cloudstack-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From benoit lair <kurushi4...@gmail.com>
Subject Re: [ACS43][MGMT Server][Load Balancing]
Date Mon, 03 Nov 2014 16:55:01 GMT
Hi Rohit,

I have two managements (java) servers :

First node called node01     : 192.168.0.10
Second node called node02 : 192.168.0.11

The database is 192.168.0.200 (not vip of a netscaler, but a fully
active/active cluster with a ha-ip), the mysql server is well joining from
the 192.168.0.10 or from the 192.168.0.11, all of two have the rights
privileges onto the mysql-server.
For information, my mysql server cluster 192.168.0.200 is replicated on two
others slaves.

I want to contact my mgmt server with 192.168.0.100 (vip on netscaler vpx)
I have this config :
------------------------------------------------------------------------------------------------------------------------
192.168.0.100 port type HTTP 8080 : with LB Methode Least connection and
persistence with sourceip, 2 mins of timeout, netmask of 255.255.255.255
------------------------------------------------------------------------------------------------------------------------
192.168.0.100 port type TCP 8250 : with LB Methode Roun-robin and
persistence with sourceip, 5 mins of timeout, netmask of 255.255.255.255
------------------------------------------------------------------------------------------------------------------------
192.168.0.100 port type TCP 8096 : with LB Methode Least connections and
persistence with sourceip, 5 mins of timeout, netmask of 255.255.255.255
------------------------------------------------------------------------------------------------------------------------
The rule with the port 8096 does not serve for the moment.

I have deployed the node01 with :
cloudstack-setup-databases user:password@192.168.0.200
--deploy-as=root:password -e file -m mypassphrase -k mypassphrase -i
192.168.0.10

I have deployed the node02 with :
cloudstack-setup-databases user:password@192.168.0.200 -e file -m
mypassphrase -k mypassphrase -i 192.168.0.11

I executed on node01 the command :
/usr/share/cloudstack-common/scripts/storage/secondary/cloud-install-sys-tmplt
-m /mnt/secondary -u
http://cloudstack.apt-get.eu/systemvm/systemvm64template-2014-01-14-master-xen.vhd.bz2
-h xenserver -s mypassphase -F

I changed the global host parameter accessing with
http://192.168.0.100:8080/client with the value : 192.168.0.100
I changed the global parameter agent.lb to true with the same way.

I restarted my two management servers, using
/etc/init.d/cloudstack-management restart, so all of twos seems to well
communicate mutually (grepping the logs entries with HeartBeat tag says me
the two servers are seing them)

Also i created a zone with local storage enabled, a pod, a cluster, a host
and some public networks, without activating the zone (for local storage
trick). Terminated the wizard of creation of the zone.
I went to primary storage of my zone, adding a nfs primary storage server,
activating the zone.

I see my 2 systems vms spawning :
 the seconday storage vm is ok, state running and agent running
 the console proxy vm is not all ok, state running but the agent as a "-"
status.

If i try to reboot the ssvm, it is ok, but if i try to reboot the cpvm, the
webui says me it is ok, but if i look at xencenter i do not see my cpvm
rebooting.
Here are my mgmt logs entries about this reboot order :

2014-11-03 17:49:54,187 DEBUG [c.c.a.m.ClusteredAgentAttache]
(StatsCollector-2:ctx-9c4fb510) Seq 1-1490813094: Unable to forward null
2014-11-03 17:49:54,187 DEBUG [c.c.s.StorageManagerImpl]
(StatsCollector-2:ctx-9c4fb510) Unable to send storage pool command to
Pool[2|NetworkFilesystem] via 1
com.cloud.exception.AgentUnavailableException: Resource [Host:1] is
unreachable: Host 1: Unable to reach the peer that the agent is connected
at
com.cloud.agent.manager.ClusteredAgentAttache.send(ClusteredAgentAttache.java:220)
at com.cloud.agent.manager.AgentAttache.send(AgentAttache.java:398)
at com.cloud.agent.manager.AgentManagerImpl.send(AgentManagerImpl.java:394)
at com.cloud.agent.manager.AgentManagerImpl.send(AgentManagerImpl.java:347)
at
com.cloud.storage.StorageManagerImpl.sendToPool(StorageManagerImpl.java:964)

If i connect to http://192.168.0.11:8080/client and go to
infrastructure>systems vms, is the 2 systems vms with the same states, but
here if i try to reboot the cpvm it is working.

