[ovirt-users] Re: Re : [ovirt-users] Re: Install hosted-engine - Task Get local VM IP failed
At this time the engine (and the cluster) is up. No problem after activating gluster and creating the volumes, then finish the install in the screen session. So... Le Vendredi, Juin 29, 2018 12:32 CEST, "fsoyer" a écrit: Hi, I must say it : I'm -totally- lost. To try to find a reason to this error, I've re-installed the first host from scratch - CentOS 7.5-1804, ovirt 4.2.3-1, gluster 3.12.9. The first attemp was made with only em1 declared. Result = SUCCESS, the install pass "Get local VM IP", then through "Wait for the host to be up" without difficulty and wait at "Please specify the storage...". At this time I even notice that I've forgot to stop/disable NetworkManager, that had no impact ! So : I re-install the host from scratch (yes, sometimes I'm a fool) to be absolutly sure that there is no problem coming from the preceding install. Now I declare em1 (10.0.0.230) and em2 (10.0.0.229, without gateway nor DNS, for futur vmnetwork). NetworkManager off and disabled. Result = SUCCESS... Oo OK : Re-install host !! Now I declare, as I did some days ago, em1, em2 and bond0(em3+em4 with IP 192.168.0.30). Result : SUCCESS !!! Oo So I'm unable to say what append tuesday. Actually I see only two differences : - gluster is not active (I don't configure it to go faster) - the version of ovirt (ovirt-release, ovirt-host, appliance...) has sligthly changed. I've no more time for another attempt re-installing the host(s) with gluster activated, I must now go on as I need an operational system for other tasks with VMs this afternoon. So I leave the first host waiting for the end of install in a screen, I re-install the 2 other hosts and activate gluster and volumes on the 3 nodes. Then I'll end the install on the gluster volume. I'll tell you if this works finally, but I hope so ! however, I'm in doubt with this problem. I have no explanation of what append tuesday, this is really annoying... Maybe have you the ability to test on a same configuration (3 hosts with 2 nics on the same network for ovirtmgmt and a futur vmnetwork, and gluster on a separate network) to try to understand ? Thank you for the time spent. Frank PS : to answer to your question : yes, tuesday I ran ovirt-hosted-engine-cleanup between each attempt. Le Jeudi, Juin 28, 2018 16:26 CEST, Simone Tiraboschi a écrit: On Wed, Jun 27, 2018 at 5:48 PM fso...@systea.fr wrote:Hi again, In fact, the hour in file is exactly 2hours before, I guess a timezone problem (in the process of install ?), as the file itself is correctly timed at 11:17am (correct hour here in France). So the messages are synchrone. Yes, sorry, fault of mine.From the logs I don't see anything strange. Can you please try again on your environment and connect to the bootstrap VM via virsh console or VNC to check what's happening there? Did you also run ovirt-hosted-engine-cleanup between one attempt and the next? Message original Objet : Re: [ovirt-users] Re: Install hosted-engine - Task Get local VM IP failed De : Simone Tiraboschi À : fso...@systea.fr Cc : users Hi,HostedEngineLocal was started at 2018-06-26 09:17:26 but /var/log/messages starts only at Jun 26 11:02:32.Can you please reattach it fro the relevant time frame? On Wed, Jun 27, 2018 at 10:54 AM fsoyer wrote:Hi Simone, here are the revelant part of messages and the engine install log (there were only this file in /var/log/libvirt/qemu) . Thanks for your time. Frank Le Mardi, Juin 26, 2018 11:43 CEST, Simone Tiraboschi a écrit: On Tue, Jun 26, 2018 at 11:39 AM fsoyer wrote:Well, unfortunatly, it was a "false-positive". This morning I tried again, with the idea that at one moment the deploy will ask for the final destination for the engine, I will restart bond0+gluster+volume engine at thos moment. Re-launching the deploy on the second "fresh" host (the first one with all errors yesterday let it in a doutful state) with em2 and gluster+bond0 off : # ip a 1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: em1: mtu 1500 qdisc mq state UP group default qlen 1000 link/ether e0:db:55:15:f0:f0 brd ff:ff:ff:ff:ff:ff inet 10.0.0.227/8 brd 10.255.255.255 scope global em1 valid_lft forever preferred_lft forever inet6 fe80::e2db:55ff:fe15:f0f0/64 scope link valid_lft forever preferred_lft forever 3: em2: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f1 brd ff:ff:ff:ff:ff:ff 4: em3: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f2 brd ff:ff:ff:ff:ff:ff 5: em4: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f3 brd ff:ff:ff:ff:ff:ff 6: bond0: mtu 9000 qdisc noqueue state DOWN group
