veeam

How to avoid typical misconfigurations when setting up Veeam

This article is aimed at giving you a smooth start with Veeam Backup & Replication. It includes some basic advice on the initial setup, and outlines the most common misconfigurations that we, at Veeam Support, find in clients’ infrastructures during our investigations.

Recommendations on Veeam backup modes

In most cases, forward incremental or forever forward incremental backup modes are recommended as the fastest ones. Forever forward incremental (no periodic full backup) requires less space and offers decent performance. Forward incremental requires more space, but is also more robust (because a backup chain is further divided in subchains by periodic full backup).

Reverse incremental backup method is our oldest backup method and consequently the slowest. Depending on the type of storage in use, it can be three or more times slower than other modes. With the reverse incremental backup, you get a full backup as the last point in the chain. This allows for faster restores in case the most recent point is used, but the difference is often negligible in comparison to a forward incremental chain (if its length is not unreasonably long, we usually suggest it to be around 30 days).

Insights on the full backup

Synthetic full operation builds a full backup file from the restore points already residing in your repository. However, not every storage type provides a good performance with synthetic operations, so we advise to use active full backup as an alternative.

When you set up a synthetic full backup mode, there is an additional “Transform previous backup chains into rollbacks” option available. Keep in mind though that this option starts a task of transforming incremental backups (.VIB) into rollbacks (.VRB), which is very laborious for your target backup repository. For example, it will help you transform your current chain into the reverse incremental one for archival purposes. However, if you use it as a main backup method, it would produce a very specific backup chain consisting of a full backup file and a mix of forward and reverse incremental restore points.

Veeam - A forward incremental backup job with periodic synthetic full

Figure 1. A forward incremental backup job with periodic synthetic full.

Guest processing tips

Guest processing is used to create consistent backups of your VMs. And if they run instances of Microsoft Exchange, Active Directory, SharePoint, SQL Server and Oracle applications, you will be able to leverage granular restores using Veeam Explorers. Please note that guest processing relies on a VSS framework (a Windows feature), which should be functioning correctly, otherwise your backup jobs will fail.

To enable guest processing, go to Guest Processing of backup job properties. You should enable “Application-aware processing” option and you should provide an administrative account under guest OS credentials.

Figure 2. Guest processing step controls application-aware processing and indexing.

If some of VMs in the job require specific credentials, you can set them by clicking on the “Credentials” button. This brings up the Credentials menu. Click on “Set User…” to specify the credentials that should be used with the VM.

Veeam - Credentials menu

Figure 3. Credentials menu allows to set up users for each VM in the job.

Clicking on the “Applications…” button brings up a menu where you can specify options for supported applications and disable the guest processing for certain VMs, if needed.

Veeam - guest processing

Figure 4. In Applications menu, you can specify options for various application or disable guest processing completely for a VM.

VM guest file system indexing

With “VM Guest File System Indexing” enabled, Veeam Backup & Replication creates a catalog of files inside the VM, allowing you to use guest file search and perform 1-click restores through our Veeam Backup Enterprise Manager.

In case you don’t use the Enterprise Manager, then you can cut some (sometimes significant) time off your backup window and save space on the C: drive of a Veeam server by disabling this option. It doesn’t affect your ability to perform file level restores from your Veeam Backup & Replication console.

Secondary backup destination

No storage vendor can guarantee an absolute data integrity. Veeam checks a backup file once it’s written to a disk, but, with millions of operations happening on the datastore, occasional bits may get swapped causing silent corruption. Veeam Backup & Replication provides features like SureBackup and health checks that help detect an early corruption. However, sometimes it may be already too late, so it’s absolutely necessary to follow the 3-2-1 rule and use different sets of media in several locations to guarantee data Availability.

To maintain the 3-2-1 rule, right after creating a primary backup job, it’s advised to set up a secondary copy job. This can be a Backup Copy Job to a secondary storage, Backup Copy Job to a cloud repository or a copy to tape.

