Re: RAC or Oracle Fail Safe
Title: RE: RAC or Oracle Fail Safe Hi! Of course you have to plan your servers' capacity accordingly, that in the event of node failure, the other node will not get too loaded. When one node crashes, the second one has to deal with queries and transactions of both servers and we must not forget, that also rollback of failed nodes uncommitted transactions has to be done! Of course when you got more nodes, then the impact of one node failure would be smaller.. Tanel. Here I would beg to differ. RAC is more safe. _Almost_ half of your online user (who are on surviving node) wont even notice the node failure when one node in a RAC environment goes down. And those who were unlucky users (connected to failed node), can connect immediately to the surviving node, without any delay. Of course you need to configure these parameters manually. While in OFS environment your users have to wait till the time surviving node brings up the database and all related services (listener etc) completely. The price you may for these features is more money and more complex database environment to manage.
RE: RAC or Oracle Fail Safe
Title: RE: RAC or Oracle Fail Safe Hi Rajesh, Thanks for the detailed reply and I really appreciate that. Certain clarifications. 1. With reference to what you said about using both the nodes, you meant that we can install another database (lets say for reporting purpose) or/and also 9iAS on one node and our main production database on the other 2. Incase of OFS, the database service fails over from one node to the other, and then it starts, but it is automatic, right? 3. When you talk about Resources and that they cant be shared, you are not referring to the shared storage vault (we are using DAS Direct Attached Storage) because that has to be at least available to both the servers. And when the database fails over from one server to the other, and it starts its services on the other server, then the obviously that other server will make use of the DAS on which reside our datafiles. 4. RAC costs roughly 20k$ per CPU and it doesnt come bundled with Oracle 9i License? (For confirmation purposes) 5. From various mails I gather that OFS is basically a safe bet, if the hardware supporting is good. And if, God-forbid, one server goes down then the only time lapse is the time it takes to start the services on the other server. 6. The transactions taking place at the time when main server goes down will be lost, I mean it will give a feeling of hung database or some other error. Thanks in advance, and waiting for your reply. Regards, Hussain -Original Message- From: RAJESH DAYAL [mailto:[EMAIL PROTECTED] Sent: Thursday, July 17, 2003 12:30 PM To: Multiple recipients of list ORACLE-L Subject: RE: RAC or Oracle Fail Safe Hi Hussain ! RAC can have Active-Active or Active-Passive combination for two nodes. While OFS can't give you Active-Active combination of two nodes. Basically a resource sharing is not possible in Oracle Fail safe so one resource can be used by only one node at one point of time. ( Resource means Hard Disk Space, Oracle Service, Listener etc). RAC can give Transparent Application Failover TAF which is out of consideration in OFS. In case of node failure on RAC System, the user will just need to attempt relogin and he will be connected _immediately_ to another surviving instance ( If tnsnames.ora is configured properly). While there is a significant amount of delay in case of OFS, the Database service has to failover from one node to another node and this may take it's own _sweet_ amount of time based on amount and type of activity being performed on the database. Of course both the systems are screwed when your (shared) storage system is down ;-) Lastly why would someone want to use only 1 out of 2 nodes when it is possible to use both the nodes? One possible answer is that, on one node you configure One server (say 9iAS) and on another node you configure database. Then under normal condition they will run on their respective nodes and when one fails they would switch-over to the surviving node. So your performance may suffer for the time during which any node is down. So you are the best judge to decide now ;-) Just my .2 c Regards, Rajesh -Original Message- HREF="" PROTECTED]mailto:[EMAIL PROTECTED] On Behalf Of Hussain Ahmed Qadri Sent: Wednesday, July 16, 2003 8:44 PM To: Multiple recipients of list ORACLE-L Subject: RAC or Oracle Fail Safe Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC or Oracle Fail Safe
Title: RE: RAC or Oracle Fail Safe Hi Hussain, Replies are inline . With reference to what you said about using both the nodes, you meant that we can install another database (lets say for reporting purpose) or/and also 9iAS on one node and our main production database on the other Yes you can do that. You can very well make use of other node, in above ways. Advantage of running some standard services would be that they will also fail-over to the surviving node. So _effectively_ you are using both the nodes at any point of time. Here we have 2 node cluster where one node is running Database and another node is running 9ias Server. and they failover to each other automatically if one of them them goes down. In fact you can decide your failback policies also so that the nodes know when (may be off-peak hour) to fail-back once the failed node is back. Incase of OFS, the database service fails over from one node to the other, and then it starts, but it is automatic, right? Yes, it is a configurable parameter and can be set to automatic mode. When you talk about Resources and that they cant be shared, you are not referring to the shared storage vault (we are using DAS Direct Attached Storage) because that has to be at least available to both the servers. And when the database fails over from one server to the other, and it starts its services on the