[ovirt-users] Re: Re : [ovirt-users] Re: Install hosted-engine - Task Get local VM IP failed
Hi, I must say it : I'm -totally- lost. To try to find a reason to this error, I've re-installed the first host from scratch - CentOS 7.5-1804, ovirt 4.2.3-1, gluster 3.12.9. The first attemp was made with only em1 declared. Result = SUCCESS, the install pass "Get local VM IP", then through "Wait for the host to be up" without difficulty and wait at "Please specify the storage...". At this time I even notice that I've forgot to stop/disable NetworkManager, that had no impact ! So : I re-install the host from scratch (yes, sometimes I'm a fool) to be absolutly sure that there is no problem coming from the preceding install. Now I declare em1 (10.0.0.230) and em2 (10.0.0.229, without gateway nor DNS, for futur vmnetwork). NetworkManager off and disabled. Result = SUCCESS... Oo OK : Re-install host !! Now I declare, as I did some days ago, em1, em2 and bond0(em3+em4 with IP 192.168.0.30). Result : SUCCESS !!! Oo So I'm unable to say what append tuesday. Actually I see only two differences : - gluster is not active (I don't configure it to go faster) - the version of ovirt (ovirt-release, ovirt-host, appliance...) has sligthly changed. I've no more time for another attempt re-installing the host(s) with gluster activated, I must now go on as I need an operational system for other tasks with VMs this afternoon. So I leave the first host waiting for the end of install in a screen, I re-install the 2 other hosts and activate gluster and volumes on the 3 nodes. Then I'll end the install on the gluster volume. I'll tell you if this works finally, but I hope so ! however, I'm in doubt with this problem. I have no explanation of what append tuesday, this is really annoying... Maybe have you the ability to test on a same configuration (3 hosts with 2 nics on the same network for ovirtmgmt and a futur vmnetwork, and gluster on a separate network) to try to understand ? Thank you for the time spent. Frank PS : to answer to your question : yes, tuesday I ran ovirt-hosted-engine-cleanup between each attempt. Le Jeudi, Juin 28, 2018 16:26 CEST, Simone Tiraboschi a écrit: On Wed, Jun 27, 2018 at 5:48 PM fso...@systea.fr wrote:Hi again, In fact, the hour in file is exactly 2hours before, I guess a timezone problem (in the process of install ?), as the file itself is correctly timed at 11:17am (correct hour here in France). So the messages are synchrone. Yes, sorry, fault of mine.From the logs I don't see anything strange. Can you please try again on your environment and connect to the bootstrap VM via virsh console or VNC to check what's happening there? Did you also run ovirt-hosted-engine-cleanup between one attempt and the next? Message original Objet : Re: [ovirt-users] Re: Install hosted-engine - Task Get local VM IP failed De : Simone Tiraboschi À : fso...@systea.fr Cc : users Hi,HostedEngineLocal was started at 2018-06-26 09:17:26 but /var/log/messages starts only at Jun 26 11:02:32.Can you please reattach it fro the relevant time frame? On Wed, Jun 27, 2018 at 10:54 AM fsoyer wrote:Hi Simone, here are the revelant part of messages and the engine install log (there were only this file in /var/log/libvirt/qemu) . Thanks for your time. Frank Le Mardi, Juin 26, 2018 11:43 CEST, Simone Tiraboschi a écrit: On Tue, Jun 26, 2018 at 11:39 AM fsoyer wrote:Well, unfortunatly, it was a "false-positive". This morning I tried again, with the idea that at one moment the deploy will ask for the final destination for the engine, I will restart bond0+gluster+volume engine at thos moment. Re-launching the deploy on the second "fresh" host (the first one with all errors yesterday let it in a doutful state) with em2 and gluster+bond0 off : # ip a 1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: em1: mtu 1500 qdisc mq state UP group default qlen 1000 link/ether e0:db:55:15:f0:f0 brd ff:ff:ff:ff:ff:ff inet 10.0.0.227/8 brd 10.255.255.255 scope global em1 valid_lft forever preferred_lft forever inet6 fe80::e2db:55ff:fe15:f0f0/64 scope link valid_lft forever preferred_lft forever 3: em2: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f1 brd ff:ff:ff:ff:ff:ff 4: em3: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f2 brd ff:ff:ff:ff:ff:ff 5: em4: mtu 1500 qdisc mq state DOWN group default qlen 1000 link/ether e0:db:55:15:f0:f3 brd ff:ff:ff:ff:ff:ff 6: bond0: mtu 9000 qdisc noqueue state DOWN group default qlen 1000 link/ether 3a:ab:a2:f2:38:5c brd ff:ff:ff:ff:ff:ff # ip r default via 10.0.1.254 dev em1 10.0.0.0/8 dev em1 proto kernel scope link src 10.0.0.227 169.254.0.0/16 dev em1 scope link metric 1002 ... does
