The systemvm iso file is copied only when a systemvm or router vm is to be started on
a host. The file gets copied to the secondary storage. The mount point used is the one
that has permissions for regular user to mount a share.
CLOUDSTACK-5275: The failure was because a secondary storage wasn't available when the
host was added. When a setup is done through wizard the hosts get added before the
secondary storage. CS was tying to copy yhe systemvm iso to secondary and it used to
fail if it wasn't available. Made a change to copy the iso only when a systemvm is
being started on a host.
CLOUDSTACK-5202: Made changes to clean up mount points on stop and start.
All the three are related fixes; so putting a fix in one commit.
Fixed few RAT issues reported for new additions of files to repo.
Signed-off-by: Santhosh Edukulla <Santhosh.Edukulla@citrix.com>
Signed-off-by: Abhinandan Prateek <aprateek@apache.org>
service and not used for LB
Fix adds a boolean flag to addNetscalerLoadBalancer api, which
will mark added NetScaler for exclusive GSLB service. A netscaler marked
as exclusive gslb service provider is not picked for any guest network's
lb provider.
Conflicts:
engine/schema/src/com/cloud/network/dao/ExternalLoadBalancerDeviceVO.java
plugins/network-elements/f5/src/com/cloud/network/element/F5ExternalLoadBalancerElement.java
plugins/network-elements/netscaler/src/com/cloud/api/commands/AddNetscalerLoadBalancerCmd.java
plugins/network-elements/netscaler/src/com/cloud/api/response/NetscalerLoadBalancerResponse.java
plugins/network-elements/netscaler/src/com/cloud/network/element/NetscalerElement.java
server/src/com/cloud/network/ExternalLoadBalancerDeviceManager.java
server/src/com/cloud/network/ExternalLoadBalancerDeviceManagerImpl.java
setup/db/db/schema-421to430.sql
Added a temporary time sleep fix post a restart command. Applies to 4.2,master,4.3
Note: Has to clean other things as well and will do more cleaner way of handling things.
From 4.2, network ACLs are no longer associated with a single network.
Multiple tiers within a VPC can share the same ACL.
Due to this, ACLs should not be cleaned up during network GC.
1. Egress default policy rules is send to the firewall provider. It is up to the
provider to configure the rules.
2. The default policy rules are send for both allow and deny default policy.
3. On network shutdown rules for delete are send.
4. For VR and SRX, by default deny the traffic. So no default rule to deny traffic is required.
scaling up vms was not considering parameter cluster.(memory/cpu).allocated.capacity.disablethreshold. Fixed it
Also added overprovisioning factor retrieval at the cluster level for host capacity check
scaling up vms was not considering parameter cluster.(memory/cpu).allocated.capacity.disablethreshold. Fixed it
Also added overprovisioning factor retrieval at the cluster level for host capacity check
SrikanteswaraRao Talluri
Devdeep Singh
Rajani Karuturi
Donal Lafferty
Abhinandan Prateek
Syed Ahmed
Nitin Mehta
Min Chen
Kelven Yang
Jessica Wang
Mike Tutkowski
Harikrishna Patnala
Alena Prokharchyk
Santhosh Edukulla
Murali Reddy
Girish Shilamkar
Jayapal
Bharat Kumar
Damodar Reddy
Sheng Yang
Ashutosh K
Anthony Xu
Will Stevens
Frank.Zhang