Failover Question

I am new to JMS. I have a cluster with two instances (A & B), and I am wanting to implement failover with the brokers. My producer creates a transactional Topic on instance A's broker. My consumer subscribes to that same Topic via instance B's broker (since they are mirrored). What is required so that when I stop instance B, the consumer will continue to receive uninterrupted messages via INSTANCE A? I have already added the IP and Port numbers to the AddressList for both instances, however, when I stop instance B, I get the "[C4000]: Packet acknowledge failed. user=guest, broker=............ " JMSException. I'm expecting the consumer to connect to instance A's broker in the addressList to continue sending/receiving messages. Any guidance? Thanks for your time.

If you have configured for conventional availability clusters, with GlassFish MQ, you should expect to see exceptions. If you have auto-reconnect enabled, you should see your session reconnect to the remaining broker, after the failure. If you are using Enhanced availability clusters, these exceptions are deferred and, the clients will attempt reconnect, as part of the message store "take-over" processing. If the session failover succeeds, you shouldn't see an exception.
If you haven't already, look at the overview documentation: [http://docs.sun.com/app/docs/doc/821-0028?l=en] -- for enhanced availability, look at chapter 4, which describes the differences between the cluster configurations. If you have already done all this, could you give a few more details about your configuration so that we can help you further?

Similar Messages

WLC Failover Question

Hi All
Can anyone give me a definitive answer to this question please?
If you are using a pair of wireless LAN controllers configured with primary and secondary controllers for the access points and the primary controller fails - do the access points reboot before associating to the secondary controller. I can't see why they would need to but documentation suggests they do.
Additionally, has anyone significantly reduced the failover time? If so, what is the lowest practical failover time. I know the actual failover time can be reduced to 3 seconds but I think that is likely to cause other problems.
Thanks guys.
Regards
Roger

As far I know, In this case the AP does not reboot, only changes its lwapp status to discovery and begins with the discovery proccess.
You can see in the AP if it is restarted; when it places registered in the second WLC, Wireless tab and select the AP affected; normaly in the first tab you can see bottom right the AP up time and the AP association time; if this AP has rebooted this value will close to 00:00.
Normaly I set the Ap heartbeat timeout to 5 seconds, I don´t know if is the best value and my failover time is bigger than your, I don´t know how critical are your network, but a prefer a higher heartbeat timeout to avoid unnecessary AP changes that spend more time.
Best Regards.

RHI failover question

Hi,
I have a pair of 6509's with Sup-720's and ACE modules. IOS is SXI3, ACE software is 3.0.
When I do a manual ACE failover from one chassis to another I see this behaviour (from debug ip routing)
switch1:
Jan 14 09:13:32.210 CET: RT(test-context): del 10.0.2.1/32 via 10.0.0.1, static metric [77/0]
switch2:
Jan 14 09:13:34.211 CET: RT(test-context): add 10.0.2.1/32 via 10.0.0.1, static metric [77/0]
My question is this - why two seconds to install the static route on the second switch? The VIP is there pretty much instantly but the static route is lagging behind by two seconds.
Is there any way to speed up the process or is it hardwired? (I've tried other versions of IOS and other versions of ACE code - no difference)
thanks,
Andrew.

sqlnet client failover is defined in either tnsnames.ora or a thin JDBC connection string on the client.

Load balancing with failover questions

If we install 2 multi-role Exchange servers in our building and a 3rd multi-role server in our remote data center, what is the best way to load balance them? Do we need two load balancers or is there some way to span a single load balancer across the
WAN ?
What about using Windows NLB as an alternative to using round robin internally?
Can a load balancer keep our interoffice Exchange CAS traffic from leaving our LAN and only failover to using the 3rd CAS/mailbox sever for internal users if both internal Exchange servers are offline?
We would also like remote users to "prefer" to use the data center CAS unless it is down. Right now we point our smart host directly to a CAS, but if we had a load balancer there, we could point the smart host to the IP of the load balancer and
the load balancer could normally send it to data center CAS if it's up and forward it to one of the servers in the office otherwise.
Is it possible to do all this without a very complicated and expensive solution?

