Re: [ovirt-users] Power Management config on Ovirt
Hi Eli/Martin, On each host power management I have given the other hosts iRMC IPMI address. Is that correct or I have to give same hosts management IP? Ovirt engine is installed on another machine and it can reach the iRMC management IP of each host. My setup is as below. 2 hosts with 6 NICs configured as bond0 on each host. Another IP given for iRMC management port on each host and this IP is configured on the other host power management. VM stored on glusterfs storage and engine installed on another server. I can able to create VM and manually migrate but if one host is down, VM is not migrating to other host. What do I add more to achieve this. Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com] Sent: Wednesday, March 18, 2015 5:51 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" > To: "Eli Mesika" > Cc: "Martin Perina" , users@ovirt.org > Sent: Wednesday, March 18, 2015 3:15:40 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Eli, > > Those 2 hosts are connected with Fujitsu iRMC management port and power > management is configured with ipmi. So it can use this connection to fence > the other node, is it correct? No, keep in mind that the one that communicates with the proxy host is the oVirt engine, so , if it is not accessable, oVirt engine can not use it > > Regards > > Renchu Mathew > > -Original Message- > From: Eli Mesika [mailto:emes...@redhat.com] > Sent: Wednesday, March 18, 2015 4:31 PM > To: Renchu Mathew > Cc: Martin Perina; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: users@ovirt.org > > Sent: Wednesday, March 18, 2015 2:24:46 PM > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > My setup meets all those requirements and I can able to migrate the VM > > from one host to another manually. Once network cable is pulled off > > from one of the server other server is also shuts down. > > Hi > Sorry for jumping in late , Yesterday was an election day in ISRAEL... > > If the other server shuts down when you plug-off the first one and you have > only 2 hosts then no fencing will take place since there is no available > proxy host to perform the operation > > > > > Regards > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > > +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | > > LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB > > www.cracknell.com > > > > This email, its content and any files transmitted with it are intended > > solely for the addressee(s) and may be legally privileged and/or > > confidential. If you are not the intended recipient please let us know > > by email reply and delete it from the system. Please note that any > > views or opinions presented in this email do not necessarily represent > > those of the company. Email transmissions cannot be guaranteed to be > > secure or error-free as information could be intercepted, corrupted, > > lost, destroyed, arrive late or incomplete, or contain viruses. The > > company therefore does not accept liability for any errors or > > omissions in the contents of this message which arise as a result of email > > transmission. > > > > > > -Original Message- > > From: Martin Perina [mailto:mper...@redhat.com] > > Sent: Tuesday, March 17, 2015 8:31 PM > > To: Renchu Mathew > > Cc: users@ovirt.org > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi, > > > > prior to the test I would check this: > > > > - Data Center status is Up > > - All hosts status is Up > > - All storage domains status is Up > > - VM is running > > > > If this is valid, you can start your fence testing. But bear in mind > > what I sent you in previous email: at least one host in DC should be > > fully functional to be able to fence non responsive host. > > > > Martin Perina > > > > - Original Message - > > > From: "Renchu Mathew" > > > To: "Martin Perina" > > > Cc: users@ovirt.org > > > Sent: Tuesday, March 17, 2015 5:03:53 PM >
Re: [ovirt-users] Power Management config on Ovirt
Hi Eli, Those 2 hosts are connected with Fujitsu iRMC management port and power management is configured with ipmi. So it can use this connection to fence the other node, is it correct? Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com] Sent: Wednesday, March 18, 2015 4:31 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org > Sent: Wednesday, March 18, 2015 2:24:46 PM > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi Martin, > > My setup meets all those requirements and I can able to migrate the VM > from one host to another manually. Once network cable is pulled off > from one of the server other server is also shuts down. Hi Sorry for jumping in late , Yesterday was an election day in ISRAEL... If the other server shuts down when you plug-off the first one and you have only 2 hosts then no fencing will take place since there is no available proxy host to perform the operation > > Regards > > Renchu Mathew | Sr. IT Administrator > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > +971 4 > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | > LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB > www.cracknell.com > > This email, its content and any files transmitted with it are intended > solely for the addressee(s) and may be legally privileged and/or > confidential. If you are not the intended recipient please let us know > by email reply and delete it from the system. Please note that any > views or opinions presented in this email do not necessarily represent > those of the company. Email transmissions cannot be guaranteed to be > secure or error-free as information could be intercepted, corrupted, > lost, destroyed, arrive late or incomplete, or contain viruses. The > company therefore does not accept liability for any errors or > omissions in the contents of this message which arise as a result of email > transmission. > > > -----Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Tuesday, March 17, 2015 8:31 PM > To: Renchu Mathew > Cc: users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi, > > prior to the test I would check this: > > - Data Center status is Up > - All hosts status is Up > - All storage domains status is Up > - VM is running > > If this is valid, you can start your fence testing. But bear in mind > what I sent you in previous email: at least one host in DC should be > fully functional to be able to fence non responsive host. > > Martin Perina > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: users@ovirt.org > > Sent: Tuesday, March 17, 2015 5:03:53 PM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin > > > > Yes, my test VM still running on this storage. Is it possible to do > > remote session and check this? > > > > Regards > > > > Renchu Mathew > > > > > > -Original Message- > > From: Martin Perina [mailto:mper...