Hi, I'm replying because I'm already interested in this topic. In fact, we'd like to monitor VR and collect logs with rsyslog. So, is it possible to modify the template image to configure an rsyslog and snmp to allow monitoring of all new deployed VR?
Thank you. On Wed, Oct 4, 2017 at 5:10 PM, Dag Sonstebo <dag.sonst...@shapeblue.com> wrote: > Hi Alexander, > > As you point out the attack surface of the VRs have been minimised and the > system VMs overall hardened. > You have two issues with updating the software on the VRs – an update has > the potential to break VR services (or agent services on SSVM/CPVM) as you > already mentioned, in addition any updates will obviously be lost the next > time you restart a network with cleanup. You would also need to somehow > automate the updates on new VRs. > > In general I would think most CloudStack users will wait for new system VM > templates to be released – but interested in hearing other thoughts on this. > > Regards, > Dag Sonstebo > Cloud Architect > ShapeBlue > > On 04/10/2017, 15:35, "Stock, Alexander" <alexander.st...@bitgroup.de> > wrote: > > Hi all, > > at the moment we try to improve our monitoring of the virtual routers > in our environment. > For this we also monitor the update status of the machines and could > see that there are some updates pending (OS Updates not Template Updates) . > > So what would like to know is if you have some experience in pathing > the virtual routers and if you have an update strategy for this (redundant > router,etc..). > I am in worry that some updates could crash the services which are > responsible for communication with the cloudstack controller. > I am also not sure if the attack surface is too small to don't worry > about patching. > > Any ideas or comments are welcome. > Thank you. > Alexander > > > > > dag.sonst...@shapeblue.com > www.shapeblue.com > 53 Chandos Place, Covent Garden, London WC2N 4HSUK > @shapeblue > > > >
