Hi all, I have made an enviroment to test Web Farm Framework and ARR. There are 3 servers with Windows Server 2008 R2. The server A is the Web Farm controller. The server B is the Primary server and C is Secundary. The load balance algorithm is Weighted Round
Robin with Even distribution. I want to have a siteA with high availability, when I try to access it, correctly I am directed alternately to SiteA in Server A or to Server B. Continuing the tests, when I stop the SiteA Application Pool in the Servidor A, instead
of my request be forwarded to Server B, simply return a error Code 503 Service Unavailable. I want to know which configuration I need to make for the ARR forward my request to other server in the Farm that is with its Application Pool running. Thanks!! This
is the configuration of WFF and ARR in the ApplicationHost.config:
According to my test, Application Request Routing (ARR) uses load balance algorithms to distribute incoming requests between several servers despite the content server condition(up or down). So, in your case, even the primary server is down, such request
is still forwarded there, 503 error message is returned.
To achieve high availability, you may want to create a NLB cluster with a failover server. For detailed information, please refer to the following article:
I read the contents in the link that you posted, it teaches how to configure High Availability and Scalability for ARR Servers. But, addition, I want to know if have some configuration to make for when ARR fowards requisitions between content servers, It
checks before if the Site's Application Pool is started, otherwise will be returned error 503, instead ARR forwards the requisition to another content server in the web farm that has the same Site and is working correctly.
I found my mistake, the correct is change the DNS entry of the test site, so that instead of pointing to the IP address of the primary Web server farm, point to the server farm controller. This form all request for access to the site will reach it, and it will
make routing decisions between servers on the farm.
danielvieira...
12 Posts
High availability if Application Pool stop in Primary Server
Apr 18, 2012 09:27 PM|LINK
rewrite>
globalRules>
rule name="ARR_RHIISHA_loadbalance" enabled="true" patternSyntax="Wildcard" stopProcessing="true">
match url="*" />
action type="Rewrite" url="http://RHIISHA/{R:0}" />
conditions logicalGrouping="MatchAny" trackAllCaptures="false">
add input="{SERVER_ADDR}" pattern="192.160.0.1" />
add input="{SERVER_ADDR}" pattern="192.168.0.2" />
/conditions>
/rule>
/globalRules>
/rewrite>
diskCache>
compression>
add mimeType="text/*" enabled="true" />
add mimeType="message/*" enabled="true" />
add mimeType="application/x-javascript" enabled="true" />
add mimeType="application/javascript" enabled="true" />
/compression>
/diskCache>
/system.webServer>
webFarms>
webFarm name="RHIISHA" enabled="true" adminUserName="{0}\admin" adminPassword="
[enc:AesProvider:adfafafadafdJo6Q4UO/9k42afafafdsfadfafStg=:enc]" primaryServer="RHIISHA01">
server address="RHIISHA01" enabled="true">
applicationRequestRouting weight="100" />
/server>
server address="RHIISHA02" enabled="true">
applicationRequestRouting weight="100" />
/server>
platformProvision syncPlatformFromPrimary="true" />
applicationProvision syncWebServerFromPrimary="true" />
applicationRequestRouting>
protocol reverseRewriteHostInResponseHeaders="true" />
loadBalancing algorithm="WeightedRoundRobin" />
/applicationRequestRouting>
/webFarm>
applicationRequestRouting>
hostAffinityProviderList>
add name="Microsoft.Web.Arr.HostNameRoundRobin" />
add name="Microsoft.Web.Arr.HostNameMemory" />
/hostAffinityProviderList>
/applicationRequestRouting>
/webFarms>
ARR AppPool WFF Application Request Routing 2.5 503 {SERVER_ADDR}
danielvieira...
12 Posts
Re: High availability if Application Pool stop in Primary Server
Apr 18, 2012 09:28 PM|LINK
Dalong Zhang...
647 Posts
Microsoft
Re: High availability if Application Pool stop in Primary Server
Apr 20, 2012 09:27 AM|LINK
Hi,
According to my test, Application Request Routing (ARR) uses load balance algorithms to distribute incoming requests between several servers despite the content server condition(up or down). So, in your case, even the primary server is down, such request is still forwarded there, 503 error message is returned.
To achieve high availability, you may want to create a NLB cluster with a failover server. For detailed information, please refer to the following article:
Achieving High Availability and Scalability - ARR and NLB
http://learn.iis.net/page.aspx/511/achieving-high-availability-and-scalability---arr-and-nlb/
Feedback to us
Develop and promote your apps in Windows Store
danielvieira...
12 Posts
Re: High availability if Application Pool stop in Primary Server
Apr 20, 2012 04:27 PM|LINK
danielvieira...
12 Posts
Re: High availability if Application Pool stop in Primary Server
Apr 20, 2012 06:15 PM|LINK
Hi Dalong Zhang,
I read the contents in the link that you posted, it teaches how to configure High Availability and Scalability for ARR Servers. But, addition, I want to know if have some configuration to make for when ARR fowards requisitions between content servers, It checks before if the Site's Application Pool is started, otherwise will be returned error 503, instead ARR forwards the requisition to another content server in the web farm that has the same Site and is working correctly.
Thanks!
danielvieira...
12 Posts
Re: High availability if Application Pool stop in Primary Server
May 02, 2012 05:10 PM|LINK
ARR AppPool WFF Application Request Routing 2.5
danielvieira...
12 Posts
Re: High availability if Application Pool stop in Primary Server
May 04, 2012 05:08 PM|LINK