I think there is a problem with my 2 management nodes installation (maybe
the agent does not communicate well between the two mgmt servers).

How can i troubleshoot this ?

Thanks for your lights Rohit.







2014-11-03 14:58 GMT+01:00 Rohit Yadav <rohit.yadav@shapeblue.com>:

> Hi Benoit,
>
> You got the db deployment right, not sure why it’s failing for you. It
> would require some debugging of your deployment environment.
>
> The only advice I have for you is that - you make sure the cloud user is
> able to log in to the mysql server from all the mgmt server IPs and
> netscaler VIPs (not sure if you’re load-balancing for port 3306/mysql).
>
> > On 03-Nov-2014, at 6:26 pm, benoit lair <kurushi4000@gmail.com> wrote:
> >
> > Any advice ?
> >
> > the 192.168.0.100 is the VIP of the netscaler (so not configured onto any
> > of the 2 mgmt servers).
> >
> > How should i setup and deploy database with the first node 192.168.0.10 ?
> > How for the second 192.168.0.11 ?
> >
> > Thanks for your help and advices.
> >
> > Regards, Benoit.
> >
> > 2014-10-31 16:27 GMT+01:00 benoit lair <kurushi4000@gmail.com>:
> >
> >> Hi,
> >>
> >> i retried an install of LB Mgmt servers :
> >>
> >> On the first mgmt server 192.168.0.10 i tried a :
> >> cloudstack-setup-databases cloud:password@192.168.0.200
> >> <cloud%3Apassword@192.168.0.10> --deploy-as=root:mypassroot -e file -m
> >> mypassphrase -k mypassphrase -i 192.168.0.100
> >>
> >> 192.168.0.100 is the vip of the netscaler
> >> the 192.168.0.10 if the first mgmt server node (192.168.0.11 is the
> second)
> >>
> >> Whe i try to start the mgmt server i have this error :
> >>
> >> 2014-10-31 16:05:14,246 DEBUG [c.c.s.ConfigurationServerImpl]
> (main:null)
> >> Execution is successful.
> >> 2014-10-31 16:05:14,285 INFO  [c.c.c.ClusterManagerImpl] (main:null)
> Start
> >> configuring cluster manager : ClusterManagerImpl
> >> 2014-10-31 16:05:14,285 INFO  [c.c.c.ClusterManagerImpl] (main:null)
> >> Cluster node IP : 192.168.0.100
> >> 2014-10-31 16:05:14,292 ERROR [o.a.c.s.l.CloudStackExtendedLifeCycle]
> >> (main:null) Failed to configure ClusterManagerImpl
> >> javax.naming.ConfigurationException: cluster node IP should be valid
> local
> >> address where the server is running, please check your configuration
> >>
> >> And the management server does not start.
> >>
> >> What did i miss ?
> >>
> >> Thanks for your responses.
> >>
> >> Regards, Benoit.
> >>
> >>
> >> 2014-10-30 10:36 GMT+01:00 Rohit Yadav <rohit.yadav@shapeblue.com>:
> >>
> >>> Hi,
> >>>
> >>>> On 30-Oct-2014, at 2:52 pm, benoit lair <kurushi4000@gmail.com>
> wrote:
> >>>>
> >>>> So when doing a "cloudstack-setup-databases" with "-i 192.168.0.100"
i
> >>> have
> >>>> to specify the vip of the netscaler ? As well i am on the first node
> or
> >>> on
> >>>> the second node or N node ?
> >>>
> >>> The -i <ip> is used as the host IP by CloudStack management server,
> among
> >>> other things this is especially used by systemvms to connect to mgmt
> >>> server. So, if you’re load balancing using netscaler, use the
> netscaler IP.
> >>> Make sure to configure ports 8080, 8250 appropriately. Make sure all
> the
> >>> ACS mgmt servers (primary/first one and others) can connect to mysql
> server
> >>> from their respective IPs.
> >>>
> >>>> Yes the management server was crashed, for the moment i don't have the
> >>>> access to this management server pool.
> >>>> As soon i'm getting back to this mgmt server pool, i give you more
> >>> infos.
> >>>>
> >>>> I'm going to reinstall some fresh mgmt servers and give you more
> >>> feedback.
> >>>
> >>> Sure.
> >>>
> >>>> Concerning the mgmt server log entries "Resp: Routing to peer", is it
> >>>> normal that the mgmt server is producing so much log entries with this