@redhat.com] > > Sent: Tuesday, March 17, 2015 7:30 PM > > To: Renchu Mathew > > Cc: users@ovirt.org > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi, > > > > this is what happened (at least what I was able to read from log): > > > > 18:18:02 - host node02 changed status to Connecting > > > > - that's OK, but prior to this I can see many errors > > from ConnectStorage*Commands. Are you sure that your storage > > was ok prior to fencing test? > > > > 18:18:51 - host node02 changed status to Non Responsive > > > > - that's OK, non responding treatment started with SSH Soft > > Fencing > > > > > > 18:18:55 - SSH Soft Fencing failed, because of no route to host > > node02 > > > > - that's OK > > > > > > 18:18:56 - power management stop was executed using node01 as fence > > proxy > > > > - problem started here, communication with node01 was timed out, we were > > not able to find any other suitable fence proxy > > > > 18:21:57 - host node01 changed status to Connection > > > > > > From this point there's is nothing that can be done, because engine > > canno
Re: [ovirt-users] Power Management config on Ovirt
Hi Martin, My setup meets all those requirements and I can able to migrate the VM from one host to another manually. Once network cable is pulled off from one of the server other server is also shuts down. Regards Renchu Mathew | Sr. IT Administrator CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB www.cracknell.com This email, its content and any files transmitted with it are intended solely for the addressee(s) and may be legally privileged and/or confidential. If you are not the intended recipient please let us know by email reply and delete it from the system. Please note that any views or opinions presented in this email do not necessarily represent those of the company. Email transmissions cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The company therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of email transmission. -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Tuesday, March 17, 2015 8:31 PM To: Renchu Mathew Cc: users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi, prior to the test I would check this: - Data Center status is Up - All hosts status is Up - All storage domains status is Up - VM is running If this is valid, you can start your fence testing. But bear in mind what I sent you in previous email: at least one host in DC should be fully functional to be able to fence non responsive host. Martin Perina - Original Message - > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org > Sent: Tuesday, March 17, 2015 5:03:53 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Martin > > Yes, my test VM still running on this storage. Is it possible to do > remote session and check this? > > Regards > > Renchu Mathew > > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Tuesday, March 17, 2015 7:30 PM > To: Renchu Mathew > Cc: users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi, > > this is what happened (at least what I was able to read from log): > > 18:18:02 - host node02 changed status to Connecting > > - that's OK, but prior to this I can see many errors > from ConnectStorage*Commands. Are you sure that your storage > was ok prior to fencing test? > > 18:18:51 - host node02 changed status to Non Responsive > > - that's OK, non responding treatment started with SSH Soft Fencing > > > 18:18:55 - SSH Soft Fencing failed, because of no route to host node02 > > - that's OK > > > 18:18:56 - power management stop was executed using node01 as fence > proxy > > - problem started here, communication with node01 was timed out, we were > not able to find any other suitable fence proxy > > 18:21:57 - host node01 changed status to Connection > > > From this point there's is nothing that can be done, because engine > cannot communicate with any host, so it cannot fix anything. > > When you are doing those tests you need at least one functional host, > otherwise you are not able to execute any fence action. > > > Martin Perina > > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: users@ovirt.org > > Sent: Tuesday, March 17, 2015 3:53:24 PM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > Please find attached the log files. > > > > Regards > > > > Renchu Mathew > > > > -Original Message- > > From: Martin Perina [mailto:mper...@redhat.com] > > Sent: Tuesday, March 17, 2015 6:40 PM > > To: Renchu Mathew > > Cc: users@ovirt.org > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi, > > > > please attach new logs, so we can investigate what has happened. > > > > Thanks > > > > Martin Perina > > > > - Original Message - > > > From: "Renchu Mathew" > > > To: "Martin Perina" > > > Cc: "Piotr Kliczewski" , > > > users@ovirt.org > > > Sent: Tuesday, March 17, 2015 3:34:11 PM > > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > > > Hi Marti
Re: [ovirt-users] Power Management config on Ovirt
Hi Martin Yes, my test VM still running on this storage. Is it possible to do remote session and check this? Regards Renchu Mathew -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Tuesday, March 17, 2015 7:30 PM To: Renchu Mathew Cc: users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi, this is what happened (at least what I was able to read from log): 18:18:02 - host node02 changed status to Connecting - that's OK, but prior to this I can see many errors from ConnectStorage*Commands. Are you sure that your storage was ok prior to fencing test? 18:18:51 - host node02 changed status to Non Responsive - that's OK, non responding treatment started with SSH Soft Fencing 18:18:55 - SSH Soft Fencing failed, because of no route to host node02 - that's OK 18:18:56 - power management stop was executed using node01 as fence proxy - problem started here, communication with node01 was timed out, we were not able to find any other suitable fence proxy 18:21:57 - host node01 changed status to Connection From this point there's is nothing that can be done, because engine cannot communicate with any host, so it cannot fix anything. When you are doing those tests you need at least one functional host, otherwise you are not able to execute any fence action. Martin Perina - Original Message ----- > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org > Sent: Tuesday, March 17, 2015 3:53:24 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Martin, > > Please find attached the log files. > > Regards > > Renchu Mathew > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Tuesday, March 17, 2015 6:40 PM > To: Renchu Mathew > Cc: users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi, > > please attach new logs, so we can investigate what has happened. > > Thanks > > Martin Perina > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: "Piotr Kliczewski" , users@ovirt.org > > Sent: Tuesday, March 17, 2015 3:34:11 PM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > I did the same as below. But is still does the same. Node01 is > > shutdown. Do you think RHEV + gluster storage will not have this > > issue. What storage do you recommend? > > > > Regards > > > > Renchu Mathew > > > > -Original Message- > > From: Martin Perina [mailto:mper...