cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From benoit lair <kurushi4...@gmail.com>
Subject Re: Production recovery procedure for VPC Virtual router into advanced zone - CS 4.0.0
Date Wed, 13 Feb 2013 17:53:22 GMT
Hello Ahmad,

I shuttted down the vpc vr during an update of acls sent by the mgmt
server, so mgmt server sais me the vpc vr has 3 rules, in reality, an
ipatbles -L give me only 2 acls.

More simply, when i create a new vpc, create 3 tiers netoworks, deployed a
vm into each tier network, all seems to be ok. Problems begin when i simply
reboot the vpc vr.

After a reboot, i have no more the interfaces of my tier (vlans) interfaces
mounted.

Even if i reboot each tier network, i can't recover access to my vlans so
no access to my vms.

In fact, i only have the control interface (1698.254.x.x) the public
interface (in my case 10.14.10.36) and the localhost one.

My master cidr is 172.28.0.0/16, my tiers are 172.28.1/24, 172.28.2.1/24and
172.28.3.1/24, before reboot i got 3 interfaces : 172.28.1.1, 172.28.2.1
and 172.28.3.1 after reboot no more interfaces...

I have already tried to destroy the vpc vr, triggered the creation of a new
vm to force cs to detect that vpc vr is missing and so relaunch the
installation of the vpc vr.

Even in this case, the vpc is redeployed with only mgmt , public and
loopback interface, no vlan interfaces.

Have you already encountered this problem ?

2013/2/11 Anthony Xu <Xuefei.Xu@citrix.com>

> You can try restart network with clean checked in UI, it will try to
> destroy Virtual router on this network and create a new one.
>
>
> Anthony
>
> > -----Original Message-----
> > From: Ahmad Emneina [mailto:aemneina@gmail.com]
> > Sent: Monday, February 11, 2013 10:48 AM
> > To: Cloudstack users
> > Subject: Re: Production recovery procedure for VPC Virtual router into
> > advanced zone - CS 4.0.0
> >
> > how did you crash the vpc router?
> >
> > did spinning up a vm, or restarting a vm, in the affected network
> > restore
> > the router? (this is probably the easiest way to correct the issue, if
> > cloudstack understands the router is gone.)
> >
> > I would only spin up a vm after a destroy router is called, insuring
> > cloudstack understands that router needs to be recreated.
> >
> >
> > On Mon, Feb 11, 2013 at 8:17 AM, benoit lair <kurushi4000@gmail.com>
> > wrote:
> >
> > > Hello,
> > >
> > > I am currently working on a model of preproduction  of cs 4.0.0 on a
> > centos
> > > 6.3 server management.
> > >
> > > I have deployed my zones, networks, primary iscsi storages onto my
> > array,
> > > configured my clusters with xcp 1.1.
> > >
> > > Now i'm looking for deploying the VPC features.
> > >
> > > I have configured my vpc, defined my CIDR, created 3 tiers networks,
> > > deployed 1 vm onto each tier network.
> > >
> > > All worked very great.
> > >
> > > Now i've tried to force crash of the vpc virtual router in order to
> > > troubleshooting the recovery of production of the vpc.
> > >
> > > I got some problems in certain cases, after crashing my vpc router,
> > my tier
> > > networks didn't mount onto the virtual router.
> > >
> > > This caused lossing the avalibility to recover production of my
> > tiered
> > > networking vms.
> > >
> > > So i'm looking for differents recovery procedures in order to recover
> > a
> > > full fonctionnal vpc cloud.
> > >
> > > If somebody got best practices in order to recover a vpc router, i
> > would
> > > appreciate any help.
> > >
> > >
> > > Thanks.
> > >
> > > Benoit Lair.
> > >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message