> >>>> message (seeing my mgmt log file growing very fast).
> >>>
> >>> Since, both the management servers are loadbalancing internal calls,
> >>> you’ll see these a lot. You may plan your log storage appropriately or
> >>> configure log4j xml to not log INFO/DEBUG etc.
> >>>
> >>> Regards,
> >>> Rohit Yadav
> >>> Software Architect, ShapeBlue
> >>> M. +91 88 262 30892 | rohit.yadav@shapeblue.com
> >>> Blog: bhaisaab.org | Twitter: @_bhaisaab
> >>>
> >>> Find out more about ShapeBlue and our range of CloudStack related
> services
> >>>
> >>> IaaS Cloud Design & Build<
> >>> http://shapeblue.com/iaas-cloud-design-and-build//>
> >>> CSForge – rapid IaaS deployment framework<
> http://shapeblue.com/csforge/>
> >>> CloudStack Consulting<http://shapeblue.com/cloudstack-consultancy/>
> >>> CloudStack Infrastructure Support<
> >>> http://shapeblue.com/cloudstack-infrastructure-support/>
> >>> CloudStack Bootcamp Training Courses<
> >>> http://shapeblue.com/cloudstack-training/>
> >>>
> >>> This email and any attachments to it may be confidential and are
> intended
> >>> solely for the use of the individual to whom it is addressed. Any
> views or
> >>> opinions expressed are solely those of the author and do not
> necessarily
> >>> represent those of Shape Blue Ltd or related companies. If you are not
> the
> >>> intended recipient of this email, you must neither take any action
> based
> >>> upon its contents, nor copy or show it to anyone. Please contact the
> sender
> >>> if you believe you have received this email in error. Shape Blue Ltd
> is a
> >>> company incorporated in England & Wales. ShapeBlue Services India LLP
> is a
> >>> company incorporated in India and is operated under license from Shape
> Blue
> >>> Ltd. Shape Blue Brasil Consultoria Ltda is a company incorporated in
> Brasil
> >>> and is operated under license from Shape Blue Ltd. ShapeBlue SA Pty
> Ltd is
> >>> a company registered by The Republic of South Africa and is traded
> under
> >>> license from Shape Blue Ltd. ShapeBlue is a registered trademark.
> >>>
> >>
> >>
>
> Regards,
> Rohit Yadav
> Software Architect, ShapeBlue
> M. +91 88 262 30892 | rohit.yadav@shapeblue.com
> Blog: bhaisaab.org | Twitter: @_bhaisaab
>
>
>
> Find out more about ShapeBlue and our range of CloudStack related services
>
> IaaS Cloud Design & Build<
> http://shapeblue.com/iaas-cloud-design-and-build//>
> CSForge – rapid IaaS deployment framework<http://shapeblue.com/csforge/>
> CloudStack Consulting<http://shapeblue.com/cloudstack-consultancy/>
> CloudStack Infrastructure Support<
> http://shapeblue.com/cloudstack-infrastructure-support/>
> CloudStack Bootcamp Training Courses<
> http://shapeblue.com/cloudstack-training/>
>
> This email and any attachments to it may be confidential and are intended
> solely for the use of the individual to whom it is addressed. Any views or
> opinions expressed are solely those of the author and do not necessarily
> represent those of Shape Blue Ltd or related companies. If you are not the
> intended recipient of this email, you must neither take any action based
> upon its contents, nor copy or show it to anyone. Please contact the sender
> if you believe you have received this email in error. Shape Blue Ltd is a
> company incorporated in England & Wales. ShapeBlue Services India LLP is a
> company incorporated in India and is operated under license from Shape Blue
> Ltd. Shape Blue Brasil Consultoria Ltda is a company incorporated in Brasil
> and is operated under license from Shape Blue Ltd. ShapeBlue SA Pty Ltd is
> a company registered by The Republic of South Africa and is traded under
> license from Shape Blue Ltd. ShapeBlue is a registered trademark.
>

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