Depends... what is the connectivity speed between two sites, is it good enough?
You can use load balance in front of all the 3 CAS if your inter-site connectivity is very good.
What about using Windows NLB as an alternative to using round robin internally? WNLB and round robin is different,. You can use DNS Roud Robin if you want to or WNLB for all three CAS Server. Or Hardware loadbalancer for all three CAS servers
Can a load balancer keep our interoffice Exchange CAS traffic from leaving our LAN and only failover to using the 3rd CAS/mailbox sever for internal users if both internal Exchange servers are offline? If you want to use the load balancer then you don't
need to fail them over one by one -- again you can use DNS Round Robing so the request will go to eah CAS servers one by one or use Hardware Load balance.
We would also like remote users to "prefer" to use the data center CAS unless it is down. Right now we point our smart host directly to a CAS, but if we had a load balancer there, we could point the smart host to the IP of the load balancer and
the load balancer could normally send it to data center CAS if it's up and forward it to one of the servers in the office otherwise.
Use DNS Server and point the A record to the Primary Data center load balanced CAS server instead using IP or host file.
Hope that helps
Where Technology Meets Talent

Two Wireless controllers load balance and failover question

I have two 4404 controllers and each can take 100 APs. I have 140 APs in total. With the default settings (no master controller, no configuration of Prime, secondary controller on APs), each controller will take 70 APs, right?
Then I will need to configure each AP with an IP address, name ...etc. My question is, when one Controller failed, these 70 APs will try to associate with another controller, right? However only 30 APs can because another controller can maximum manage 100 APs. Then in this case, will these 30 APs lose their static IP addresses and names? When the failed controller came back online, will the 70 APs automatically go back to this controller and have their IP, name configuration back?
Thanks!

With default setting you have no control how many ap's go to what wlc. It doesn't matter, because you will need to specify the primary and secondary. You might as well stage all the ap's you want on one wlc first and set that wlc to master, then when you have finished that, set the other wlc to master and have the ap's join that wlc which will be the primary fro those ap's.
You only can support 100 ap's so depending what code you use, 30 ap's that are not able to join will just keep trying. If you run 5.2 (I think is buggy) you can set the priority on the ap's so that ap's that you set up with a higher priority will be able to join and the others will again sit there until the othe wlc comes back up. Static IP address will not dissapear because the wlc doesnt' accept any more connections. Once both wlc are up, the ap's will go back to their primary wlc as long as ap fallback is enabled and mobility is configured right.

9i Standby Database - Failover Question - REF 227196.1

Hi ,
Im researching the use of a Standby DB for out Production DB....
The Metalink DOC 227196.1 mentions that in failover , the Production going down , Standby becomes Primary and cannot be come Standby again , whereas with Switchover , they can be swapped as many times as you need for upgrades etc...
"The Difference between a Switchover and a Failover is, that after a Failover, the Standby Database becoming a Primary now, cannot switchback to become a Standby Database again. In Opposition a Switchover exchanges the Database Roles (The Primary becomes a Standby and the Standby becomes a Primary). Switchovers can be performed arbitrarily, a Failover only once."
If failover happens can someone tell me or point me in the right direction:
a) Can you config Failover to happen automatically?
b) If failover occurs , do you need to reconfig all the Primary/Standby DB again in the event that failover happens again?
c) Any clear step by step guide to config of standby... The Oracle docs whilst very good are robust and some "summary" steps would be great
thanks in advance
Stephen

HI,
Please check this link:
Physical Standby Database SWITCHOVER &amp; FAILOVER Operations | Talip Hakan Ozturk&#039;s ORACLE BLOG
Thank you

CSS 11503 adv-bal-stcky-srcip and failover question

From the documentation I have read about failover (when a service fails) it lists several loadbalancing types, but the advanced-balance-sticky-srcip is not one of them. Is it possible to configure failover linear or failover next when using adv-bal-stcky-srcip? The CSS is configured for a kal (type=tcp / port=8080), but does not do anything when the service does not respond the kal.

