----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/17736/#review33792 -----------------------------------------------------------
server/src/com/cloud/vm/UserVmManagerImpl.java <https://reviews.apache.org/r/17736/#comment63409> The nested for loop of select queries is a scalability issue. Please refactor to join the nics and routers to create one list to traverse. server/src/com/cloud/vm/UserVmManagerImpl.java <https://reviews.apache.org/r/17736/#comment63410> The content of this log message makes no sense to an operator. Why do they care that a router has been found? Please change to something like the following: "Restarting virtual router <name> (<id>) due to guest VM <guest vm name> restart." This log message feels more like an INFO not a WARN. - John Burwell On Feb. 5, 2014, 3:46 a.m., Saksham Srivastava wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/17736/ > ----------------------------------------------------------- > > (Updated Feb. 5, 2014, 3:46 a.m.) > > > Review request for cloudstack and Murali Reddy. > > > Bugs: CLOUDSTACK-5999 > https://issues.apache.org/jira/browse/CLOUDSTACK-5999 > > > Repository: cloudstack-git > > > Description > ------- > > When a guest is rebooted from CloudStack, if the virtual router managing the > guest network of that guest is down, CloudStack will not start the virtual > router. > However the router is started in case the guest vm is stopped and then > started. > To mantain similarity between the 2 process it is necessary to start the VR > in case it is not running. > The fix will address the same. > > > Diffs > ----- > > server/src/com/cloud/vm/UserVmManagerImpl.java 81be19f > > Diff: https://reviews.apache.org/r/17736/diff/ > > > Testing > ------- > > Testing: > 1) vm in a single guest network : > vm Reboot : If the VR is stopped: VR is first started and then > the VM reboots. > vm Reboot : If the VR is running, VM reboots as it used to. > vm Stop/Start continue to work the same. > > 2) vm having nics in multi networks : > vm Reboot : If VR in any/all network is stopped: VRs are first > started and then the VM reboots. > vm Reboot : If the VRs are running, VM reboots as it used to. > vm Stop/Start continue to work the same. > > 3) vpc : > Tested the above scenarios for vpc also, works fine. > > Patch applies cleanly. > > > Thanks, > > Saksham Srivastava > >
