Distributed services with load balancing and failover?

Hullo;
What platform would you use to implement something like the following:
* easy registration of various services
* delegation of a service request to the best candidate of many, based on some measure (probably reported by the services themselves)
* quick failover and location of an alternate service in case the best candidate does not respond (real-life environment, uncertain networks and servers;)
RMI could be a starting point, with a custom SocketFactory to take care of the timeouts and redelegations and a good delegator service to work through. The service concept sounds a lot like JINI, but I don't see any provision for best candidate selection, and wonder whether JINI would really save any time compared to RMI in this case.
Is there anything else I should be aware of? I wouldn't mind finding a pre-built wheel. (Cougaar (http://www.cougaar.org/) is on my reading list; a quick glance gives me the impression it may be a bit too heavy on the communication level, but maybe I'm wrong.)
Thanks for your thoughts;
//ata

ata,
Before jumping to anything so bloated and limited as cougaar, take time to consider what you really need. Before grabing at the fanciest Java features like RMI, JINI, and custom SocketFactories; focus on what you are trying to accomplish.
There are plenty of great answers right here, at this forum.
Good hunting,
John

Similar Messages

Dual WLAN links with load balancing and failover

Hello,
I am in a scenario where I am in need of two WLAN links between two buildings. There is a distance of 100-150 meters and minimum bandwidth required for both links together is 300Mbit/s. The thing is that both links should use load balancing between them and if one of them goes down, the last one should act as fail over.
I have been looking at Cisco Aironet 1550 Series though I have no idea what is needed to get load balancing and fail over to work, so I am searching here for suggestions on what equipment is needed.
Something like this:
              ---------------WLAN Link 150-300Mbit/s-----------
Building                    Load balancing and fail over               Building
          ---------------WLAN Link 150-300Mbit/s-----------
Thanks in advance!

Several points.
When an AP is doing 300Mbps, that's NOT the real throughput you have. It's the data rate at which traffic is sent.
All in all, if your AP/client are doing 300MBps association, you will see max 150Mbps with a file transfer.
From there, I'm not even sure that 11n supports dual spatial streams over such long distances (you can't have multipath in open air) so afaik the 1550 only do 150Mbps association rate (=dual channel with one spatial stream). That means 75Mbps real speed.
I couldn't test a 1550 yet so don't take my word for official statement but that's what I'm thinking.
the wireless links will always be both up and they can be on different channels.
That will then mean that it will be "as if" the remote switch was connected directly to the central switch (where WLC is connected) as the WLC tunnels traffic all the way. So you could do a spanning-tree config on this one I guess to block the port onthe remote switch.
Regards,
Nicolas

Distributed HA cluster with load-balancing and failover: advice?

My workplace has a Xeon Xserve, which acts as our primary external server, with an attached ActiveStorage XRAID. We have just purchased a second Xserve/XRAID set to act as a mirror, which we will colocate. Both have Leopard Server installed, along with an array of additional software.
What we want to do is have both servers load-balance between the two, with failover in case of a server or XRAID fault. I plan on using RSYNC to mirror static files between the two, and I'm looking into PostgreSQL replication and load-balancing solutions for our database. I gather that Apache supports web-server failover and load-balancing, as well. But, that still leaves the actual host and network setup to arrange.
Does Leopard server support such a thing? The only information I found on IP failover instructs the user to place the two servers on the same subnet, directly connected via ethernet cable; obviously, this would not work in my case.
Ideally, what we'd end up with is a situation in which the two systems kept each other in sync, both in static files and database data, and load-balanced between themselves; in cases of failure, the remaining system would transparently assume all duties until the other was restored, at which time they would resynchronize
Any suggestions on how I could arrange such a thing?

Interesting. Does this DNS-based approach support session tracking, though? I would need to have a user directed to just one of the two servers for the duration of their session, to avoid having to synchronize temporary files and such.
You can't have it both ways. You need to build tolerance into the app.
At the simplest level where you run all traffic to one site and use the second site as a failover/standby site you'll be OK most of the time - all users will go to the same server and their sessions will be intact.
However, under any failover situation (your primary site is down for some reason), there is going to be some level of session traffic that it going to switch over to the other site. If your site depends on sessions then you're going to need to tolerate this kind of situation - your app will need to be able to fail gracefully if a user comes in with an invalid session cookie.
Note, though, that this may be less of an issue than you at first think - all DNS clients will cache DNS data for whatever TTL you set. This means that if a user looks up your site name and you return an IP address with a 30 minute TTL, then that user is going to use the same IP address for the next 30 minutes and isn't going to ask the server again. This should negate most chance of a user suddenly switching from one server location to the other in mid-session.
The trick comes in setting the DNS TTL low enough to effect a failover, yet not so long that you impact performance - e.g. you don't want the user to perform a DNS lookup on every page load. You may find that 10 minutes is appropriate. Just bear in mind that this affects how long a user could see your site 'down' before the failover DNS kicks in. Clearly you don't want to set the DNS TTL to a day since that may prevent the user switching to the secondary site for 24 hours by which time, hopefully, the primary site is back up, anyway.
The 'right' TTL value may take some analysis on your traffic to see how long a typical user 'session' is. If the average user spends 20 minutes on your site, then it would make sense to set your TTL to somewhere around 20 minutes to give the best chance of their entire session staying on the same server.