other server, then the obviously that other server will make use of the DAS on which reside our datafiles. I think you got me wrong here. They cant be shared at the same point of time. Actually the shared storage would be physically connected to both the nodes. And they can/will have multiple drives defined on them ( say U,V,W,X).So at one point of time say UV would be owned by node1 and WX would be owned by node2. If node1 fails then node2 will own all the 4 drives U,V,W and X. RAC costs roughly 20k$ per CPU and it doesnt come bundled with Oracle 9i License? (For confirmation purposes) Absolutely right. From various mails I gather that OFS is basically a safe bet, if the hardware supporting is good. And if, God-forbid, one server goes down then the only time lapse is the time it takes to start the services on the other server. Here I would beg to differ. RAC is more safe. _Almost_ half of your online user (who are on surviving node) wont even notice the node failure when one node in a RAC environment goes down. And those who were unlucky users (connected to failed node), can connect immediately to the surviving node, without any delay. Of course you need to configure these parameters manually. While in OFS environment your users have to wait till the time surviving node brings up the database and all related services (listener etc) completely. The price you may for these features is more money and more complex database environment to manage. The transactions taking place at the time when main server goes down will be lost, I mean it will give a feeling of hung database or some other error. I think users will have a hung screen with/without some ORA error message. Hope this helps you decide, Regards, Rajesh -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]On Behalf Of Hussain Ahmed Qadri Sent: Saturday, July 19, 2003 11:44 AM To: Multiple recipients of list ORACLE-L Subject: RE: RAC or Oracle Fail Safe Hi Rajesh, Thanks for the detailed reply and I really appreciate that. Certain clarifications. 1. With reference to what you said about using both the nodes, you meant that we can install another database (lets say for reporting purpose) or/and also 9iAS on one node and our main production database on the other 2. Incase of OFS, the database service fails over from one node to the other, and then it starts, but it is automatic, right? 3. When you talk about Resources and that they cant be shared, you are not referring to the shared storage vault (we are using DAS Direct Attached Storage) because that has to be at least available to both the servers. And when the database fails over from one server to the other, and it starts its services on the other server, then the obviously that other server will make use of the DAS on which reside our datafiles. 4. RAC costs roughly 20k$ per CPU and it doesnt come bundled with Oracle 9i License? (For confirmation purposes) 5. From various mails I gather that OFS is basically a safe bet, if the hardware supporting is good. And if, God-forbid, one server goes down then the only time lapse is the time it takes to start the services on the other server. 6. The transactions taking place at the time when main server goes down will be lost, I mean it will give a feeling of hung database or some other error. Thanks in advance, and waiting for your reply. Regards, Hussain -Original Message- From: RAJESH DAYAL [mailto:[EMAIL PROTECTED] Sent: Thursday, July 17, 2003 12:30 PM To: Multiple recipients
RE: RAC or Oracle Fail Safe
Thank you to all those who replied and who intend to reply to this one :) We are using Dell PE4600 servers. My concern was that I read it somewhere that if one of the node goes down, then the clients have to restart the application to log in again to the database, is that true? And is it any different on RAC? We have the licensed RAC and Oracle Fail Safe CDs, so that is not an issue. I also believe that setting up RAC is more complex, so we wanted to confirm that if Fail Safe gives us HA, 24x7, and any change in node shifting (when active is down, passive becomes active) is transparent to the user then we should go for Fail safe. By the way, we intend to keep cluster settings of ACTIVE and PASSIVE, is there any issue regarding this? Tanel Poder: you said that, in event of failure you can start up instance on another node that will mount, recover and open the database. Doesn't the second node automatically take over? And this means that the user has to log in again, do I get it right? Best Regards, Hussain -Original Message- Sent: Wednesday, July 16, 2003 10:44 PM To: Multiple recipients of list ORACLE-L Hi! RAC - One database, two (or more) instances servicing it concurrently. If one node crashes, second one starts recovering, your uncommitted transactions and session state variables (package variables) on failed instance are lost. FailSafe - One database, one instance servicing it at any time, in event of failure you can start up instance on another node that will mount, recover and open the database. With RAC you can have some kind of transparent failover which is able to reexecute your queries on survived instance, but AFAIK for uncommitted transaction survivability your app has to be coded to support it. Also, with RAC you may scale your system up to support more users in some cases. RAC is additional $20k per CPU, FailSafe comes with EE, I believe. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 16, 2003 7:44 PM Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC or Oracle Fail Safe
