent call last): File "/usr/share/vdsm/supervdsmServer.py", line 78, in wrapper return func(*args, **k
Re: [Users] Testday aftermath
On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 Joop ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. To work around this issue, you could try: yum --disablerepo="*" --enablerepo="gluster-nieuw" install glusterfs glusterfs-fuse glusterfs-geo-replication glusterfs-server Joop ___ 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] Testday aftermath
Shireesh Anjal wrote: On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. To work around this issue, you could try: yum --disablerepo="*" --enablerepo="gluster-nieuw" install glusterfs glusterfs-fuse glusterfs-geo-replication glusterfs-server [root@st01 ~]# yum --disablerepo="*" --enablerepo="gluster-nieuw" install glusterfs glusterfs-fuse glusterfs-geo-replication glusterfs-server Loaded plugins: langpacks, presto, refresh-packagekit Package matching glusterfs-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Package matching glusterfs-fuse-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Package matching glusterfs-server-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Resolving Dependencies --> Running transaction check ---> Package glusterfs-geo-replication.x86_64 0:v3.4.0qa7-1.el6 will be installed --> Processing Dependency: glusterfs = v3.4.0qa7-1.el6 for package: glusterfs-geo-replication-v3.4.0qa7-1.el6.x86_64 --> Finished Dependency Resolution Error: Package: glusterfs-geo-replication-v3.4.0qa7-1.el6.x86_64 (gluster-nieuw) Requires: glusterfs = v3.4.0qa7-1.el6 Installed: glusterfs-3.3.1-8.fc18.x86_64 (@updates) glusterfs = 3.3.1-8.fc18 Available: glusterfs-v3.4.0qa7-1.el6.x86_64 (gluster-nieuw) glusterfs = v3.4.0qa7-1.el6 You could try using --skip-broken to work around the problem You could try running: rpm -Va --nofiles --nodigest [root@st01 ~]# yum --disablerepo="*" --enablerepo="gluster-nieuw" install glusterfs glusterfs-fuse glusterfs-geo-replication glusterfs-server --skip-broken Loaded plugins: langpacks, presto, refresh-packagekit Package matching glusterfs-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Package matching glusterfs-fuse-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Package matching glusterfs-server-v3.4.0qa7-1.el6.x86_64 already installed. Checking for update. Resolving Dependencies --> Running transaction check ---> Package glusterfs-geo-replication.x86_64 0:v3.4.0qa7-1.el6 will be installed --> Processing Dependency: glusterfs = v3.4.0qa7-1.el6 for package: glusterfs-geo-replication-v3.4.0qa7-1.el6.x86_64 gluster-nieuw/filelists | 7.2 kB 00:00:00 Packages skipped because of dependency problems: glusterfs-geo-replication-v3.4.0qa7-1.el6.x86_64 from gluster-nieuw Last post, probably, until sunday-evening/monday morning, off to fosdem ;-) Joop ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
Shireesh Anjal wrote: On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. The 'v' is 99.9% the culprit. I had 3.4.0qa6 before I wiped and just had a look that folder and repo doesn't have the 'v' in front of it. Is there someone on this list that has the 'powers' to change that ?? Joop ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
On 02/01/2013 06:47 PM, Joop wrote: Shireesh Anjal wrote: On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. The 'v' is 99.9% the culprit. I had 3.4.0qa6 before I wiped and just had a look that folder and repo doesn't have the 'v' in front of it. Thats correct. [kanagaraj@localhost ~]$ rpmdev-vercmp glusterfs-3.3.1-8.fc18.x86_64 glusterfs-v3.4.0qa7-1.el6.x86_64 glusterfs-3.3.1-8.fc18.x86_64 > glusterfs-v3.4.0qa7-1.el6.x86_64 Is there someone on this list that has the 'powers' to change that ?? [Adding Vijay] Joop ___ 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] Testday aftermath
