Re: Failover and replication problem - Weblogic

This is a discussion on Re: Failover and replication problem - Weblogic ; How are you load balancing? Is it possible that the test you are running is getting "sticky" behavior and thus not balancing? Peace, Cameron Purdy Tangosol, Inc. http://www.tangosol.com/coherence.jsp Tangosol Coherence: Clustered Replicated Cache for Weblogic "Campot" wrote in message news:3f159eb8$1@newsgroups.bea.com... ...

+ Reply to Thread
Results 1 to 8 of 8

Thread: Re: Failover and replication problem

  1. Re: Failover and replication problem

    How are you load balancing?

    Is it possible that the test you are running is getting "sticky" behavior
    and thus not balancing?

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    http://www.tangosol.com/coherence.jsp
    Tangosol Coherence: Clustered Replicated Cache for Weblogic


    "Campot" wrote in message
    news:3f159eb8$1@newsgroups.bea.com...
    >
    > I have the following problem.
    >
    > I have configured a WebLogic 8.1 cluster with two servers, one running on

    Windows
    > and the other running on Linux Red Hat 8.0
    > Apparently everything works ok but when I test the fail over and load

    balancing
    > mechanism of the cluster I have the following problem:
    >
    > I am using HttpClusterServlet to redirect the requests to the cluster.

    Although
    > it should use a round-robin algorithm, al the request are directed to the

    server
    > running on windows. Only when the windows server goes down, the request

    are forwarded
    > to the server running on Linux, but the session information is lost.
    >
    > When I tried the same situation but with all the server running on

    windows, everything
    > worked ok.
    >
    > Any ideas why this is happening?
    >
    > Thanks
    >




  2. Re: Failover and replication problem

    I am using HttpClusterServlet as the loas balancer.

    Any ideas why this "sticky" behavior is happening?

    Thanks.
    "Cameron Purdy" escribió en el mensaje
    news:3f1b5555@newsgroups.bea.com...
    > How are you load balancing?
    >
    > Is it possible that the test you are running is getting "sticky" behavior
    > and thus not balancing?
    >
    > Peace,
    >
    > Cameron Purdy
    > Tangosol, Inc.
    > http://www.tangosol.com/coherence.jsp
    > Tangosol Coherence: Clustered Replicated Cache for Weblogic
    >
    >
    > "Campot" wrote in message
    > news:3f159eb8$1@newsgroups.bea.com...
    > >
    > > I have the following problem.
    > >
    > > I have configured a WebLogic 8.1 cluster with two servers, one running

    on
    > Windows
    > > and the other running on Linux Red Hat 8.0
    > > Apparently everything works ok but when I test the fail over and load

    > balancing
    > > mechanism of the cluster I have the following problem:
    > >
    > > I am using HttpClusterServlet to redirect the requests to the cluster.

    > Although
    > > it should use a round-robin algorithm, al the request are directed to

    the
    > server
    > > running on windows. Only when the windows server goes down, the request

    > are forwarded
    > > to the server running on Linux, but the session information is lost.
    > >
    > > When I tried the same situation but with all the server running on

    > windows, everything
    > > worked ok.
    > >
    > > Any ideas why this is happening?
    > >
    > > Thanks
    > >

    >
    >




  3. Re: Failover and replication problem

    > I am using HttpClusterServlet as the loas balancer.
    > Any ideas why this "sticky" behavior is happening?


    Could be that the same cookie or session ID (in the URL) gets sent in with
    every test request.

    Peace,

    Cameron Purdy
    Tangosol, Inc.
    http://www.tangosol.com/coherence.jsp
    Tangosol Coherence: Clustered Replicated Cache for Weblogic


    "Campot" wrote in message
    news:3f1c3192@newsgroups.bea.com...
    >




  4. Re: Failover and replication problem

    I ran utils.MulticastTest and it worked ok. The machines are able to
    communicate via Multicast. Both server logs are OK and both servers
    succeeded in joining the cluster.

    One server is running on Linux and the other on Windows. Is it possible that
    this is causing the error?

    Thanks.

    "Prasad Peddada" escribió en el mensaje
    news:3f1c91e7$1@newsgroups.bea.com...
    > Cameron Purdy wrote:
    > >>I am using HttpClusterServlet as the loas balancer.
    > >>Any ideas why this "sticky" behavior is happening?