@redhat.com] > > Sent: Tuesday, March 17, 2015 5:35 PM > > To: Renchu Mathew > > Cc: Piotr Kliczewski; users@ovirt.org > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi, > > > > I don't know much about gluster configuration, but one of the > > following should work: > > > > 1. Turn off you host networking completely for example using /sbin/ifdown > >(you will need to access your node locally or using remote serial > >console, for example IPMI) > > > > 2. Block engine and all gluster hosts using iptables > > > > 3. Disable "Skip fencing if host has live lease on storage" option > >in Cluster Fencing Policy and block connection between host and engine > >(in this case storage connection won't be checked prior to > > fencing) > > > > > > Martin Perina > > > > > > - Original Message - > > > From: "Renchu Mathew" > > > To: "Martin Perina" > > > Cc: "Piotr Kliczewski" , > > > users@ovirt.org > > > Sent: Tuesday, March 17, 2015 2:26:29 PM > > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > > > Martin, > > > Yes, I need to test the VM migration. So can you please tell me > > > how to block the host connection to storage? It is a glusterfs > > > volume with distribute option as below. > > > > > > Volume Name: VM-Store > > > Type: Distribute > > > Volume ID: 59e40310-476d-449f-b700-cab69b829930 > > > Status: Started > > > Number of Bricks: 1 > > > Transport-type: tcp > > > Bricks: > > > Brick1: storage1:/VM/VM-Store > > > Options Reconfigured: > > > server.allow-insecure: on > > > storage.owner-gid: 36 > > > storage.owner-uid: 36 > > > > > > Many thanks for all of your support. > > > > &
Re: [ovirt-users] Power Management config on Ovirt
Hi Martin, I did the same as below. But is still does the same. Node01 is shutdown. Do you think RHEV + gluster storage will not have this issue. What storage do you recommend? Regards Renchu Mathew -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Tuesday, March 17, 2015 5:35 PM To: Renchu Mathew Cc: Piotr Kliczewski; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi, I don't know much about gluster configuration, but one of the following should work: 1. Turn off you host networking completely for example using /sbin/ifdown (you will need to access your node locally or using remote serial console, for example IPMI) 2. Block engine and all gluster hosts using iptables 3. Disable "Skip fencing if host has live lease on storage" option in Cluster Fencing Policy and block connection between host and engine (in this case storage connection won't be checked prior to fencing) Martin Perina - Original Message ----- > From: "Renchu Mathew" > To: "Martin Perina" > Cc: "Piotr Kliczewski" , users@ovirt.org > Sent: Tuesday, March 17, 2015 2:26:29 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Martin, > Yes, I need to test the VM migration. So can you please tell me how to > block the host connection to storage? It is a glusterfs volume with > distribute option as below. > > Volume Name: VM-Store > Type: Distribute > Volume ID: 59e40310-476d-449f-b700-cab69b829930 > Status: Started > Number of Bricks: 1 > Transport-type: tcp > Bricks: > Brick1: storage1:/VM/VM-Store > Options Reconfigured: > server.allow-insecure: on > storage.owner-gid: 36 > storage.owner-uid: 36 > > Many thanks for all of your support. > > Regards > > Renchu Mathew | Sr. IT Administrator > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > +971 4 > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | > LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB > www.cracknell.com > > This email, its content and any files transmitted with it are intended > solely for the addressee(s) and may be legally privileged and/or > confidential. If you are not the intended recipient please let us know > by email reply and delete it from the system. Please note that any > views or opinions presented in this email do not necessarily represent > those of the company. Email transmissions cannot be guaranteed to be > secure or error-free as information could be intercepted, corrupted, > lost, destroyed, arrive late or incomplete, or contain viruses. The > company therefore does not accept liability for any errors or > omissions in the contents of this message which arise as a result of email > transmission. > > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Tuesday, March 17, 2015 5:02 PM > To: Renchu Mathew > Cc: Piotr Kliczewski; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi, > > your VM was not migrated, because host node02 was not fenced. And host > node02 was not fenced, because option "Skip fencing if host has live > lease on storage" in your Cluster Fencing Policy is enabled. Prior > fencing of the host we check its storage connection (renewal of its sanlock > lease). > You should see an alert about it in Events/Alerts tab in webadmin. > > So if you want to test VM migration on your cluster, have also block > host connection to storage (and not only to engine). > > Also please take a look at "Skip fencing on cluster connectivity issues" > option in Cluster Fencing Policy. If it's enabled and set for example > to 50% and you have only two hosts in cluster, then your hosts won't > be fenced at all due to it. > > Thanks > > Martin Perina > > > - Original Message - > > From: "Renchu Mathew" > > To: "Piotr Kliczewski" > > Cc: "Piotr Kliczewski" , users@ovirt.org > > Sent: Tuesday, March 17, 2015 1:37:28 PM > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Piotr, > > > > Please find attached log files. If I stop the network of node02, the > > VM > > (Printserver) running on that is not automatically migrating to > > node01 instead node01 is getting shutdown. > > > > Regards > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > > +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 7386
