Same thing happened to me, upgrading from 4.5.2 to 4.8 and I had to restart network with clean-up option checked for each network which rebuilds the router VM. If your SSVM works And if you can test it then I don’t see any problem rebuilding network(router vm) with new template which took about 30 minutes for about 20 networks in total...may not be an option for large installations...how many routers you are running?
-----Original Message----- From: Cloud List [mailto:cloud-l...@sg.or.id] Sent: Wednesday, June 8, 2016 12:23 AM To: users@cloudstack.apache.org; d...@cloudstack.apache.org Subject: Re: Upgrading from ACS 4.2 to 4.8 - VR upgrade issue Dear all, Could it be the VR is not able to use the new systemVM template because we registered the template as Routing = NO? Will it resolve the problem if during upgrade, we register the template as Routing = YES? Based on the upgrade instruction at http://cloudstack-release-notes.readthedocs.io/ja/stable/upgrade/upgrade-4.2.html , it says that routing option should be "no", is this correct? Since the systemVM template will also be used by VR? Any advice is appreciated. Looking forward to your reply, thank you. Cheers. -ip- On Tue, Jun 7, 2016 at 10:22 AM, Cloud List <cloud-l...@sg.or.id> wrote: > Hi Kirk, > > Thanks for your reply. > > Here's the status of the CPVM and SSVM in the UI: > > Name v-2-VM > ID fc6d23bf-58ac-4641-af9b-506d1dc46df3 > State Running > Type Console Proxy VM > Zone sc-test-cs42-zone > Public IP Address 192.168.0.162 > Private IP Address 192.168.0.160 > Link Local IP Address 169.254.3.147 > Host sc-test-kvm-03 > Gateway 192.168.0.1 > Created 31 May 2016 15:24:20 > Active Sessions 0 > > Name s-29-VM > ID d26f54f8-1c1f-4f3d-9862-fc0405159620 > State Running > Type Secondary Storage VM > Zone sc-test-cs42-zone > Public IP Address 192.168.0.161 > Private IP Address 192.168.0.151 > Link Local IP Address 169.254.1.207 > Host sc-test-kvm-03 > Gateway 192.168.0.1 > Created 06 Jun 2016 00:31:04 > Active Sessions > > Both system VMs seem to be on "Running" state but I can't verify what > template version they are using, whether they are already on 4.6/4.8 > template or still using 4.2 template, are you able to advise how can I > tell? The SSVM creation date seem to be after we performed the upgrade > but the CPVM created date is before the upgrade. > > How to ensure that all systemVMs using the new version template? Shall > I destroy the system VMs so new ones will be created, or shall I > re-run the /usr/bin/cloudstack-sysvmadm again in full? > > Looking forward to your reply, thank you. > > Cheers. > > -ip- > > > > On Tue, Jun 7, 2016 at 7:22 AM, Kirk Kosinski > <kirk.kosin...@shapeblue.com > > wrote: > >> Hi, what is the status of the CPVM and SSVM in the UI? Were they >> really upgraded successfully? >> >> Also, are there any errors in the management-server.log? >> >> kirk.kosin...@shapeblue.com >> www.shapeblue.com >> 53 Chandos Place, Covent Garden, London WC2N 4HS @shapeblue >> >> >> >> -----Original Message----- >> From: Cloud List [mailto:cloud-l...@sg.or.id] >> Sent: Monday, June 06, 2016 9:51 AM >> To: users@cloudstack.apache.org; d...@cloudstack.apache.org >> Subject: Upgrading from ACS 4.2 to 4.8 - VR upgrade issue >> >> Hi, >> >> I tried upgrading our CloudStack 4.2.0 on our test environment to 4.8.0, >> and I am having issues on VR upgrade. I followed the upgrade instructions >> at below URL: >> >> >> http://cloudstack-release-notes.readthedocs.io/ja/stable/upgrade/upgrade-4.2.html >> >> On the last step to restart all the system VMs and VRs, it seems that the >> cloudstack-sysvmadm script managed to restart the system VMs (SSVM and >> CPVM) but not the VR. >> >> # nohup cloudstack-sysvmadm -d IPaddress -u cloud -p password -a > >> sysvm.log 2>&1 & >> >> Here's the output of the sysvm.log file: >> >> ==== >> nohup: ignoring input >> /usr/bin/cloudstack-sysvmadm: line 21: /etc/rc.d/init.d/functions: No >> such file or directory >> >> Stopping and starting 1 secondary storage vm(s)... >> Done stopping and starting secondary storage vm(s) >> >> Stopping and starting 1 console proxy vm(s)... >> Done stopping and starting console proxy vm(s) . >> >> Stopping and starting 1 running routing vm(s)... >> *ERROR: Failed to restart domainRouter with id 5* >> >> Done restarting router(s). >> ==== >> >> Output of the cloud.log file: >> >> ==== >> [2016.06.05-22.28.40] Stopping and starting 1 secondary storage vm(s)... >> [2016.06.05-22.28.40] INFO: Stopping secondary storage vm with id 3 >> [2016.06.05-22.28.46] INFO: Starting secondary storage vm with id 3 >> [2016.06.05-23.29.00] ERROR: Failed to startSystemVm id=3; jobId is >> 121c5d89-745a-4e8e-9f96-0b348ef42abf >> [2016.06.05-23.29.00] ERROR: Failed to start secondary storage vm with id >> 3 [2016.06.05-23.29.00] ERROR: Failed to start secondary storage vm with id >> 3 [2016.06.05-23.29.00] Done stopping and starting secondary storage vm(s). >> [2016.06.05-23.29.00] Stopping and starting 1 console proxy vm(s)... >> [2016.06.05-23.29.00] INFO: Stopping console proxy with id 2 >> [2016.06.05-23.29.06] INFO: Starting console proxy vm with id 2 >> [2016.06.05-23.30.49] Done stopping and starting console proxy vm(s) . >> [2016.06.05-23.30.49] Stopping and starting 1 running routing vm(s)... >> [2016.06.05-23.30.49] INFO: Restarting router with id 5 >> [2016.06.06-00.30.59] ERROR: Failed to restart domainRouter with id 5; >> jobId 4a3f681e-c4a6-4856-9d42-0d3b79bffdae >> [2016.06.06-00.31.04] Done restarting router(s). >> ==== >> >> Initially I noted that the VR remains in 4.2.0 version (instead of the >> expected 4.8.0 version), tried to destroy the router so it'll be recreated >> but it doesn't get created successfully. >> >> Tried to run the cloudstack-sysvmadm manually with the -r option to >> restart only the VR, and the problem still persists. >> >> At the moment, the version of the VR is "UNKNOWN" and the "Require >> Upgrade" >> state is "Yes". >> >> Any advice how to make the VR back up and running? >> >> Looking forward to your reply, thank you. >> >> Cheers. >> >> -ip- >> > >