Instant VM recovery as it should be

Instant VM Recovery allows you to start a VM in minimal time right from a backup file. However, you need to keep in mind that a recovered VM still sits in your backup repository and consumes its resources. To finalize the restore process, the VM must be migrated back to the production. Too often we at Veeam Support see critical VMs working for weeks in the Instant VM Recovery mode until a datastore fills up and data is lost.

For those of you looking for a deep dive on the topic, I recommend the recent blog post on Instant VM Recovery by Veeam Vanguard Didier Van Hoye.

Veeam - VM Recovery mode

Figure 5. Soon after VM is started in the Instant VM Recovery mode you should initiate its migration back to the production.

Mind the CIFS as a main target repository

Veeam is storage agnostic and supports several types of backup repositories. Over the years, it was proven that a Windows or Linux physical server with internal storage gives the best performance in most cases.

Backup repository on a CIFS share still remains a popular choice, yet it generally offers the poorest performance of all options. Many modern NAS devices support iSCSI, so a better choice would be to create an iSCSI disk and present it to a Veeam server/proxy. Note though, that it’s also not recommended to use reverse incremental backup mode for repositories on NAS because it puts heavy IO load on the target.

Target proxy for replication

When replicating over the WAN, it is advised to deploy a backup proxy on the target site and configure it as a target proxy in replication job settings. This will create a robust channel between the two sites. We recommend setting a target proxy to NBD/Network mode, as using hot-add for replica can cause stuck and orphaned snapshots.

Note that when using WAN accelerators, a target proxy should still be deployed. Target WAN accelerator and target proxy can be installed on different or on a single machine, given it has enough resources.

Veeam - For replication over WAN, you should specify source and target proxy

Figure 6. For replication over WAN, you should specify source and target proxy.

Veeam - Set the target proxy mode to Network.

Figure 7. Set the target proxy mode to Network.

 

A must-do for a tape server

Tape server is a component responsible for communication with a tape device. It is installed on a physical machine to which a tape device is connected (“pass through” connections via ESXi host to a virtual machine are not supported!).

Veeam Backup & Replication gets the information about the library from the OS, so you should make sure that the latest drivers are installed and the tape device is visible correctly in the device manager.

You can find more info on using tapes with Veeam Backup & Replication in the previous blog post.

Watchdog your entire IT environment with Veeam ONE

With data becoming so valuable these days, organizations can’t afford to have their IT systems unavailable even for minutes. Monitoring your environment plays a key role to ensure Availability. You need to be alerted when things could go wrong and when it’s time to fix issues before they negatively impact your business. Veeam ONE does just that, not only for your VMware vSphere and Microsoft Hyper-V virtual environments, but also integrates with Veeam Backup & Replicationand, starting with Update 3, provides visibility for Veeam Agents as well.

In this blog post, I will talk about some of the Veeam ONE capabilities that will help you keep an eye on your IT systems.

Categorizing your infrastructure objects

Veeam ONE helps you categorize objects inside your infrastructure by business unit, department, purpose or SLA by means of its Business View component. This business categorization is integrated with Veeam ONE Monitor, enabling you to monitor, troubleshoot and report on business groups of VMware and Hyper-V objects.

Veeam ONE

The Configuration tab allows you to configure the basic application settings

After adding business categories into Veeam ONE, you can start monitoring your business groups through the Business View tab in Veeam ONE Monitor.

Business View in Veeam ONE Monitor, showing a virtual machine in the VMs with Snapshots category

You can also build reports for specified categories of objects. In the Workspace view of Veeam ONE Reporter, when you select a report, you can either choose to get details on the entire virtual environment, or on specific business view objects. For example, if you group VMs by department, you can create reports for a specific department in your organization.