Let me add some clarity then. :) I am running PeopleSoft through the CSS (which is the reason for adv-bal sticky-srcip). I have two services in my content rule:
service pplsft-web1
ip 172.27.144.63
port 8080
kal tcp / 8080
service pplsft-web2
ip 172.27.144.63
port 8080
kal tcp / 8080
When the KAL fails and the service is marked down, what is the CSS supposed to do?...you can configure the failover of a a service (ie: failover linear or failover next). In all the docs I have read, I read that 'failover linear' and 'failover next' were for regular load balancing techniques (ie: domain, url, srcip, destip, domainhash and urlhash). Can I use it if I have the advanced-balance sticky-srcip load balancing command on the content?
Does that clarify any?
I did get an answer from TAC...by default, 'failover linear' is enabled. But what may be happening is that because of the sticky config, the user ip is still in the sticky table, which over-rides load balancing. I have to define the settings for "sticky serer down failover" to either 'balance' or 'reject' entries in the sticky table and any requests that come in.

Failover Questions

Hey guys,
Just doing some studying and running into something that I am not quiete understanding...
If i have 2 firewall's in Active/Active Stateful failover mode and 2 contexts (E1 and E2). Let's say ASA1 has E1 as the active context and ASA2 has E2 as the active context. E2 is the only context used to connect Router_X. If I need to permit traffic to Router_X, would I make the ACL in the ASA1 E2 context (secondary) or in the ASA2 E2 context (primary)?
I completed an Active\Active Statuful failover configuration between 2 firewalls, but once I was finished I remembered that i didn't configure the failover group 2 as secondary (problem). So i went ahead and make the configuration change, once I did so I entered the commands NO FAILOVER/FAILOVER to "resynch" the configurations between the 2 firewalls. Is this necessary or couldn't I just perform a WRITE on the primary ASA?
Is there any command that will verify that each of the configurations on both firewalls are syncrhonized?
Thanks ahead of time guys!

Hi Kenneth,
Here are your answers:
1. If you need to make changes, always do that on the active context, replication is always done from active to standby, so you need to make changes on E2 active context on ASA2.
2. You need not do this everytime, just do a write mem or write standby, that would save teh configuration on the standby context as well.
3. There is no command to verify the command replication, you can check the status of the contexts through "show failover" in the system context, if they show active and standby, then everything is fine. You can study different failovers status's from here:
http://www.cisco.com/en/US/docs/security/asa/asa72/command/reference/s3_72.html#wp1285409
Hope that helps
Thanks,
Varun Rao
Security Team,
Cisco TAC

EJB- Db failover question

Can anyone help explain what are my options for database failover? I am
          using WLAS 5.1. I have seen a lot of information on servlets replication,
          web proxying, and EJB clustering, but what about Db failure and rollover to
          a secondary Db? Is this something better handled at the Db level or is there
          anything at the application for managing this? I was intending to delve into
          Db replication and have a complete backup Db system, but am fuzzy on how to
          implement the EJB-JDBC connection / connection pools.
          Thanks
          - Ben Embree


