![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#3
| |||
| |||
|
|
You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode—only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object’s ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#4
| |||
| |||
|
|
Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode—only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object’s ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#5
| |||
| |||
|
|
No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. |
|
No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Clustering works on OS level so u can use clustering and log shipping together. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Rogers" wrote: Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode—only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object’s ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#6
| |||
| |||
|
|
Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? |
|
Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode—only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object’s ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#7
| |||
| |||
|
|
No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Are you sure about that? Note that even though they both works on transaction logs, log shipping works on the transaction log backups, which are independent from database mirroring. Linchi "Manpreet Singh" wrote: No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Clustering works on OS level so u can use clustering and log shipping together. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Rogers" wrote: Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode—only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object’s ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP –DBA, DD - SQL server 2005 MCTS – SQL server 2005 MCTS –SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#8
| |||
| |||
|
|
Linchi Shea is right. Maybe I get confused somewhere. Its perfectly safe to take transaction log backup in database mirroring scenario. which means that you can use log shipping with database mirroring. . Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Linchi Shea" wrote: No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Are you sure about that? Note that even though they both works on transaction logs, log shipping works on the transaction log backups, which are independent from database mirroring. Linchi "Manpreet Singh" wrote: No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Clustering works on OS level so u can use clustering and log shipping together. Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Rogers" wrote: Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode-only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object's ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
#9
| |||
| |||
|
|
Linchi Shea is right. Maybe I get confused somewhere. Its perfectly safe to take transaction log backup in database mirroring scenario. which means that you can use log shipping with database mirroring. . Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Linchi Shea" wrote: No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Are you sure about that? Note that even though they both works on transaction logs, log shipping works on the transaction log backups, which are independent from database mirroring. Linchi "Manpreet Singh" wrote: No, database mirroring and log shipping both works on transaction logs. Which, means that you can use only one method at a time. Clustering works on OS level so u can use clustering and log shipping together. Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Rogers" wrote: Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode-only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object's ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
|
Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Why would you want to do that regardless whether it can or cannot be done? I can see reasons for doing both if they are for different destinations. But you seem to talking about the same destination. Note that failover clustering protects you from failed host, whereas database mirroring can also protect you from failed storage. Because they protect against different kinds of disasters, it may make sense to use both. Linchi "Rogers" wrote: Thanks Manpreet. As you mentioned SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. Can I use Database Mirroring and Log Shipping together to achieve automatic failover and disaster recovery at one time? Thanks "Manpreet Singh" wrote: You are using a high performance mode (asynchronously) of database mirroring. The witness serves serve no purpose in high performance mode-only the principal and mirror are needed. In fact, adding the witness to a high performance mode configuration will add risk since its addition to this setup will enforce quorum, which requires two or more SQL Server instances. I would not use a witness in this scenario, as it provides no value. If quorum is lost, there are two scenarios: when the principal is down, you will only be able to manually force the mirror to become the new primary if it can connect to the witness; and when the mirror is down, the principal will be taken offline if it cannot connect to the witness. Therefore, it is recommended that if you plan on implementing high performance mode, you do not configure a witness. You can use High Safety Mode (synchronously) for automatic failover. You need to code your application to specifically know about database mirroring to be able to handle the primary and mirror. First, you will need the version of MDAC (or later) that is compatible with SQL Server 2005 as well as either SQL Server Native Client or ADO.NET. If you are coding your own application, use the SqlConnection object's ConnectionString. Also, the connection string has to contain not only the primary but the mirror. The connection string also assumes the same database name on each partner. The following is a sample connection string: "Server=My_Instance; Failover Partner=fo_my_Instance;Database=myDB;Network=dbmss ocn" As per your 2nd question, you can use SQL server clustering and log shipping together to achieve automatic failover and disaster recovery at one time. The only new thing you have to do is to configure file share for log shipping as a resource in windows cluster service. Manpreet Singh http://crazysql.wordpress.com/ MCITP -DBA, DD - SQL server 2005 MCTS - SQL server 2005 MCTS -SharePoint Server 2007 "Noor" wrote: Dear Professional, I would really appreciate if anyone can help me out. Currently I have database mirroring but I am using Asynchronous mode and using manual failover strategy but going forward I have to implement following strategy. 1. Automate failover Question: if I go with automate failover that means I have to configure separate Witness Server and application will point to Witness Server right? and witness will responsible for redirecting to active server right? 2. Disaster Recovery: Question: For disaster recovery I have to implement Log shipping offsite but the problem is if I go with Automate failover then I can't implement Log shipping for DR, can you tell me any alternate solution that provide both solution: Automate Failover as well as DR offsite. Thanks in advance. |
![]() |
| Thread Tools | |
| Display Modes | |
| |