Hello,
I am backing up SQL Server VMs in a Hyper-V environment and have several questions regarding the use of persistent agents and application-aware processing:
1. Persistent Agent Installation and Usage:
- If I enable the "Use persistent guest agent" option in the application-aware processing settings, will the persistent agent automatically install during the first job run?
- How does this differ from manually installing the agent using the Veeam Installer Service?
2. Transaction Log Backups:
- When configuring transaction log backups every 15 minutes, does this process rely on the persistent agent? Does it involve running the full backup job every 15 minutes, or is it limited to just backing up the transaction logs?
3. Application-Aware Processing:
- Does using a persistent agent on SQL VMs provide better application-aware processing compared to non-persistent runtime components? Specifically, does it improve handling of SQL Server transactional logs and ensure better consistency for backups?
4. Recovery Capabilities:
- Does having a persistent agent installed enhance recovery options (e.g., point-in-time recovery or database/table-level recovery) compared to using non-persistent components?
5. Persistent vs Non-Persistent Components:
- Can you clarify the differences between persistent agent components and non-persistent runtime components in terms of functionality, performance, and use cases for SQL Server backups? Are there any specific scenarios where one is preferred over the other?
6. Best Practices:
- For production SQL VMs, would you recommend always using a persistent agent for optimal backup and recovery of databases and transaction logs? If not, under what circumstances would non-persistent components be sufficient?"
Thank you so much
Best regards
-
- Enthusiast
- Posts: 45
- Liked: 1 time
- Joined: May 30, 2023 12:58 pm
- Contact:
-
- Veeam Software
- Posts: 2604
- Liked: 608 times
- Joined: Jun 28, 2016 12:12 pm
- Contact:
Re: Hyper-V SQL Backup questions
Hi joshima,
Quick question and then I will answer on persistent vs non-persistent guest processing agents.
As I see you are doing a lot of research and planning for your SQL log backups, but your questions about the Persistent vs Non-Persistent Guest Processing agents seem to indicate you're actually asking about the pros and cons of an image level backup for your SQL servers versus use of plugins for Enterprise Applications.
Your questions seem to be more about having a dedicated process installed on the source server as opposed to our Persistent vs Non-Persistent guest processing agents, which are for a fairly specific operation for image level backups. Can you confirm if I'm understanding your inquiry correctly? Do the plugins seem more like what you're interested in?
To address your questions on the guest processing agents, I will link the User Guide page for Persistent Agents, as it covers a lot of these questions.
Persistent Agents are primarily about security considerations, namely in that it greatly reduces the amount of ports:
For your questions:
1. As per User Guide, if the option is checked, on first run the Persistent Agent is installed. Fundamentally there's no difference between manual installation and automatic, but let Veeam do it for you.
2. Persistent Agents can be used in logshipping, correct. Log Shipping runs as a "child" job outside of the normal job. The log shipping job will pause itself whenever the parent job runs, so no special configuration required. The interval specified is just how frequently to check for new transaction logs, this does not require a full backup on every interval. Each interval will be "tied" to the parent job backup which has a full backup of the database. (Full here refers to VSS Full vs VSS Copy, not a VBK file. So it's not a full _veeam_ backup, it's a full database backup)
3. No difference, see the quoted user guide section above
4. No difference, it's purely a security consideration as noted in the quoted user guide section above
5. No specific situation except for security concerns about required ports for using non-persistent agents
6. No specific recommendations, this is a decision that you and your team should make based on your security and networking interests
Quick question and then I will answer on persistent vs non-persistent guest processing agents.
As I see you are doing a lot of research and planning for your SQL log backups, but your questions about the Persistent vs Non-Persistent Guest Processing agents seem to indicate you're actually asking about the pros and cons of an image level backup for your SQL servers versus use of plugins for Enterprise Applications.
Your questions seem to be more about having a dedicated process installed on the source server as opposed to our Persistent vs Non-Persistent guest processing agents, which are for a fairly specific operation for image level backups. Can you confirm if I'm understanding your inquiry correctly? Do the plugins seem more like what you're interested in?
To address your questions on the guest processing agents, I will link the User Guide page for Persistent Agents, as it covers a lot of these questions.
Persistent Agents are primarily about security considerations, namely in that it greatly reduces the amount of ports:
So the main point is not about functionality or performance, but about securely allowing Application Consistent backups.If you want to perform guest processing in a highly secure way, you can use persistent agent components (Guest Helper, Log Shipping Service) on protected VMs. Persistent agent components require very limited and clearly defined ports to communicate with Veeam Backup & Replication. For more information about ports, see Guest Processing Components.
For your questions:
1. As per User Guide, if the option is checked, on first run the Persistent Agent is installed. Fundamentally there's no difference between manual installation and automatic, but let Veeam do it for you.
2. Persistent Agents can be used in logshipping, correct. Log Shipping runs as a "child" job outside of the normal job. The log shipping job will pause itself whenever the parent job runs, so no special configuration required. The interval specified is just how frequently to check for new transaction logs, this does not require a full backup on every interval. Each interval will be "tied" to the parent job backup which has a full backup of the database. (Full here refers to VSS Full vs VSS Copy, not a VBK file. So it's not a full _veeam_ backup, it's a full database backup)
3. No difference, see the quoted user guide section above
4. No difference, it's purely a security consideration as noted in the quoted user guide section above
5. No specific situation except for security concerns about required ports for using non-persistent agents
6. No specific recommendations, this is a decision that you and your team should make based on your security and networking interests
David Domask | Product Management: Principal Analyst
Who is online
Users browsing this forum: No registered users and 6 guests