          Ben wrote:
          >
          > Can anyone help explain what are my options for database failover? I am
          > using WLAS 5.1. I have seen a lot of information on servlets replication,
          > web proxying, and EJB clustering, but what about Db failure and rollover to
          > a secondary Db? Is this something better handled at the Db level or is there
          > anything at the application for managing this? I was intending to delve into
          > Db replication and have a complete backup Db system, but am fuzzy on how to
          > implement the EJB-JDBC connection / connection pools.
          >
          > Thanks
          > - Ben Embree
          Hi Ben. Database failover is best handled by the DBMS offering some sort of
          transparent failover to a hot-standby DBMS instance, using dual-ported disks
          so the user's transactional state is retained. Oracle purports to offer such
          for OCI clients, and if so, our JDBC driver for Oracle would fully, passively
          benefit. The driver and application would never know the connection had been
          switched.
          However, the fundamental issue is that relational DBMSes currently enforce
          a tight bond between a transaction and a single unbroken socket connection
          to a single client OS process. Whenever this socket connection is broken,
          whether by the client dying, a firewall closing it for idleness, etc, the DBMS
          will roll back the tx. No other connection to that database can ever contribute
          to, adopt, or resurrect the lost tx. Therefore, at any higher level than the DBMS,
          the only viable response to a connection failure is to assume a total tx rollback,
          and to reconnect to the DBMS, and restart the transaction from the very beginning.
          We are shortly going to offer 'multipools', which will look like a basic pool
          to clients, but will consist of an ordered list of basic pools, and when a connection
          request comes in, if the DBMS for the first pool in the list is down, we will try
          to return a connection from the next pool, presumably to another DBMS. This is a
          failover pool. This can only work during connection attempts, so any DBMS failure
          except at connect time will still require a manual re-connection as a part of redoing
          the whole tx. With respect to clustering, as soon as any JDBC is started for a given tx,
          it must be pursued to completion through the given cluster instance. No other cluster
          instance can contribute to JDBC within the tx to that DBMS except by making a
          type-3 JDBC connection to the other cluster instance to pass to the one real DBMS
          connection.
          I hope this helps,
          Joe Weinstein
          PS: Folks: BEA WebLogic is in S.F., and now has some entry-level positions for
          people who want to work with Java and E-Commerce infrastructure products. Send
          resumes to [email protected]
          The Weblogic Application Server from BEA
          JavaWorld Editor's Choice Award: Best Web Application Server
          Java Developer's Journal Editor's Choice Award: Best Web Application Server
          Crossroads A-List Award: Rapid Application Development Tools for Java
          Intelligent Enterprise RealWare: Best Application Using a Component Architecture
          http://www.bea.com/press/awards_weblogic.html

VIP failover question

We recently had a node failure in our 2 node Solaris 10 and Oracle 10.2.0.3 cluster. When the failed node was restored and came back online the Load balancing in the cluster was still off. The node that had not failed was still taking all the requests where the restored node was idle.
The final resolution was to restart crs on the node that had not failed. This balanced the cluster again. At first I thought it was related to the listener and how it handles the load balancing but now Oracle and I believe it may be due to the VIP not failing back.
Has anyone else seen this? Is there a way to manually fail a VIP from one node to another? I am still waiting to here more from Oracle but was hoping someone in the forums might have some insight.
Thanks

fjlee, the ML note you are referring, says :
"*What happens when the Public adapter comes back?*
The VIP will be relocated back to its host node, this will happen the next time Oracle Clusterware checks the instance. This check is carried out, by default, every 600 seconds so this is the maximum amount of time that could pass before the relocate is triggered. The VIP can manually be relocated back to its home node using the srvctl start nodeapps –n <nodename> command.
then, if the relocation did not appear after 600 seconds, we have got an issue => SR should be open

Clustering for failover question

We are using Weblogic 5.1 and TopLink for container managed Entity beans.
          We are able to use TopLink/EJB cache when weblogic is running on only one
          machine. We are trying to cluster and still use caching feature. We wanted
          to use Weblogic cluster only for fail-over purpose (not for load balancing)
          i.e we want to use only one weblogic instance all the time until the machine
          dies/hangs and start using second machine until it dies then come back to
          first one etc., I wanted to know if this is possible with weblogic cluster.
          The only reason we wanted to dothis, is we need to use cache of entity beans
          and there is no distributed sync of caching feature available.
          Thanks in advance -- Sree