Load Balancing and Failover with 10G Standard Edition

Hi,
I am new to Oracle Replication and need some help setting up replication for load balancing and failover. Is this possible using Oracle 10G Standard Edition? I plan on having all updates done on the master site and both databases will be for reads. In case of failure of the master site, I would need to be able to failover to the other database.
Also, if anyone knows of any documention for Basic Replication in 10G, please let me know.
Thanks.

Simple nnapshot replication of data would require significant manual effort to configure to load balance or failover. One the load balancing side, you would generally be limited to to static load balancing-- assigning half the users to one machine and the other half of the users to the other machine, regardless of who is actively using the machine. Failover would be a significant manual effort, particularly to bring the failed machine back into the cluster. You would be implementing the guts of multi-master replication.
Frankly, if you actually have a system which is valuable enough to need load balancing and disaster recovery, I'm going to wager that it will be far cheaper even in the short run to buy more boxes and/or enterprise edition licenses than to try to implement this sort of thing yourself. In the long run, it will be far cheaper, since it will be far easier to maintain. Building all this yourself would probably be penny wise and pound foolish.
Justin
Distributed Database Consulting, Inc.
http://www.ddbcinc.com/askDDBC

Load balancing and Failover

Hello,
We are wondering how load-balancing and failover of tpcall() work with
WTC:
The scenario:
We have one WLS Domain and two Tuxedo Domains. The Tuxedo Domains offer
the same set of services.
In the bdmconfig.xml, we specify connection_policy as 'ON_STARTUP' for
both Remote Tuxedo Domains. We also Import (T_DM_IMPORT) the same
Tuxedo Service from both Tuxedo Domains.
Questions:
1. Is there any load-balancing of the tpcall between the two Domains? If
so, is it round-robin? If round-robin, what determines the order?
2. If it is ONLY Failover, what determines the order of the tpcall? And,
is the Failover automatic? Or do we need to code for retry on failure?
3. ON_DEMAND vs ON_STARTUP: Does ON_DEMAND drop the connection to the
remote domain upon tpterm? And does ON_STARTUP use a pool of
TuxedoConnection objects?
4. Are there any configuration parameters for
'max_number-of_connections? What determines how many simultaneous
connections can be made?
Thanks,
Suresh Mohan.

Hi Suresh,
The following are my answers to your questions.
Suresh Mohan wrote:
Hello,
We are wondering how load-balancing and failover of tpcall() work with
WTC:
The scenario:
We have one WLS Domain and two Tuxedo Domains. The Tuxedo Domains offer
the same set of services.
In the bdmconfig.xml, we specify connection_policy as 'ON_STARTUP' for
both Remote Tuxedo Domains. We also Import (T_DM_IMPORT) the same
Tuxedo Service from both Tuxedo Domains.
Questions:
1. Is there any load-balancing of the tpcall between the two Domains? If
so, is it round-robin? If round-robin, what determines the order?Yes there is a load balancing between two remote Tuxedo TDomain Gateways.
The algorithm is random, not RR. Over time this should give equal
opportunities to both remote TDomain.
>
2. If it is ONLY Failover, what determines the order of the tpcall? And,
is the Failover automatic? Or do we need to code for retry on failure?The load balancing is always there. The failover is automatic. When a
connection to a remote TDomain encountered a problem (ie network) the remote
domain will be put on retry open connection (in ON_STARTUP) and the load
balancing will not select it until the connection re-established.
However, the tpcall() that encountered the error will not be retried to send
to different destination. It is up to the application to decide whether it
want to resend. Any requests called after the error will not select the
failed Remote TDomain.
>
3. ON_DEMAND vs ON_STARTUP: Does ON_DEMAND drop the connection to the
remote domain upon tpterm? And does ON_STARTUP use a pool of
TuxedoConnection objects?TPTERM() only terminate your application session to WTC. WTC still maintain
a secured T-session to remote Tuxedo TDomain. WTC does not use a pool of
TuxedoConnection Objects, the object stored in the JNDI refers to WTC.
>
4. Are there any configuration parameters for
'max_number-of_connections? What determines how many simultaneous
connections can be made?No. As described in #3, there is no need to use connection pool in WTC. WTC
uses session and virtual circuit design concept as Tuxedo TDOMAIN, the
logical pool is created/destroyed dynamically. That is the reason why you
can have a lot of TPACALL() outstanding at the same time. (The limitation is
the availability system resource.)
>
>
Thanks,
Suresh Mohan.Regards,
Hong-Hsi :-)