Title: RE: RAC or Oracle Fail Safe Hi Hussain ! RAC can have Active-Active or Active-Passive combination for two nodes. While OFS can't give you Active-Active combination of two nodes. Basically a resource sharing is not possible in Oracle Fail safe so one resource can be used by only one node at one point of time. ( Resource means Hard Disk Space, Oracle Service, Listener etc). RAC can give Transparent Application Failover TAF which is out of consideration in OFS. In case of node failure on RAC System, the user will just need to attempt relogin and he will be connected _immediately_ to another surviving instance ( If tnsnames.ora is configured properly). While there is a significant amount of delay in case of OFS, the Database service has to failover from one node to another node and this may take it's own _sweet_ amount of time based on amount and type of activity being performed on the database. Of course both the systems are screwed when your (shared) storage system is down ;-) Lastly why would someone want to use only 1 out of 2 nodes when it is possible to use both the nodes? One possible answer is that, on one node you configure One server (say 9iAS) and on another node you configure database. Then under normal condition they will run on their respective nodes and when one fails they would switch-over to the surviving node. So your performance may suffer for the time during which any node is down. So you are the best judge to decide now ;-) Just my .2 c Regards, Rajesh -Original Message- HREF="">mailto:[EMAIL PROTECTED]] On Behalf Of Hussain Ahmed Qadri Sent: Wednesday, July 16, 2003 8:44 PM To: Multiple recipients of list ORACLE-L Subject: RAC or Oracle Fail Safe Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC or Oracle Fail Safe
Title: RE: RAC or Oracle Fail Safe Answers in line... -- I also believe that setting up RAC is more complex, That's not totally true, For a DBA it shouldn't matter much, whether he is setting up RAC or OFS. But yes relatively RAC is a bit complex to manage. -- and any change in node shifting (when active is down, passive becomes active) is transparent to the user Not true really, user's session will flash error message and he will have to re login. This is true in both the cases(unless you implement TAF Transparent Application Failover logics in your application for RAC. One major difference would be that in RAC user can re-login immediately, while in OFS he has to wait till the other node bring the database up. -- Doesn't the second node automatically take over? Yes other node can be configured to take over automatically base on some simple in-built logic. But a DB recovery might be required depending on specific cases. -- And this means that the user has to log in again, do I get it right? Yes you are right here. HTH, Rajesh -Original Message- HREF="">mailto:[EMAIL PROTECTED]] On Behalf Of Hussain Ahmed Qadri Sent: Thursday, July 17, 2003 11:04 AM To: Multiple recipients of list ORACLE-L Subject: RE: RAC or Oracle Fail Safe Thank you to all those who replied and who intend to reply to this one :) We are using Dell PE4600 servers. My concern was that I read it somewhere that if one of the node goes down, then the clients have to restart the application to log in again to the database, is that true? And is it any different on RAC? We have the licensed RAC and Oracle Fail Safe CDs, so that is not an issue. I also believe that setting up RAC is more complex, so we wanted to confirm that if Fail Safe gives us HA, 24x7, and any change in node shifting (when active is down, passive becomes active) is transparent to the user then we should go for Fail safe. By the way, we intend to keep cluster settings of ACTIVE and PASSIVE, is there any issue regarding this? Tanel Poder: you said that, in event of failure you can start up instance on another node that will mount, recover and open the database. Doesn't the second node automatically take over? And this means that the user has to log in again, do I get it right? Best Regards, Hussain -Original Message- Sent: Wednesday, July 16, 2003 10:44 PM To: Multiple recipients of list ORACLE-L Hi! RAC - One database, two (or more) instances servicing it concurrently. If one node crashes, second one starts recovering, your uncommitted transactions and session state variables (package variables) on failed instance are lost. FailSafe - One database, one instance servicing it at any time, in event of failure you can start up instance on another node that will mount, recover and open the database. With RAC you can have some kind of transparent failover which is able to reexecute your queries on survived instance, but AFAIK for uncommitted transaction survivability your app has to be coded to support it. Also, with RAC you may scale your system up to support more users in some cases. RAC is additional $20k per CPU, FailSafe comes with EE, I believe. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 16, 2003 7:44 PM Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting services --
RE: RAC or Oracle Fail Safe
One other note is that we have not had to failover due to a Win2K of Oracle problem in 18 months of running failsafe. We have found it to be extremely stable just not scalable froma CPU standpoint. I totally agree with this. I am running FailSafe here also. The *only* failover's I experience is when the morons running the data center install software on the primary server and reboot the box. Failsafe brings the DB up on the other box without any problems. Works great. Tom Mercadante Oracle Certified Professional -Original Message- Sent: Wednesday, July 16, 2003 5:59 PM To: Multiple recipients of list ORACLE-L FailSafe comes with EE and works very well. It might even come with SE but I am not sure. Our production environment fails over in less than 2 minutes. It is much simpler to set up ( ie no SAN, raw devices or OCFS) and a heck of a lotr cheaper ( 20K$ / CPU for RAC ). One other note is that we have not had to failover due to a Win2K of Oracle problem in 18 months of running failsafe. We have found it to be extremely stable just not scalable froma CPU standpoint. -Original Message- Sent: Wednesday, July 16, 2003 9:44 AM To: Multiple recipients of list ORACLE-L Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tony Johnson INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Mercadante, Thomas F INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Re: RAC or Oracle Fail Safe