Yes, you can use CallRouter for this. Look in the documentation for more
          information.
          - Prasad
          http://www.weblogic.com/docs51/classdocs/API_ejb/EJB_reference.html#1026570
          xxx wrote:
          > We are using Weblogic 5.1 and TopLink for container managed Entity beans.
          > We are able to use TopLink/EJB cache when weblogic is running on only one
          > machine. We are trying to cluster and still use caching feature. We wanted
          > to use Weblogic cluster only for fail-over purpose (not for load balancing)
          > i.e we want to use only one weblogic instance all the time until the machine
          > dies/hangs and start using second machine until it dies then come back to
          > first one etc., I wanted to know if this is possible with weblogic cluster.
          > The only reason we wanted to dothis, is we need to use cache of entity beans
          > and there is no distributed sync of caching feature available.
          >
          > Thanks in advance -- Sree

Reliability: OD Master & Replica, or IP-based failover?

I have an Xserve that serves approx. 60 users and around 200GB of shared files. The primary Xserve is the OD master for my workgroup, and the backup Xserve is an OD replica.
I'm new to OS X Server (but not client), so I don't know what happens in the event of, say, the OD master dying. Do client login/access requests automatically get answered by the OD replica? The reason I'm asking this is I'm trying to decide between using two separate servers with an OD master & replica relationship, or using IP-based failover where the backup server would effectively "pretend" to be the primary server in case of a problem.
But with IP-based failover, I don't know if the backup server will automatically know to become an OD master or not. Or, perhaps, the OD integration is completely separate; and I can choose to use IP-based failover and do OD replication whichever way I want to. Does anyone have concrete info?

If your OD Master fails the OD Replica will automatically start handling authentication requests so for the most part you're covered (at least users can still log in).
As noted in that .pdf, though, the replica doesn't allow changes to the directory other than password changes for users with Open Directory passwords.
The main issue is how long your master is going to be down, and the relative pain of promoting the replica to be a master.
For various reasons there is no automatic promote-to-master function because most failures of the master are temporary (e.g. a simple reboot after a software update).
For the failover question I would use IP failover for any services that don't have their own failover mechanism. In all cases you need to write failover scripts to startup whatever processes you need your standby machine to take over. In the case of OD there is a built-in, automatic failover process so I would use that. The only exception to that would be if you wanted the replica to become a master server should the master fail, but I can't think of any circumstance where that would be the case.

Questions on Client Failver and Fast Failover

've few questions regarding Client Failover and Fast Faiolver of Oracle Database HA. Before I ask those questions I would like to explain my environment. Below are the details.
- We have two physical locations called 'ABC' and 'PQR'
- ABC is the primary site.
- PQR is the standby site.
- In ABC, we have Oracle RAC database (11.2.0.2) with two nodes.
- In PQR, we have a single server standalone database (11.2.0.2) with ASM. This is not a RAC.
- Data Guard has been configured between ABC and PQR and it is working as expected.
- Please note that we have a licence for Active Data Guard.
- We have Oracle Identity Management products at both ABC and PQR and they are going to use RAC database as a primary database which is in ABC.
- We did not configure Data Guard broker yet.
We want to achieve below goals:
Goal 1:
Whenever RAC primary goes down completely, standby database should become primary database AUTOMATICALLY and it should allow read/write operation.
I guess this is called 'Fast Failover'. Please let me know if I am wrong.
Questions :
- To make this happen, Do I need to configure Data Guard Broker so that standy database becomes primary when RAC goes down completely with planned or unplanned outage.
- Let's say RAC goes down completely, how long does Data Guard broker take to make standby db as Primary.
- What about the client/application who already connected to RAC.
- Let's standby DB has become as primary and after sometime if RAC comes back , Does data guard automatically changes the role of RAC to primary ?
Goal 2:
As I explained above, all Oracle IDM products and applications speak to RAC database they only know about RAC database which is primary.They are not aware of standby database.
- Whenever a client session is in progress with RAC primary database, if RAC goes down completely , we would like to expect client session should get transferred standby datbase without loosing session information . However before this happens, standby database should become primary becuase client session may perform write operations.
- Whenever a client is trying to connect to RAC prmary and assume RAC is completely down, we would like to expect client connections should get transferred to standby database.
However before this happens, standby database should become primary becuase client session may perform write operations.
As per my knowledge, above scenarios are called 'client failver'. Please let me know if I am wrong.
Questions:
1. Please throw some light to achieve above features.
2. As per my understanding, before client failover happens, fast failover should have already occured and standby should get switch to primary role. I guess all this happens through TIMEOUT parameters. What are those.
Could you please help ?
Thanks