    > >
    > >
    > > Could be that the same cookie or session ID (in the URL) gets sent in

    with
    > > every test request.
    > >
    > > Peace,
    > >
    > > Cameron Purdy
    > > Tangosol, Inc.
    > > http://www.tangosol.com/coherence.jsp
    > > Tangosol Coherence: Clustered Replicated Cache for Weblogic
    > >
    > >
    > > "Campot" wrote in message
    > > news:3f1c3192@newsgroups.bea.com...
    > >
    > >
    > >

    >
    > Make sure those two machines are able to communicate via multicast. Did
    > you run utils.MulticastTest?
    >
    > Cheers,
    >
    > -- Prasad
    >




  5. Re: Failover and replication problem

    Campot wrote:
    > I ran utils.MulticastTest and it worked ok. The machines are able to
    > communicate via Multicast. Both server logs are OK and both servers
    > succeeded in joining the cluster.
    >
    > One server is running on Linux and the other on Windows. Is it possible that
    > this is causing the error?
    >
    > Thanks.
    >
    > "Prasad Peddada" escribió en el mensaje
    > news:3f1c91e7$1@newsgroups.bea.com...
    >
    >>Cameron Purdy wrote:
    >>
    >>>>I am using HttpClusterServlet as the loas balancer.
    >>>>Any ideas why this "sticky" behavior is happening?
    >>>
    >>>
    >>>Could be that the same cookie or session ID (in the URL) gets sent in

    >
    > with
    >
    >>>every test request.
    >>>
    >>>Peace,
    >>>
    >>>Cameron Purdy
    >>>Tangosol, Inc.
    >>>http://www.tangosol.com/coherence.jsp
    >>>Tangosol Coherence: Clustered Replicated Cache for Weblogic
    >>>
    >>>
    >>>"Campot" wrote in message
    >>>news:3f1c3192@newsgroups.bea.com...
    >>>
    >>>
    >>>

    >>
    >>Make sure those two machines are able to communicate via multicast. Did
    >>you run utils.MulticastTest?
    >>
    >>Cheers,
    >>
    >>-- Prasad
    >>

    >
    >
    >

    No, there is nothing that is OS specific. Check in the console to make
    sure each server recognizes other server in the cluster.

    Cheers,

    -- Prasad


  6. Re: Failover and replication problem

    Date: Wed, 23 Jul 2003 09:41:49 -0700
    From: Prasad Peddada
    User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.3.1) Gecko/20030425
    X-Accept-Language: en-us, en
    MIME-Version: 1.0
    Newsgroups: weblogic.developer.interest.clustering
    Subject: Re: Failover and replication problem
    References: <3f159eb8$1@newsgroups.bea.com> <3f1b5555@newsgroups.bea.com> <3f1c3192@newsgroups.bea.com> <3f1c6d51@newsgroups.bea.com> <3f1c91e7$1@newsgroups.bea.com> <3f1d65db@newsgroups.bea.com> <3f1d8a52$1@newsgroups.bea.com> <3f1e91d2$1@newsgroups.bea.com>
    In-Reply-To: <3f1e91d2$1@newsgroups.bea.com>
    Content-Type: text/plain; charset=ISO-8859-1; format=flowed
    Content-Transfer-Encoding: 8bit
    NNTP-Posting-Host: 216.148.48.18
    X-Original-NNTP-Posting-Host: 216.148.48.18
    Message-ID: <3f1ebad1@newsgroups.bea.com>
    X-Trace: newsgroups.bea.com 1058978513 216.148.48.18 (23 Jul 2003 09:41:53 -0700)
    X-Original-Trace: 23 Jul 2003 09:41:53 -0700, 216.148.48.18
    Organization: BEA NEWS SITE
    Lines: 88
    XPident: Unknown
    Path: newsgroups.bea.com!not-for-mail
    Xref: newsgroups.bea.com weblogic.developer.interest.clustering:9480

    Campot wrote:
    > It seems to be a problem with the HttpClusterServlet. When I display a list
    > of the Managed Servers that the proxy servlet can see using the page:
    > http://IPort/foo.jsp?__WebLogicBridgeConfig
    >
    > Only ONE of the two servers appear, but when I force shutdown that server,
    > and check again, the other server of the cluster appears alone.
    >
    > Any ideas?
    >
    > Thanks.
    >
    > "Prasad Peddada" escribió en el mensaje
    > news:3f1d8a52$1@newsgroups.bea.com...
    >
    >>Campot wrote:
    >>
    >>>I ran utils.MulticastTest and it worked ok. The machines are able to
    >>>communicate via Multicast. Both server logs are OK and both servers
    >>>succeeded in joining the cluster.
    >>>
    >>>One server is running on Linux and the other on Windows. Is it possible