Re: [ovirt-users] Power Management config on Ovirt
Martin, Yes, I need to test the VM migration. So can you please tell me how to block the host connection to storage? It is a glusterfs volume with distribute option as below. Volume Name: VM-Store Type: Distribute Volume ID: 59e40310-476d-449f-b700-cab69b829930 Status: Started Number of Bricks: 1 Transport-type: tcp Bricks: Brick1: storage1:/VM/VM-Store Options Reconfigured: server.allow-insecure: on storage.owner-gid: 36 storage.owner-uid: 36 Many thanks for all of your support. Regards Renchu Mathew | Sr. IT Administrator CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB www.cracknell.com This email, its content and any files transmitted with it are intended solely for the addressee(s) and may be legally privileged and/or confidential. If you are not the intended recipient please let us know by email reply and delete it from the system. Please note that any views or opinions presented in this email do not necessarily represent those of the company. Email transmissions cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The company therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of email transmission. -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Tuesday, March 17, 2015 5:02 PM To: Renchu Mathew Cc: Piotr Kliczewski; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi, your VM was not migrated, because host node02 was not fenced. And host node02 was not fenced, because option "Skip fencing if host has live lease on storage" in your Cluster Fencing Policy is enabled. Prior fencing of the host we check its storage connection (renewal of its sanlock lease). You should see an alert about it in Events/Alerts tab in webadmin. So if you want to test VM migration on your cluster, have also block host connection to storage (and not only to engine). Also please take a look at "Skip fencing on cluster connectivity issues" option in Cluster Fencing Policy. If it's enabled and set for example to 50% and you have only two hosts in cluster, then your hosts won't be fenced at all due to it. Thanks Martin Perina - Original Message - > From: "Renchu Mathew" > To: "Piotr Kliczewski" > Cc: "Piotr Kliczewski" , users@ovirt.org > Sent: Tuesday, March 17, 2015 1:37:28 PM > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Piotr, > > Please find attached log files. If I stop the network of node02, the > VM > (Printserver) running on that is not automatically migrating to node01 > instead node01 is getting shutdown. > > Regards > > Renchu Mathew | Sr. IT Administrator > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > +971 4 > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | > LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB > www.cracknell.com > > This email, its content and any files transmitted with it are intended > solely for the addressee(s) and may be legally privileged and/or > confidential. If you are not the intended recipient please let us know > by email reply and delete it from the system. Please note that any > views or opinions presented in this email do not necessarily represent > those of the company. Email transmissions cannot be guaranteed to be > secure or error-free as information could be intercepted, corrupted, > lost, destroyed, arrive late or incomplete, or contain viruses. The > company therefore does not accept liability for any errors or > omissions in the contents of this message which arise as a result of email > transmission. > > > -Original Message- > From: Piotr Kliczewski [mailto:piotr.kliczew...@gmail.com] > Sent: Tuesday, March 17, 2015 3:35 PM > To: Renchu Mathew > Cc: Piotr Kliczewski; Eli Mesika; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > On Tue, Mar 17, 2015 at 12:18 PM, Renchu Mathew wrote: > > Hi Piotr, > > > > I have re-installed the node with ovirt node 3.5 latest version and > > tried the power management and it shows test succeeded, on but if I > > stop the network of one of the node the running vm is not migrating > > to the other instead both the nodes are getting down. Can anybody help me > > on this? > > > > Please provide engine and both nodes logs for analysis. > > > > > Also is it possible to use the resources
Re: [ovirt-users] Power Management config on Ovirt
Hi Piotr, I have re-installed the node with ovirt node 3.5 latest version and tried the power management and it shows test succeeded, on but if I stop the network of one of the node the running vm is not migrating to the other instead both the nodes are getting down. Can anybody help me on this? Also is it possible to use the resources (CPU cores and memory) from other node in the same cluster or combine 2 nodes in same clusters and create a vm on that. Regards Renchu Mathew -Original Message- From: Piotr Kliczewski [mailto:pklic...@redhat.com] Sent: Tuesday, January 27, 2015 2:04 PM To: Renchu Mathew Cc: Piotr Kliczewski; Eli Mesika; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" > To: "Piotr Kliczewski" > Cc: "Piotr Kliczewski" , "Eli Mesika" > , users@ovirt.org > Sent: Tuesday, January 27, 2015 10:48:12 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Piotr, > > Can you please provide me the link to download the new node iso? I can > see only ovirt live iso. > I saw that Douglas sent following links in other thread [1] on this list: centos7: http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-el7_merged/ Fedora20: http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-fc20_merged/ Please note that "We should share the official iso this week." [1] http://lists.ovirt.org/pipermail/users/2015-January/030846.html > > Regards > > Renchu Mathew > > -Original Message- > From: Piotr Kliczewski [mailto:pklic...@redhat.com] > Sent: Tuesday, January 27, 2015 12:21 PM > To: Renchu Mathew > Cc: Piotr Kliczewski; Eli Mesika; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi Mathew, > > It seems like this is quite old version (September 2014). There were > good number of fixes in this area. > > Looking at provided logs I am not able to tell which call is exactly failing. > Do you have any steps to reproduce? > > Thanks, > Piotr > > - Original Message - > From: "Renchu Mathew" > To: "Piotr Kliczewski" > Cc: "Eli Mesika" , "Piotr Kliczewski" > , users@ovirt.org > Sent: Monday, January 26, 2015 5:15:46 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Piotr, > VDSM Version is vdsm-4.16.4-0.el6 > > Regards > > Renchu Mathew > . > > From: Piotr Kliczewski [mailto:piotr.kliczew...