Hi! RAC - One database, two (or more) instances servicing it concurrently. If one node crashes, second one starts recovering, your uncommitted transactions and session state variables (package variables) on failed instance are lost. FailSafe - One database, one instance servicing it at any time, in event of failure you can start up instance on another node that will mount, recover and open the database. With RAC you can have some kind of transparent failover which is able to reexecute your queries on survived instance, but AFAIK for uncommitted transaction survivability your app has to be coded to support it. Also, with RAC you may scale your system up to support more users in some cases. RAC is additional $20k per CPU, FailSafe comes with EE, I believe. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 16, 2003 7:44 PM Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
RE: RAC or Oracle Fail Safe
FailSafe comes with EE and works very well. It might even come with SE but I am not sure. Our production environment fails over in less than 2 minutes. It is much simpler to set up ( ie no SAN, raw devices or OCFS) and a heck of a lotr cheaper ( 20K$ / CPU for RAC ). One other note is that we have not had to failover due to a Win2K of Oracle problem in 18 months of running failsafe. We have found it to be extremely stable just not scalable froma CPU standpoint. -Original Message- Sent: Wednesday, July 16, 2003 9:44 AM To: Multiple recipients of list ORACLE-L Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tony Johnson INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
Re: RAC or Oracle Fail Safe
EE is not required to FailSafe. It comes at no cost with Standard Edition as well. Tanel Poder tanel.poder.00To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] [EMAIL PROTECTED] cc: Sent by: Subject: Re: RAC or Oracle Fail Safe [EMAIL PROTECTED] ty.com 07/16/2003 12:44 PM Please respond to ORACLE-L Hi! RAC - One database, two (or more) instances servicing it concurrently. If one node crashes, second one starts recovering, your uncommitted transactions and session state variables (package variables) on failed instance are lost. FailSafe - One database, one instance servicing it at any time, in event of failure you can start up instance on another node that will mount, recover and open the database. With RAC you can have some kind of transparent failover which is able to reexecute your queries on survived instance, but AFAIK for uncommitted transaction survivability your app has to be coded to support it. Also, with RAC you may scale your system up to support more users in some cases. RAC is additional $20k per CPU, FailSafe comes with EE, I believe. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 16, 2003 7:44 PM Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling
Re: RAC or Oracle Fail Safe
And to add to that, I don't believe that FailSafe is even required to use Oracle with MS Clusters. IIRC, it's just to simplify the cluster setup. FailSafe certainly isn't need for setting up failover withi Veritas Cluster Server. Jared Allen R. Lucas [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 07/16/2003 10:59 AM Please respond to ORACLE-L To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] cc: Subject:Re: RAC or Oracle Fail Safe EE is not required to FailSafe. It comes at no cost with Standard Edition as well. Tanel Poder tanel.poder.00To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] [EMAIL PROTECTED] cc: Sent by: Subject: Re: RAC or Oracle Fail Safe [EMAIL PROTECTED] ty.com 07/16/2003 12:44 PM Please respond to ORACLE-L Hi! RAC - One database, two (or more) instances servicing it concurrently. If one node crashes, second one starts recovering, your uncommitted transactions and session state variables (package variables) on failed instance are lost. FailSafe - One database, one instance servicing it at any time, in event of failure you can start up instance on another node that will mount, recover and open the database. With RAC you can have some kind of transparent failover which is able to reexecute your queries on survived instance, but AFAIK for uncommitted transaction survivability your app has to be coded to support it. Also, with RAC you may scale your system up to support more users in some cases. RAC is additional $20k per CPU, FailSafe comes with EE, I believe. Tanel. - Original Message - To: Multiple recipients of list ORACLE-L [EMAIL PROTECTED] Sent: Wednesday, July 16, 2003 7:44 PM Hello all, We have setup a Microsoft Cluster Server (MSCS) cluster, with two nodes, using W2K, and now are in the process of deploying Oracle 9i on it. For the purpose of high availability (HA), we are deliberating on setting up either RAC or Oracle Fail Safe. The confusion is over the fact that if we get HA with oracle Fail Safe, i.e. if one node is down due to any problem, then the other takes over, then can we do without RAC? Which of these two is more transparent to the user, i.e. which will take less time to shift the load from one node (server) to the other node? Can any one explain the benefits of using RAC over Oracle Fail Safe, or vice versa. Regards, Hussain DBA SKMCHRC -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Hussain Ahmed Qadri INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Tanel Poder INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: Allen R. Lucas INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services - To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.net -- Author: INET: [EMAIL PROTECTED] Fat City Network Services-- 858-538-5051 http://www.fatcity.com San Diego, California-- Mailing list and web hosting services