    >
    > that
    >
    >>>this is causing the error?
    >>>
    >>>Thanks.
    >>>
    >>>"Prasad Peddada" escribió en el mensaje
    >>>news:3f1c91e7$1@newsgroups.bea.com...
    >>>
    >>>
    >>>>Cameron Purdy wrote:
    >>>>
    >>>>
    >>>>>>I am using HttpClusterServlet as the loas balancer.
    >>>>>>Any ideas why this "sticky" behavior is happening?
    >>>>>
    >>>>>
    >>>>>Could be that the same cookie or session ID (in the URL) gets sent in
    >>>
    >>>with
    >>>
    >>>
    >>>>>every test request.
    >>>>>
    >>>>>Peace,
    >>>>>
    >>>>>Cameron Purdy
    >>>>>Tangosol, Inc.
    >>>>>http://www.tangosol.com/coherence.jsp
    >>>>>Tangosol Coherence: Clustered Replicated Cache for Weblogic
    >>>>>
    >>>>>
    >>>>>"Campot" wrote in message
    >>>>>news:3f1c3192@newsgroups.bea.com...
    >>>>>
    >>>>>
    >>>>>
    >>>>
    >>>>Make sure those two machines are able to communicate via multicast. Did
    >>>>you run utils.MulticastTest?
    >>>>
    >>>>Cheers,
    >>>>
    >>>>-- Prasad
    >>>>
    >>>
    >>>
    >>>

    >>No, there is nothing that is OS specific. Check in the console to make
    >>sure each server recognizes other server in the cluster.
    >>
    >>Cheers,
    >>
    >>-- Prasad
    >>

    >
    >
    >

    I suggest you try with some other plugin. In the meantime I will try to
    find out if there is a bug in HTTPClusterServlet.

    Cheers,

    - Prasad


  7. Re: Failover and replication problem

    Date: Wed, 23 Jul 2003 10:50:24 -0700
    From: Prasad Peddada
    User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.3.1) Gecko/20030425
    X-Accept-Language: en-us, en
    MIME-Version: 1.0
    Newsgroups: weblogic.developer.interest.clustering
    Subject: Re: Failover and replication problem
    References: <3f159eb8$1@newsgroups.bea.com> <3f1b5555@newsgroups.bea.com> <3f1c3192@newsgroups.bea.com> <3f1c6d51@newsgroups.bea.com> <3f1c91e7$1@newsgroups.bea.com> <3f1d65db@newsgroups.bea.com> <3f1d8a52$1@newsgroups.bea.com> <3f1e91d2$1@newsgroups.bea.com>
    In-Reply-To: <3f1e91d2$1@newsgroups.bea.com>
    Content-Type: text/plain; charset=ISO-8859-1; format=flowed
    Content-Transfer-Encoding: 8bit
    NNTP-Posting-Host: 216.148.48.18
    X-Original-NNTP-Posting-Host: 216.148.48.18
    Message-ID: <3f1ecae5@newsgroups.bea.com>
    X-Trace: newsgroups.bea.com 1058982629 216.148.48.18 (23 Jul 2003 10:50:29 -0700)
    X-Original-Trace: 23 Jul 2003 10:50:29 -0700, 216.148.48.18
    Organization: BEA NEWS SITE
    Lines: 88
    XPident: Unknown
    Path: newsgroups.bea.com!not-for-mail
    Xref: newsgroups.bea.com weblogic.developer.interest.clustering:9481

    Campot wrote:
    > It seems to be a problem with the HttpClusterServlet. When I display a list
    > of the Managed Servers that the proxy servlet can see using the page:
    > http://IPort/foo.jsp?__WebLogicBridgeConfig
    >
    > Only ONE of the two servers appear, but when I force shutdown that server,
    > and check again, the other server of the cluster appears alone.
    >
    > Any ideas?
    >
    > Thanks.
    >
    > "Prasad Peddada" escribió en el mensaje
    > news:3f1d8a52$1@newsgroups.bea.com...
    >
    >>Campot wrote:
    >>
    >>>I ran utils.MulticastTest and it worked ok. The machines are able to
    >>>communicate via Multicast. Both server logs are OK and both servers
    >>>succeeded in joining the cluster.
    >>>
    >>>One server is running on Linux and the other on Windows. Is it possible

    >
    > that
    >
    >>>this is causing the error?
    >>>
    >>>Thanks.
    >>>
    >>>"Prasad Peddada" escribió en el mensaje
    >>>news:3f1c91e7$1@newsgroups.bea.com...
    >>>
    >>>
    >>>>Cameron Purdy wrote:
    >>>>
    >>>>
    >>>>>>I am using HttpClusterServlet as the loas balancer.
    >>>>>>Any ideas why this "sticky" behavior is happening?
    >>>>>
    >>>>>
    >>>>>Could be that the same cookie or session ID (in the URL) gets sent in
    >>>
    >>>with
    >>>
    >>>
    >>>>>every test request.
    >>>>>
    >>>>>Peace,
    >>>>>
    >>>>>Cameron Purdy
    >>>>>Tangosol, Inc.
    >>>>>http://www.tangosol.com/coherence.jsp
    >>>>>Tangosol Coherence: Clustered Replicated Cache for Weblogic
    >>>>>
    >>>>>
    >>>>>"Campot" wrote in message
    >>>>>news:3f1c3192@newsgroups.bea.com...
    >>>>>
    >>>>>
    >>>>>
    >>>>
    >>>>Make sure those two machines are able to communicate via multicast. Did
    >>>>you run utils.MulticastTest?
    >>>>
    >>>>Cheers,
    >>>>
    >>>>-- Prasad
    >>>>
    >>>
    >>>
    >>>