859875 wrote:
Goal 1:
Whenever RAC primary goes down completely, standby database should become primary database AUTOMATICALLY and it should allow read/write operation.
I guess this is called 'Fast Failover'. Please let me know if I am wrong.You are correct.
>
Questions :
- To make this happen, Do I need to configure Data Guard Broker so that standy database becomes primary when RAC goes down completely with planned or unplanned outage.Yes (you can also use Grid Control that will use Data Guard Broker).
>
Goal 2:
As I explained above, all Oracle IDM products and applications speak to RAC database they only know about RAC database which is primary.They are not aware of standby database.
- Whenever a client session is in progress with RAC primary database, if RAC goes down completely , we would like to expect client session should get transferred standby datbase without loosing session information:This is not possible: it is possible only for SELECT statement and only in a single RAC database.
You can find some interesting documents on MAA home page (best pratices, case studies...):
http://www.oracle.com/technetwork/database/features/availability/maa-090890.html

SQL Server 2012 Failover Cluster configuration questions

Hi,
I have few questions on , SQL Server 2012 Failover cluserting pleasse provide suggestions:
1) In SQL Server 2012 is there a configuration for active / passive Failover Cluster installation? If so how is it done? if you could provide any links or articles that would help.
OR
Is this been replaced by Always ON Availability Groups?
2) Also in our environment we have done active/active installation but at a time my understanding is only node has the ownership to the shared storage versus both the nodes, is that correct? If not please provide an explanation?
Any additional information would be valuable in clearing my doubts?
Thank youy
Malini=

Hi malinisethi,
If you install SQL Server in a cluster and configure Active-Passive cluster. In the first node, select “New SQL Server Failover Cluster installation” option , when we are installing Active-Passive cluster, we have to specify one virtual/network name. (Note:
For Active-Active clustering you have specify different network names as per the number of nodes). on the other nodes is similar to installing on the first node except that we have select the Add Node to a SQL Server failover cluster option from the initial
menu. For more information, there is a similar issue about SQL Server 2008 Active-Passive/Active-Active cluster installation, you can review the following article.
http://sqldbpool.com/2009/10/07/sql-server-2008-active-passive-cluster-installation/
About Active/Active SQL Cluster, two clustered SQL Server instance are created in different nodes. Then apply Active /Active configuration of both instance. There is an example about creating an Active/Active SQL Cluster using Hyper-V. you can review the
following article.
http://blogs.msdn.com/b/momalek/archive/2012/04/11/creating-an-active-active-sql-cluster-using-hyper-v-part2-the-clustered-instances.aspx
Regards,
Sofiya Li
Sofiya Li
TechNet Community Support

Question about updating SQL Server on failover cluster

Hi all,
I would like to better understand the update process of SQL Server (2008 r2) in a failover cluster environment.
In our cluster we have 4 instances of SQL running, where two of them I am able to at anytime stop the instance and update it.
The question is... Applying the latest update, will update forcibly all instances installed on that node, or do I have the ability to choose which instance I want to update?
This way, I could "test" the update process in those two SQL Instances, as described in http://support.microsoft.com/kb/958734 and after finishing I would update the remaining instances.
Please let me know.
Thanks in advance.
Nuno Silva

Hi Nuno,
It's a failover cluster issue, I moved it to SQL Server High Availability and Disaster Recovery. So that you can get better help.
Regards,
Charlie Liao
TechNet Community Support

Failover Question

Similar Messages

Maybe you are looking for