Network Load Balancing and failover for AFP Sharing

Dear all,
Somebody kindly teach me to use round robin DNS to perform the network load balancing, it's success but not the failover.
I have 4 xserve and want to do the load balancing and failover at the same time.
I have read the IP failover document and setup it successfully, but anyone know is it possible to do the IP failover for more than 2 server?
For example, 4 server serving the AFP service at the same time, maybe I have 1 more extra server to do the IP failover for thoese 4 servers.
As I know, IP failover require Firewire as the heartbeat detection. But one xserve only have 2 firewire ports. May I setting up the IP failover only by a ethernet port and an IP address? does it possible to detect and failover to any server after server down has been detected?
I believe load balancer maybe the best solution but its cost is too high.
Thanks any advance!
Karllee

well, u have 2 options here
software load balancing
request comes it foo.com -> ws7u2 hosting foo.com is configured to run as reverse proxy . this server sends any incoming requests to one of the four back end web server 7 handling your incoming request
hardware load balancing (this you need to invest)
request comes to hardware load balancer who responds for foo.com -> sends requests to four ws7 server hosting your application
you could try out how software load balancing works out for you before you invest in hardware load balancing
here is more instruction on configuring ws7 + reverse proxy (software load configuration)
- install ws7 on foo.com
- create a new configuration (choose port 80, disable java

OCS on a cluster with Load balancing and fail safe environment

Dear all,
i want to ask is there any documat or hints on how to do an OCS R2 installaiotn on 3 server with RAC option (clustered Fail Safe), how can i install OCS on a cluster with Load balancing and fail safe environment.
plz i need ur help
thanking u
[email protected]

Dear all,
i want to ask is there any documat or hints on how to do an OCS R2 installaiotn on 3 server with RAC option (clustered Fail Safe), how can i install OCS on a cluster with Load balancing and fail safe environment.
plz i need ur help
thanking u
[email protected]

Load balancing and failover in Embedded LDAP in weblogic

How to handle load balancing and failover in Embedded LDAP in weblogic server?

You should consider posting this to the Weblogic and/or LDAP support forums. This forum is meant for Sun Web Server questions.
Thanks
Manish

Load Balancing and Failover in RMI.

Hi All,
I have just one RMI object doing mathematical
calculations.I want failover (and perhaps load
balancing also) for this rmi object.Few Application
servers(like weblogic) provide a wrapper over sun's
rmi to produce replica aware rmi stubs but they force
to use vendor specific (like weblogic rmi-registry)
rmi-registry, which runs within the app server.
My problem is :
I dont want to run the whole app server to just run
one rmi object ?
Is there a way using which I can make replica aware
rmi stubs without using the app server ?
Thanks in advance for the help !!
Regards
Rais

your question is pretty vague. In case you are trying to do load balancing using load balancers then your application layer is actually supposed to be blind as far as the working of the networking layer is concerned (and thats where the load balancers exist). Then the question would also be do you want to achieve this to make the system highly available or make it more resilient and improvr performance. There would be different implementations for both of them. However if you are planning to introduce load balancing programmatically you need to start more than one rmi services.

ASA Vpn load balancing and failover

Hello all.
We have two asa5520 configured as primary and standby unit in failover configuration, and all is working properly.
Is it possible, with this configuration (failover), to configure vpn load balancing/clustering?
Thanks
Daniele

Hi Wajih,
I am testing this right now. In my case, I want A and B are failover pairs with A as the primary, (A+B) together as one member in cluster with other ASAs C and D. Here is what I found out:
1, After the active/standby working, configure the load banlancing in the master, the cluster IP worked.
2, after "no fail ac" in A, cluster IP stopped working. Seems the vpn load banlance configuration wasn't copied over to the standby B.
3, In the active (now it's the secondary B), manually configure vpn load banlancing, then the cluster IP worked.
4, "no fail ac" in the B and make the the primary A active, the cluster IP still worked.
5, after "no fail ac" in A, cluster IP stopped working. show vpn load and found out the load banlance was disabled.
6, "no fail ac" in the B and make the the primary A active, the cluster IP then worked.
Based on above, the secondary B's VPN load banlance will be disabled when B becomes active in failover role. If that's true, these two features can't work together. Or maybe there is some configuration I'm missing -- maybe having C or D as the cluster master will help. The ASAs are 5510 with 8.4(2)
Thanks,
Rick.