@gmail.com] > Sent: Sunday, January 25, 2015 6:08 PM > To: Renchu Mathew > Cc: Eli Mesika; Piotr Kliczewski; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > On Sun, Jan 25, 2015 at 2:07 PM, Renchu Mathew > mailto:ren...@cracknell.com>> wrote: > > Hi Eli, > > > > I have 2 hypervisor host and both are installed with below iso file. > Master data domain is glusterfs which is configured on another server. > Is there any latest iso available? > > > > ovirt-node-iso-3.5.0.ovirt35.20140912.el6 (Edited). > > @Eli - Do we know which vdsm version is in this iso? > > > > > Please find below details. > > > > [cid:image001.png@01D0393F.9A5CF2D0] > > > > Thanks & Regards > > > > Renchu Mathew > > > > -Original Message- > From: Eli Mesika > [mailto:emes...@redhat.com<mailto:emes...@redhat.com>] > Sent: Sunday, January 25, 2015 4:46 PM > To: Renchu Mathew > Cc: Martin Perina; users@ovirt.org<mailto:users@ovirt.org>; Piotr > Kliczewski > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > > > > > - Original Message - > > > From: "Renchu Mathew" > > mailto:ren...@cracknell.com>> > > > To: "Eli Mesika" mailto:emes...@redhat.com>> > > > Cc: "Martin Perina" mailto:mper...@redhat.com>>, > > users@ovirt.org<mailto:users@ovirt.org> > > > Sent: Sunday, January 25, 2015 11:19:21 AM > > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > > > Hi Eli, > > > > > > Please use any details. > > > > > > Thanks for your support. > > > > > > Regards > > > > > > Renchu Mathew > > > > Hi again > > > > Looking in the logs I see that this is a issue we handled when we > switched from XML RPC to JSON RPC protocol. > > Which vdsm is installed in the proxy host ? > > I am adding Piotr who is on charge of the JSON RPC protocol an
Re: [ovirt-users] Power Management config on Ovirt
Hi Piotr, Can you please provide me the link to download the new node iso? I can see only ovirt live iso. Regards Renchu Mathew -Original Message- From: Piotr Kliczewski [mailto:pklic...@redhat.com] Sent: Tuesday, January 27, 2015 12:21 PM To: Renchu Mathew Cc: Piotr Kliczewski; Eli Mesika; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi Mathew, It seems like this is quite old version (September 2014). There were good number of fixes in this area. Looking at provided logs I am not able to tell which call is exactly failing. Do you have any steps to reproduce? Thanks, Piotr - Original Message - From: "Renchu Mathew" To: "Piotr Kliczewski" Cc: "Eli Mesika" , "Piotr Kliczewski" , users@ovirt.org Sent: Monday, January 26, 2015 5:15:46 AM Subject: RE: [ovirt-users] Power Management config on Ovirt Hi Piotr, VDSM Version is vdsm-4.16.4-0.el6 Regards Renchu Mathew . From: Piotr Kliczewski [mailto:piotr.kliczew...@gmail.com] Sent: Sunday, January 25, 2015 6:08 PM To: Renchu Mathew Cc: Eli Mesika; Piotr Kliczewski; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt On Sun, Jan 25, 2015 at 2:07 PM, Renchu Mathew mailto:ren...@cracknell.com>> wrote: Hi Eli, I have 2 hypervisor host and both are installed with below iso file. Master data domain is glusterfs which is configured on another server. Is there any latest iso available? ovirt-node-iso-3.5.0.ovirt35.20140912.el6 (Edited). @Eli - Do we know which vdsm version is in this iso? Please find below details. [cid:image001.png@01D0393F.9A5CF2D0] Thanks & Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com<mailto:emes...@redhat.com>] Sent: Sunday, January 25, 2015 4:46 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org<mailto:users@ovirt.org>; Piotr Kliczewski Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" mailto:ren...@cracknell.com>> > To: "Eli Mesika" mailto:emes...@redhat.com>> > Cc: "Martin Perina" mailto:mper...@redhat.com>>, > users@ovirt.org<mailto:users@ovirt.org> > Sent: Sunday, January 25, 2015 11:19:21 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Eli, > > Please use any details. > > Thanks for your support. > > Regards > > Renchu Mathew Hi again Looking in the logs I see that this is a issue we handled when we switched from XML RPC to JSON RPC protocol. Which vdsm is installed in the proxy host ? I am adding Piotr who is on charge of the JSON RPC protocol and who helped me in the past when I got a similar error on PM operations It looks like marshaling issue but I am not sure whether it was already fixed or it is something new. I am not able to match a call when looking at engine.log. > > -Original Message- > From: Eli Mesika [mailto:emes...@redhat.com<mailto:emes...@redhat.com>] > Sent: Sunday, January 25, 2015 12:58 PM > To: Renchu Mathew > Cc: Martin Perina; users@ovirt.org<mailto:users@ovirt.org> > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi Renchu > > I am looking currently on he case and will respond later on today > after investigating the logs. > If I will not find anything, can I use the details for the host PM > only for status command to try to debug and find the source of the problem ? > > Thanks > > Eli > > - Original Message - > > From: "Renchu Mathew" mailto:ren...@cracknell.com>> > > To: "Martin Perina" mailto:mper...@redhat.com>> > > Cc: users@ovirt.org<mailto:users@ovirt.org>, "Eli Mesika" > > mailto:emes...@redhat.com>> > > Sent: Sunday, January 25, 2015 6:50:00 AM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > I got the below output when I run the command and it works from both > > the nodes. > > > > fence_ipmilan -a 192.168.1.114 -l admin -p admin -o status -v -P > > > > Getting status of IPMI:192.168.1.114...Spawning: '/usr/bin/ipmitool > > -I lanplus -H '192.168.1.114' -U 'admin' -P '[set]' -v chassis power > > status'... > > Chassis power = On > > Done > > > > Thanks > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > > +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 > > 7386484 ABU DHABI | DUBAI > > | L
Re: [ovirt-users] Power Management config on Ovirt
