FR12 introduce basic support for comma-separated list of management
servers for agents, while makes an explicit LB unnecessary.
On the agent side, the 'host' is saved as:
<comma separated addresses>@<algorithm name>.
Where the algorithm name is the name of the lb algorithm. The
This FR introduces two new global settings:
- indirect.agent.lb.algorithm: The algorithm for the indirect agent LB.
- indirect.agent.lb.check.interval: The preferred host check interval
for the agent's background task that checks and switches to agent's
preferred host.
Any changes to the above two global settings and the 'host' setting does
not require restarting of the management server(s).
The indirect.agent.lb.algorithm supports following algorithm options:
- static: use the list as provided.
- roundrobin: evenly spreads hosts across management servers.
- shuffle: (pseudo) randomly sorts the list (not recommended for production).
From the agent's perspective, the first address in the propagated list
will be considered the preferred host. A new background task can be
activated by configuring the indirect.agent.lb.check.interval which is
a cluster level global setting from CloudStack or admins can override
this by configuring the 'host.lb.check.interval' in the host's
agent.properties file.
Comma-separated management server list is propagated to agents on
following cases:
- Addition of a host (including ssvm, cpvm systevms).
- Connection or reconnection by the agents to a management server.
- After admin changes the 'host' and/or the
'indirect.agent.lb.algorithm' global settings.
First the agent connects to the management server and sends its current
management server list, which is compared by the management server and
in case of failure a new/update list is sent for the agent to persist.
Every time agent gets a ms-host list and the algorithm, the host specific
background check interval is also sent and it dynamically reconfigures
the background task without need to restart agents.
The 'static' and 'roundrobin' algorithms, strictly checks for the order
as expected by them, however, the 'shuffle' algorithm just checks for
content and not the order of the comma separate ms host addresses.
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
From b3f18e7d74, the zone level systemvm local
storage setting never worked as it needed to be moved to config depot.
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
1. getRouters() doesn't handle RestartNetwork with cleanup=true for basic zone,
because pod wouldn't be specific at the time.
2. The regression caused by the following fix. The variable "routers" was
overrided with some local values, result in only one of the routers in multiple
pods would return, thus only one router would be started.
commit 6dd5c3fd42
Author: Rohit Yadav <bhaisaab@apache.org>
Date: Thu Oct 11 18:30:00 2012 +0530
CLOUDSTACK-70: Improve restart network behaviour for basic network
(cherry picked from commit aaeadc5c44)
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
On default iptables rules are updated to add ACCEPT egress traffic.
If the network egress default policy is false, CS remove ACCEPT and adds the DROP rule which
is egress default rule when there are no other egress rules.
If the CS network egress default policy is true, CS won't configure any default rule for egress because
router already came up to accept egress traffic. If there are already egress rules for network then the
egress rules get applied on VR.
For isolated network with out firewall service, VR default allows egress traffic (guestnetwork --> public network)
The last commit 513adab51b didn't fully fix it. Scenario #1 was not handled, only #2 was handled.
Fixed#1 as part of this commit.
1. If VM is in stopped state in CS due to 'PowerMissing' report from old host (hostId is null) and then there is a 'PowerOn' report from new host
2. If VM is in running state in CS and there is a 'PowerOn' report from new host
The previous fix tried to access StatsCollector from UsageManagerImpl
which is not possible due to dependency cycle.
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
(cherry picked from commit 1e0880cbab)
Signed-off-by: Rohit Yadav <rohit.yadav@shapeblue.com>
Conflicts:
server/src/com/cloud/network/router/VirtualNetworkApplianceManagerImpl.java
In some network environment, 1*3 seconds by default make RvR setup too
sensitive. A configurable parameter would be better for fitting different
network environments.
Corrects problems from previous attempt. Fixes based on help comments from
the community and conflict resolution
Signed-off-by: Daan Hoogland <daan@onecht.net>
Rohit Yadav
Remi Bergsma
Daan Hoogland
Koushik Das
Sheng Yang
Damodar
Kishan Kavala
Jayapal
Ilia Shakitko
David Nalley
Frank Zhang
Anthony Xu
Fred Clift
Hugo Trippaers
Daan Hoogland
Funs
Marcus Sorensen
Min Chen
Rajani Karuturi
Harikrishna Patnala
Kelven Yang
Antonio Fornie
Alena Prokharchyk