Load Balance and failover over 2 lines.

We have 2 x Cat 6509 with Sup720 and are running Native IOS.
On Switch A, we have a 1GB link configured as an access link on Vlan 2 to DR site connected to a Cat 3550
On Switch B, we have a 1GB Link configured as a trunk to the same DR connected to another Cat 3550
We configured devices to be in VLAN 2 as we want them to use the 1GB link. And the 2nd 1GB link is a logical extension of our main site.
Now, I want to configure these 2 x 1GB link so that they now act as failover as well as Load balancing.
1. How do I do this?
2. Will I need to move the 1GB link on Switch A to switch B
3. Should I configure the Switch A link also as Trunk?
Please help.
Thanks

Do your 3550s run EMI or SMI? If they run EMI, why not run dynamic routing protocol between the 3550s and the 6500s? OSPF would be a good one to use, if you have hierarchical IP addressing. If your addressing is not hierarchical, then use EIGRP.
Other than this, the only other possible way I can think of is GLBP (http://www.cisco.com/en/US/products/sw/iosswrel/ps1839/products_white_paper09186a00801541c8.shtml)
Prior to GLBP, you could also have used HSRP. But it's a bloody nuisance to configure, 'cuz you need to configure two different HSRP groups for the same pair of switches, then manually assigning different hosts/PCs to use different default gateways. As I said, bloody nuisance.
HTH.

Jms in cluster / load balancing and failover

          Did I get it right ???
          I have 1 admin server and 4 managed servers in a 2 clusters, a development cluster
          and a test cluster.
          I now want to have loadbalancing with my jms server and I want to be able to migrate
          my jms server in case of failer.
          for each cluster I have created
          connectionFactory targeted to the cluster
          Distributed destination with 2 queue members
          One JMSServer migratable targeted on managed server 1, with destination 1 from
          the ditributed destination
          One JMSServer migratable targeted on managed server 2, with destination 2 from
          the ditributed destination
          I expect this to make loadbalancing between the 2 servers in the cluster, and
          I can migrate the jms server if one of the server fails to the running server.
          One thing is now.....If one server fails and I migrate the jms server to the other
          server that is running, and I then restart the server that was down, what is then
          happening, do I then have 3 jms servers ???
          [config.xml]


          "Kris" <[email protected]> wrote in message news:[email protected]...
          >
          > "Kawaljit Singh Sunny" <[email protected]> wrote:
          > >
          > >"Kris" <[email protected]> wrote in message
          > >news:[email protected]...
          > >>
          > >> Did I get it right ???
          > >>
          > >> I have 1 admin server and 4 managed servers in a 2 clusters, a
          development
          > >cluster
          > >> and a test cluster.
          > >> I now want to have loadbalancing with my jms server and I want to be
          > >able
          > >to migrate
          > >> my jms server in case of failer.
          > >>
          > >> for each cluster I have created
          > >> connectionFactory targeted to the cluster
          > >> Distributed destination with 2 queue members
          > >> One JMSServer migratable targeted on managed server 1, with destination
          > >1
          > >from
          > >> the ditributed destination
          > >> One JMSServer migratable targeted on managed server 2, with destination
          > >2
          > >from
          > >> the ditributed destination
          > >
          > >If the server where your JMSConnections are loadBalanced to goes down,
          > >the
          > >producers and consumers using this JMSConnection are closed.
          > >You have to recreate these producers and consumers.
          > >If the server where your Destination resides goes dow, the consumers
          > >are
          > >closed.
          > >If the producers JMSConnection is not on this server, the producer stays
          > >up.
          > >
          > >>
          > >>
          > >> I expect this to make loadbalancing between the 2 servers in the
          cluster,
          > >and
          > >> I can migrate the jms server if one of the server fails to the running
          > >server.
          > >>
          > >> One thing is now.....If one server fails and I migrate the jms server
          > >to
          > >the other
          > >> server that is running, and I then restart the server that was down,
          > >what
          > >is then
          > >> happening, do I then have 3 jms servers ???
          > >
          > >No you still have 2 JMSServers. JMS Migration is manual.
          > >
          > >>
          >
          > you say : No you still have 2 JMSServers. JMS Migration is manual.
          >
          > But if I manual migrate the jmsserver that was down to the running wls
          server,
          > that already have one jms server running, this wls server must then have 2
          jms
          > servers. And I boot the wls server that hosted the jms server that was
          down, this
          > will now have a running jms server. isn't that 3 jms servers ?
          Once you migrate a JMSServer from a WeblogicServer1 to WeblogicServer2,
          and then you boot WeblogicServer1, this JMSServer which was migrated should
          NOT be on WeblogicServer1.
          (You have migrated the JMSServer from WeblogicServer1 to WeblogicServer2)
          >
          > But I was thinking about that I could spare the migration part. If I have
          2 wls
          > servers and a jms server on each of them, and a destributed destination
          with 2
          > queue members that are persistent in a database. If a wls og just a jms
          server
          > goes down, I just have to reboot the server and it will run again. This
          way I
          > dont have to think about migration, or what ?
          Yes that is true.
          Irrespective of whether you have migration or not,
          only thing you need to do take care is to reconnect to weblogic server, if
          the the server where your JMSConnection is loadBalanced to goes down.
          There is no failover of JMSConnections. Producers inside this JMSConnection
          will be closed. You will have to create a new JMSConnection and a new
          Producer and continue with your production of JMS Messages.
          -sunny