Hi Piotr, VDSM Version is vdsm-4.16.4-0.el6 Regards Renchu Mathew . From: Piotr Kliczewski [mailto:piotr.kliczew...@gmail.com] Sent: Sunday, January 25, 2015 6:08 PM To: Renchu Mathew Cc: Eli Mesika; Piotr Kliczewski; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt On Sun, Jan 25, 2015 at 2:07 PM, Renchu Mathew mailto:ren...@cracknell.com>> wrote: Hi Eli, I have 2 hypervisor host and both are installed with below iso file. Master data domain is glusterfs which is configured on another server. Is there any latest iso available? ovirt-node-iso-3.5.0.ovirt35.20140912.el6 (Edited). @Eli - Do we know which vdsm version is in this iso? Please find below details. [cid:image001.png@01D0393F.9A5CF2D0] Thanks & Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com<mailto:emes...@redhat.com>] Sent: Sunday, January 25, 2015 4:46 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org<mailto:users@ovirt.org>; Piotr Kliczewski Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" mailto:ren...@cracknell.com>> > To: "Eli Mesika" mailto:emes...@redhat.com>> > Cc: "Martin Perina" mailto:mper...@redhat.com>>, > users@ovirt.org<mailto:users@ovirt.org> > Sent: Sunday, January 25, 2015 11:19:21 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Eli, > > Please use any details. > > Thanks for your support. > > Regards > > Renchu Mathew Hi again Looking in the logs I see that this is a issue we handled when we switched from XML RPC to JSON RPC protocol. Which vdsm is installed in the proxy host ? I am adding Piotr who is on charge of the JSON RPC protocol and who helped me in the past when I got a similar error on PM operations It looks like marshaling issue but I am not sure whether it was already fixed or it is something new. I am not able to match a call when looking at engine.log. > > -Original Message- > From: Eli Mesika [mailto:emes...@redhat.com<mailto:emes...@redhat.com>] > Sent: Sunday, January 25, 2015 12:58 PM > To: Renchu Mathew > Cc: Martin Perina; users@ovirt.org<mailto:users@ovirt.org> > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi Renchu > > I am looking currently on he case and will respond later on today > after investigating the logs. > If I will not find anything, can I use the details for the host PM > only for status command to try to debug and find the source of the problem ? > > Thanks > > Eli > > - Original Message - > > From: "Renchu Mathew" mailto:ren...@cracknell.com>> > > To: "Martin Perina" mailto:mper...@redhat.com>> > > Cc: users@ovirt.org<mailto:users@ovirt.org>, "Eli Mesika" > > mailto:emes...@redhat.com>> > > Sent: Sunday, January 25, 2015 6:50:00 AM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > I got the below output when I run the command and it works from both > > the nodes. > > > > fence_ipmilan -a 192.168.1.114 -l admin -p admin -o status -v -P > > > > Getting status of IPMI:192.168.1.114...Spawning: '/usr/bin/ipmitool > > -I lanplus -H '192.168.1.114' -U 'admin' -P '[set]' -v chassis power > > status'... > > Chassis power = On > > Done > > > > Thanks > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > > +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 > > 7386484 ABU DHABI | DUBAI > > | LONDON | MUSCAT | DOHA | JEDDAH EMAIL > > ren...@cracknell.com<mailto:ren...@cracknell.com> | WEB > > www.cracknell.com<http://www.cracknell.com> > > > > This email, its content and any files transmitted with it are > > intended solely for the addressee(s) and may be legally privileged > > and/or confidential. If you are not the intended recipient please > > let us know by email reply and delete it from the system. Please > > note that any views or opinions presented in this email do not > > necessarily represent those of the company. Email transmissions > > cannot be guaranteed to be secure or error-free as information could > > be intercepted, corrupted, lost, destroyed, arrive late or > > incomplete, or contain viruses. The company therefore does not > > accept liability for any errors or omiss
Re: [ovirt-users] Power Management config on Ovirt
Hi Eli, I have 2 hypervisor host and both are installed with below iso file. Master data domain is glusterfs which is configured on another server. Is there any latest iso available? ovirt-node-iso-3.5.0.ovirt35.20140912.el6 (Edited). Please find below details. [cid:image001.png@01D038BF.FC58A040] Thanks & Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com] Sent: Sunday, January 25, 2015 4:46 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org; Piotr Kliczewski Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" > To: "Eli Mesika" > Cc: "Martin Perina" , users@ovirt.org > Sent: Sunday, January 25, 2015 11:19:21 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Eli, > > Please use any details. > > Thanks for your support. > > Regards > > Renchu Mathew Hi again Looking in the logs I see that this is a issue we handled when we switched from XML RPC to JSON RPC protocol. Which vdsm is installed in the proxy host ? I am adding Piotr who is on charge of the JSON RPC protocol and who helped me in the past when I got a similar error on PM operations > > -Original Message- > From: Eli Mesika [mailto:emes...@redhat.com] > Sent: Sunday, January 25, 2015 12:58 PM > To: Renchu Mathew > Cc: Martin Perina; users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi Renchu > > I am looking currently on he case and will respond later on today > after investigating the logs. > If I will not find anything, can I use the details for the host PM > only for status command to try to debug and find the source of the problem ? > > Thanks > > Eli > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: users@ovirt.org, "Eli Mesika" > > Sent: Sunday, January 25, 2015 6:50:00 AM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > I got the below output when I run the command and it works from both > > the nodes. > > > > fence_ipmilan -a 192.168.1.114 -l admin -p admin -o status -v -P > > > > Getting status of IPMI:192.168.1.114...Spawning: '/usr/bin/ipmitool > > -I lanplus -H '192.168.1.114' -U 'admin' -P '[set]' -v chassis power > > status'... > > Chassis power = On > > Done > > > > Thanks > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > > +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI > > | LONDON | MUSCAT | DOHA | JEDDAH EMAIL > > ren...@cracknell.com<mailto:ren...@cracknell.com> | WEB > > www.cracknell.com<http://www.cracknell.com> > > > > This email, its content and any files transmitted with it are > > intended solely for the addressee(s) and may be legally privileged > > and/or confidential. If you are not the intended recipient please > > let us know by email reply and delete it from the system. Please > > note that any views or opinions presented in this email do not > > necessarily represent those of the company. Email transmissions > > cannot be guaranteed to be secure or error-free as information could > > be intercepted, corrupted, lost, destroyed, arrive late or > > incomplete, or contain viruses. The company therefore does not > > accept liability for any errors or omissions in the contents of this > > message which arise as a result of email transmission. > > > > > > -Original Message- > > From: Martin Perina [mailto:mper...@redhat.com] > > Sent: Thursday, January 22, 2015 6:12 PM > > To: Renchu Mathew > > Cc: users@ovirt.org; Eli Mesika > > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > > > > > - Original Message - > > > From: "Renchu Mathew" > > > To: "Martin Perina" > > > Cc: users@ovirt.org > > > Sent: Thursday, January 22, 2015 2:39:43 PM > > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > > > Hi Martin, > > > > > > > > > > > > Yes. The vdsm host.log is from node-02. > > > > That's strange, I cannot find any call to fenceNode which should > > appear, if PM status is g