On 02/01/2013 07:38 PM, Kanagaraj wrote: On 02/01/2013 06:47 PM, Joop wrote: Shireesh Anjal wrote: On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. The 'v' is 99.9% the culprit. I had 3.4.0qa6 before I wiped and just had a look that folder and repo doesn't have the 'v' in front of it. Thats correct. [kanagaraj@localhost ~]$ rpmdev-vercmp glusterfs-3.3.1-8.fc18.x86_64 glusterfs-v3.4.0qa7-1.el6.x86_64 glusterfs-3.3.1-8.fc18.x86_64 > glusterfs-v3.4.0qa7-1.el6.x86_64 Is there someone on this list that has the 'powers' to change that ?? [Adding Vijay] 3.4.0qa8 is available now. Can you please check with that? Thanks, Vijay ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
Vijay Bellur wrote: On 02/01/2013 07:38 PM, Kanagaraj wrote: On 02/01/2013 06:47 PM, Joop wrote: Shireesh Anjal wrote: On 02/01/2013 05:13 PM, noc wrote: On 1-2-2013 11:07, Kanagaraj wrote: Hi Joop, Looks like the problem is because of the glusterfs version you are using. vdsm could not parse the output from gluster. Can you update the glusterfs to http://bits.gluster.org/pub/gluster/glusterfs/v3.4.0qa7/x86_64/ and check it out? How?? I tried adding this repo but but yum says that there are no updates available, atleast yesterday it did. [gluster-nieuw] name=GlusterFS baseurl=http://bits.gluster.org/pub/gluster/glusterfs/stage/ gpgcheck=0 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-Gluster enabled=1 My yumfoo isn't that good so I don't know how to force it. Besides I tried through yum localinstall but it will revert when yum update is run. It looks like it thinks that 3.3.1 is newer than 3.4 The problem is that, released glusterfs rpms in fedora repository are of the form 3.3.1-8, whereas the ones from above QA release are v3.4.0qa7. I think because of the "v" before 3.4, these are considered as lower version, and by default yum picks up the rpms from fedora repository. The 'v' is 99.9% the culprit. I had 3.4.0qa6 before I wiped and just had a look that folder and repo doesn't have the 'v' in front of it. Thats correct. [kanagaraj@localhost ~]$ rpmdev-vercmp glusterfs-3.3.1-8.fc18.x86_64 glusterfs-v3.4.0qa7-1.el6.x86_64 glusterfs-3.3.1-8.fc18.x86_64 > glusterfs-v3.4.0qa7-1.el6.x86_64 Is there someone on this list that has the 'powers' to change that ?? [Adding Vijay] 3.4.0qa8 is available now. Can you please check with that? Thanks, Vijay I can report back that 3.4.0qa8 does work. oVirt also picked up on the volumes that I created but didn't show up in the interface. Could start them an will test if they are fully usable. Thanks for the quick respons. Joop ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
On 3-2-2013 19:59, Joop wrote: I can report back that 3.4.0qa8 does work. oVirt also picked up on the volumes that I created but didn't show up in the interface. Could start them an will test if they are fully usable. I added my export domain and re-imported the VMs and all is well despite that the original creation of the glustervolumes failed because of gluster version 3.3.1. So lets TEST ;-) Joop ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Re: [Users] Testday aftermath
On 22-2-2013 11:37, Josué Delgado wrote: Hello, I had the same XML ERROR when i tried to configure GlusterFS on a node. Followed up the instructions and i updated glusterfs, now i have the following packages: glusterfs-server-3.4.0alpha-1.el6.x86_64 glusterfs-fuse-3.4.0alpha-1.el6.x86_64 vdsm-gluster-4.10.3-8.fc18.noarch glusterfs-3.4.0alpha-1.el6.x86_64 But now i have the following error when i tried to start glusterd: " systemctl start glusterd Failed to issue method call: Access denied " On the logs i have this: "Starting glusterd (via systemctl): Failed to issue method call: Access denied Feb 22 05:32:22 grc003 systemd[1]: Failed to get security context on /usr/lib/systemd/system/glusterd.service: No such file or directory [FAILED] " There's no systemd init script for gluster, but there's a script in /etc/init.d . Is that the cause of problem? It's the package somehow broken? What can i do now to fix and start glusterfs in order to test again? check your owner:group on the gluster folders, they should be 36:36. You can achieve this by using the 'optimize' button in the webui while looking at the Volume. Joop -- irc: jvandewege ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users