Load Balancing and Failover in Dual Ethernet

I have a cisco 2911/K9 router with two 4Mbps Leased line connection from two different ISPs to my remote office. Remote office has cisco 2811 router
Main office has MPLS connection with static Ip routing apart from the two leased lines
All handoffs are ethernet
Is it possible to do load sharing as well as fail over between the two ISPs, if so how am i to achieve that
Kindly help me

Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
If your MPLS vendor supports no dynamic routing, they why do you ask about BGP? Or, do they only support dynamic routing with BGP?
You can do equal cost multi-path with BGP (may require a hidden command to fully utilize).
You could do GRE tunnels across the MPLS cloud and dynamically route between them using your choice of a dynamic routing protocol.
Both your devices should support OER/PfR (may require a feature upgrade). OER/PfR will actually dynamically load balance.
SLA features should also be available on both your routers, those too might require a IOS feature upgrade.
Configuration examples might be found on Cisco's main web site.

Webdispatcher with load-balancing and local icm configuration

Dear all,
we have a webdispatcher in place to load-balance the requests to our three application servers (ABAP). The webdispatcher has a hostname that is resolved by the DNS. The application servers use local hostnames that can't be resolved in our infrastructure at the moment.
That configuration works fine for all requests that reach the infrastructure from outside.
But, if a user logs on with SAP GUI and then calls a web transaction like a webdynpro, the browser opens with the local application server hostname. However that hostname can't be resolved.
Is it possible to change the three local icm configurations in a way that all local generated web requests refers to the webdispatcher too? What is the preferred configuration for this more or less usual scenario?
Thanks in advance,
Daniel

Hello,
In your scenario, I would try to create entries in table HTTPURLLOC. After that the 3 app servers should generate URLs using the SAP Web Dispatcher.
Best Regards,
Olivier

Need help with load balancing and DNS proxy

Hi,
I need help on how to configure my router so it will work with my DNS proxy and load balancing.
I have a Linksys LRT224 router. I have two broadband connections from two separate ISPs,500Mbps each (WAN1 & WAN2). WAN1 has a static IP and WAN2 is dynamic assigned. I use Unlocator (www.unlocator.com) so I can access geographically restricted sites (Pandora, Netflix, etc.).
The problem I have is that unlocator registers only one IP address (WAN1 address) and since I am doing load balancing I have no way of knowing if the DNS request will go through the registered IP (WAN1) or through the other (WAN2). I am not an expert in routing or networking but I'm guessing I have a way of configuring the router so all the DNS requests go out through WAN1, right?
In the router's Dual WAN config page there is a section for Protocol Binding. I tried to configure but only managed to screw up the internet at home. I used:
DNS[UDP/53-53]->192.168.1.1-192.168.1.254(0.0.0.0-0.0.0.0)WAN2
Any help or suggestions are appreciated.
Alex

Good solution though. That's probably the only way you could do true Load Balancing anyway.
Please remember to Kudo those that help you.
Linksys
Communities Technical Support

Distributed services with load balancing and failover?

Similar Messages

Maybe you are looking for