Re: [ovirt-users] Power Management config on Ovirt
Hi Eli, Please use any details. Thanks for your support. Regards Renchu Mathew -Original Message- From: Eli Mesika [mailto:emes...@redhat.com] Sent: Sunday, January 25, 2015 12:58 PM To: Renchu Mathew Cc: Martin Perina; users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi Renchu I am looking currently on he case and will respond later on today after investigating the logs. If I will not find anything, can I use the details for the host PM only for status command to try to debug and find the source of the problem ? Thanks Eli - Original Message - > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org, "Eli Mesika" > Sent: Sunday, January 25, 2015 6:50:00 AM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Martin, > > I got the below output when I run the command and it works from both > the nodes. > > fence_ipmilan -a 192.168.1.114 -l admin -p admin -o status -v -P > > Getting status of IPMI:192.168.1.114...Spawning: '/usr/bin/ipmitool -I > lanplus -H '192.168.1.114' -U 'admin' -P '[set]' -v chassis power status'... > Chassis power = On > Done > > Thanks > > Renchu Mathew | Sr. IT Administrator > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T > +971 4 > 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | > LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB > www.cracknell.com > > This email, its content and any files transmitted with it are intended > solely for the addressee(s) and may be legally privileged and/or > confidential. If you are not the intended recipient please let us know > by email reply and delete it from the system. Please note that any > views or opinions presented in this email do not necessarily represent > those of the company. Email transmissions cannot be guaranteed to be > secure or error-free as information could be intercepted, corrupted, > lost, destroyed, arrive late or incomplete, or contain viruses. The > company therefore does not accept liability for any errors or > omissions in the contents of this message which arise as a result of email > transmission. > > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Thursday, January 22, 2015 6:12 PM > To: Renchu Mathew > Cc: users@ovirt.org; Eli Mesika > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > - Original Message - > > From: "Renchu Mathew" > > To: "Martin Perina" > > Cc: users@ovirt.org > > Sent: Thursday, January 22, 2015 2:39:43 PM > > Subject: RE: [ovirt-users] Power Management config on Ovirt > > > > Hi Martin, > > > > > > > > Yes. The vdsm host.log is from node-02. > > That's strange, I cannot find any call to fenceNode which should > appear, if PM status is gathered. > > Eli, any idea? > > > I have checked the link and it says > > we need to install the sever view management agent & net-snmp on the hosts. > > Do we need to try this? How to install this on node? I used the > > below irmc admin user. > > Not, AFAIK this is needed only from Clustersuite, but not for oVirt. > > But could you please execute this command on node-20 just to confirm > that Fujitsu Primergy is communicating through IPMI: > > fence_ipmilan -a IP -l USER -p PASS -o status -v -P > > where IP is IPMI address (from the log it should be 192.168.1.114) > USER and PASS please set according to you setup. > > Thanks > > Martin > > > > > > > > > [cid:image001.png@01D03669.DD48B4E0] > > > > > > > > Regards > > > > > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 > > 3445417 | F +971 4 3493675 | M +971 50 7386484 > > > > ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH > > > > EMAIL ren...@cracknell.com | WEB www.cracknell.com > > > > > > > > This email, its content and any files transmitted with it are > > intended solely for the addressee(s) and may be legally privileged > > and/or confidential. If you are not the intended recipient please > > let us know by email reply and delete it from the system. Please > > note that any views or opinions presented in this email do not > > necessarily represent those of the company. Email t
Re: [ovirt-users] Power Management config on Ovirt
Hi Martin, I got the below output when I run the command and it works from both the nodes. fence_ipmilan -a 192.168.1.114 -l admin -p admin -o status -v -P Getting status of IPMI:192.168.1.114...Spawning: '/usr/bin/ipmitool -I lanplus -H '192.168.1.114' -U 'admin' -P '[set]' -v chassis power status'... Chassis power = On Done Thanks Renchu Mathew | Sr. IT Administrator CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB www.cracknell.com This email, its content and any files transmitted with it are intended solely for the addressee(s) and may be legally privileged and/or confidential. If you are not the intended recipient please let us know by email reply and delete it from the system. Please note that any views or opinions presented in this email do not necessarily represent those of the company. Email transmissions cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The company therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of email transmission. -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Thursday, January 22, 2015 6:12 PM To: Renchu Mathew Cc: users@ovirt.org; Eli Mesika Subject: Re: [ovirt-users] Power Management config on Ovirt - Original Message - > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org > Sent: Thursday, January 22, 2015 2:39:43 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Martin, > > > > Yes. The vdsm host.log is from node-02. That's strange, I cannot find any call to fenceNode which should appear, if PM status is gathered. Eli, any idea? > I have checked the link and it says > we need to install the sever view management agent & net-snmp on the hosts. > Do we need to try this? How to install this on node? I used the below > irmc admin user. Not, AFAIK this is needed only from Clustersuite, but not for oVirt. But could you please execute this command on node-20 just to confirm that Fujitsu Primergy is communicating through IPMI: fence_ipmilan -a IP -l USER -p PASS -o status -v -P where IP is IPMI address (from the log it should be 192.168.1.114) USER and PASS please set according to you setup. Thanks Martin > > > > [cid:image001.png@01D03669.DD48B4E0] > > > > Regards > > > > Renchu Mathew | Sr. IT Administrator > > > > > > > > CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 > 3445417 | F +971 4 3493675 | M +971 50 7386484 > > ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH > > EMAIL ren...