    >>No, there is nothing that is OS specific. Check in the console to make
    >>sure each server recognizes other server in the cluster.
    >>
    >>Cheers,
    >>
    >>-- Prasad
    >>

    >
    >
    >


    Could you post your HTTPClusterServlet configuration?

    Thanks

    - Prasad


  8. Re: Failover and replication problem

    I found the following bug that is fixed with WLS 8.1 SP1.
    Can this be the reason of my problem?

    "When an HTTP request was proxied by a WebLogic proxy to a cluster using the
    non-default network channel, the cluster list returned was with respect to
    the default channel. Thereafter, the plug-in started proxying to the default
    channel, instead of the original network channel, which may not have been
    reachable.
    This problem was resolved by modifying the code so that the dynamic server
    list is aware of non-default network channels. "

    I did not configure any network channels.
    cheers

    "Campot" escribió en el mensaje
    news:3f20088f@newsgroups.bea.com...
    > These are my HttpClusterServlet`s configuration files
    >
    > Thanks for helping me.
    >
    > "Prasad Peddada" escribió en el mensaje
    > news:3f1ecae5@newsgroups.bea.com...
    > > Campot wrote:
    > > > It seems to be a problem with the HttpClusterServlet. When I display a

    > list
    > > > of the Managed Servers that the proxy servlet can see using the page:
    > > > http://IPort/foo.jsp?__WebLogicBridgeConfig
    > > >
    > > > Only ONE of the two servers appear, but when I force shutdown that

    > server,
    > > > and check again, the other server of the cluster appears alone.
    > > >
    > > > Any ideas?
    > > >
    > > > Thanks.
    > > >
    > > > "Prasad Peddada" escribió en el mensaje
    > > > news:3f1d8a52$1@newsgroups.bea.com...
    > > >
    > > >>Campot wrote:
    > > >>
    > > >>>I ran utils.MulticastTest and it worked ok. The machines are able to
    > > >>>communicate via Multicast. Both server logs are OK and both servers
    > > >>>succeeded in joining the cluster.
    > > >>>
    > > >>>One server is running on Linux and the other on Windows. Is it

    possible
    > > >
    > > > that
    > > >
    > > >>>this is causing the error?
    > > >>>
    > > >>>Thanks.
    > > >>>
    > > >>>"Prasad Peddada" escribió en el mensaje
    > > >>>news:3f1c91e7$1@newsgroups.bea.com...
    > > >>>
    > > >>>
    > > >>>>Cameron Purdy wrote:
    > > >>>>
    > > >>>>
    > > >>>>>>I am using HttpClusterServlet as the loas balancer.
    > > >>>>>>Any ideas why this "sticky" behavior is happening?
    > > >>>>>
    > > >>>>>
    > > >>>>>Could be that the same cookie or session ID (in the URL) gets sent

    in
    > > >>>
    > > >>>with
    > > >>>
    > > >>>
    > > >>>>>every test request.
    > > >>>>>
    > > >>>>>Peace,
    > > >>>>>
    > > >>>>>Cameron Purdy
    > > >>>>>Tangosol, Inc.
    > > >>>>>http://www.tangosol.com/coherence.jsp
    > > >>>>>Tangosol Coherence: Clustered Replicated Cache for Weblogic
    > > >>>>>
    > > >>>>>
    > > >>>>>"Campot" wrote in message
    > > >>>>>news:3f1c3192@newsgroups.bea.com...
    > > >>>>>
    > > >>>>>
    > > >>>>>
    > > >>>>
    > > >>>>Make sure those two machines are able to communicate via multicast.

    > Did
    > > >>>>you run utils.MulticastTest?
    > > >>>>
    > > >>>>Cheers,
    > > >>>>
    > > >>>>-- Prasad
    > > >>>>
    > > >>>
    > > >>>
    > > >>>
    > > >>No, there is nothing that is OS specific. Check in the console to make
    > > >>sure each server recognizes other server in the cluster.
    > > >>
    > > >>Cheers,
    > > >>
    > > >>-- Prasad
    > > >>
    > > >
    > > >
    > > >

    > >
    > > Could you post your HTTPClusterServlet configuration?
    > >
    > > Thanks
    > >
    > > - Prasad
    > >

    >
    >
    >




+ Reply to Thread