@cracknell.com | WEB www.cracknell.com > > > > This email, its content and any files transmitted with it are intended > solely for the addressee(s) and may be legally privileged and/or > confidential. If you are not the intended recipient please let us know > by email reply and delete it from the system. Please note that any > views or opinions presented in this email do not necessarily represent > those of the company. Email transmissions cannot be guaranteed to be > secure or error-free as information could be intercepted, corrupted, > lost, destroyed, arrive late or incomplete, or contain viruses. The > company therefore does not accept liability for any errors or > omissions in the contents of this message which arise as a result of email > transmission. > > > > > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Thursday, January 22, 2015 4:49 PM > To: Renchu Mathew > Cc: users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > > > Hi, > > > > according to [1] Fujitsu iRMC should be accessible using IPMI. > > Now, I looked at logs, but I cannot see any fenceNode exection in vdsm.log. > According to engine.log, host Node-02.cracknell.com was used as fence proxy. > Are you sure that vdsm.log is from this host? > > > > Thanks > > > > Martin > > > > [1] > https://www.redhat.com/archives/linux-cluster/2010-January/msg00056.ht > ml > > > > > > - Original Message - > > > From: "Renchu Mathew" > > > To: "Martin Perina" > > > Cc: users@ovirt.org > > > Sent: Thursday, January 22, 2015 12:37:51 PM > > > Subject: RE: [ovirt-users] Power Management confi
Re: [ovirt-users] Power Management config on Ovirt
Hi Martin, Yes. The vdsm host.log is from node-02. I have checked the link and it says we need to install the sever view management agent & net-snmp on the hosts. Do we need to try this? How to install this on node? I used the below irmc admin user. [cid:image001.png@01D03669.DD48B4E0] Regards Renchu Mathew | Sr. IT Administrator CRACKNELL DUBAI | P.O. Box 66231 | United Arab Emirates | T +971 4 3445417 | F +971 4 3493675 | M +971 50 7386484 ABU DHABI | DUBAI | LONDON | MUSCAT | DOHA | JEDDAH EMAIL ren...@cracknell.com | WEB www.cracknell.com This email, its content and any files transmitted with it are intended solely for the addressee(s) and may be legally privileged and/or confidential. If you are not the intended recipient please let us know by email reply and delete it from the system. Please note that any views or opinions presented in this email do not necessarily represent those of the company. Email transmissions cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The company therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of email transmission. -Original Message- From: Martin Perina [mailto:mper...@redhat.com] Sent: Thursday, January 22, 2015 4:49 PM To: Renchu Mathew Cc: users@ovirt.org Subject: Re: [ovirt-users] Power Management config on Ovirt Hi, according to [1] Fujitsu iRMC should be accessible using IPMI. Now, I looked at logs, but I cannot see any fenceNode exection in vdsm.log. According to engine.log, host Node-02.cracknell.com was used as fence proxy. Are you sure that vdsm.log is from this host? Thanks Martin [1] https://www.redhat.com/archives/linux-cluster/2010-January/msg00056.html - Original Message - > From: "Renchu Mathew" > To: "Martin Perina" > Cc: users@ovirt.org > Sent: Thursday, January 22, 2015 12:37:51 PM > Subject: RE: [ovirt-users] Power Management config on Ovirt > > Hi Martin, > > Please find attached log files from engine and proxy host. Not sure > whether IPMI or which one can be used. It is Fujitsu iRMC port. I also > tried the rsb but same error. > > Can we configure Power management using Fujitsu iRMC? > > Thanks & Regards > > Renchu Mathew > > -Original Message- > From: Martin Perina [mailto:mper...@redhat.com] > Sent: Thursday, January 22, 2015 1:58 PM > To: Renchu Mathew > Cc: users@ovirt.org > Subject: Re: [ovirt-users] Power Management config on Ovirt > > Hi, > > first of all are you sure that Fujitsu PRIMERGY is accessible using > IPMI protocol? If so, could you provide engine.log and also vdsm.log > from host that was used as fencing proxy? You can find out which host > is used as a proxy either in Events tab after you execute the test or in > engine.log. > > Thanks > > Martin Perina > > - Original Message - > > From: "Renchu Mathew" > > To: users@ovirt.org > > Sent: Thursday, January 22, 2015 7:41:43 AM > > Subject: [ovirt-users] Power Management config on Ovirt > > > > > > > > Dear all, > > > > > > > > I am trying to configure power management on ovirt v3.5 > > (ovirt-node-iso-3.5.0.ovirt35.20140912.el6) and using two Fujitsu > > PRIMERGY > > RX2540 M1 as node hypervisor. I used Fujitsu iRMC port on power > > management configuration. The test gives a message “Test failed, > > argument of type ‘NoneType’ is not iterable” as below . Also tried > > rsb type as per redhat document. > > > > > > > > Fence-agents-3.1.5-35.el6_5.4.x86_64 is there on both nodes. Do we > > need to specify anything in Options? > > > > > > > > Please help me. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Best Regards > > > > > > > > > > > > Renchu Mathew > > > > > > > > > > > > ___ > > Users mailing list > > Users@ovirt.org<mailto:Users@ovirt.org> > > http://lists.ovirt.org/mailman/listinfo/users > > > ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
[ovirt-users] Power Management config on Ovirt
Dear all, I am trying to configure power management on ovirt v3.5 (ovirt-node-iso-3.5.0.ovirt35.20140912.el6) and using two Fujitsu PRIMERGY RX2540 M1 as node hypervisor. I used Fujitsu iRMC port on power management configuration. The test gives a message "Test failed, argument of type 'NoneType' is not iterable" as below. Also tried rsb type as per redhat document. Fence-agents-3.1.5-35.el6_5.4.x86_64 is there on both nodes. Do we need to specify anything in Options? Please help me. [Description: cid:image001.png@01D032FC.C8A4DF20] [Description: cid:image002.png@01D034BF.32500FD0] Best Regards Renchu Mathew ___ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