The Business View objects window in Veeam ONE Reporter

Using alarms

There are predefined alarms in Veeam ONE for VMware vSphere and vCloud Director, Microsoft Hyper-V, Veeam Cloud Connect, Veeam Agents, and for internal Veeam ONE issues. All of them are designed to alert IT admins when any notable events or issues occur in their environment. With these alarms, you can easily identify, troubleshoot issues and quickly act to keep business operations running.

If they are used properly, alarms can be a critical method to notify you about the performance of your virtual environment. It’s important to know that too many alarms can lead you to ignore them, while too sensitive triggers can lead to false alarms.

The predefined alarms are built on best practices thresholds and trigger when the parameters defined in the alarm are different against collected data from the virtual and backup servers.

 

Alarm details in the Alarm Management tab. You can create custom alarms for any kind of tasks and events

When an alarm is triggered, the Veeam ONE console displays details about the root cause of the issue and some ways of resolution. Veeam ONE alarms are customizable and you can edit them to fit your business needs by adding different rules, changing thresholds or assigning them to different objects. You can also choose to send an email to a group when alarm’s severity changes or to suppress an alarm during scheduled activities. Moreover, in the Knowledge base tab you can add custom text to help you with solving the alarm next time it’s triggered.

Alarm Settings

Performance and health state monitoring

Veeam ONE enables you to monitor the performance of your VMware vSphere or Microsoft Hyper-V environment through comprehensive charts. In Infrastructure View, you can find information about the latest alarms, CPU, memory and disk resources for different timeframes, as well as network usage details. As you browse in the Infrastructure tree, the Summary tab will display different information for different objects and the rest of the tabs will vary too as you move deeper in the environment.

If the parent object is selected, the Infrastructure Summary tab will display the health state overview, including the Host State, Datastores State and the Virtual Machines state.

Along the navigation menu, Veeam ONE Monitor also provides details about Alarms (as you saw earlier in this post), as well as very well-organized metrics on resource consumption. Here you can check the VMs and hosts that use the most resources (CPU, memory, network usage and more), but also the hosts with the lowest load. These stats are available for both VMware vSphere and Microsoft Hyper-V environments.

The CPU chart shows the amount of used processor resources on a machine where a backup infrastructure component runs. Graphs in the CPU chart illustrate the level of processor usage for every separate CPU on the machine. The Total graph shows the cumulative processor utilization for all CPUs.

Capacity planning

While Veeam ONE Monitor provides extensive visibility over your IT infrastructure, the Reporter makes it easy for IT administrators to obtain detailed analysis of their virtual infrastructure and helps to take the guesswork out of capacity planning.

In virtual environments, capacity planning is a critical task for housekeeping and allows IT admins to make correct decisions based on accurate forecasts. The first thing we can see in Veeam ONE Reporter is the VMware Capacity Planning dashboard, which includes few widgets that offer a sneak peek into the more detailed reports of the infrastructure.

Veeam ONE 07

The VMware Capacity Planning dashboards displays details about Hosts and Clusters, Datastores, loss of a host and how many VMs can be added to the infrastructure without having to purchase more resources

Now let’s take a look at the capacity planning reports that we have available: Capacity Planning, Host Failure Modelling, How Many More VMs Can be Provisioned and Over-provisioned Datastores.

The reports are based on collected and analyzed historical data and they are very customizable, allowing you to choose individual or all datastores, set limits for CPU and Memory, select timeframe of the analyzed performance data, as well as the period of planning.

Veeam ONE 08

The Capacity Planning report provides very detailed forecast on my virtual infrastructure

At the end of the day, the capacity planning exercise helps you answer some simple questions: Do I have enough resources for more virtual machines? When will I run out of resources? Don’t underestimate these questions, they may save you time and money one day.

Agents monitoring and reporting

Update 3 came with many new features for Veeam Backup & Replication 9.5 and Veeam ONE 9.5, and one of the most highly-anticipated is agent management. Veeam ONE 9.5 now supports agent monitoring and reporting for Veeam Agent for Microsoft Windows and Veeam Agent for Linux, enabling you to have visibility into both your virtual and physical infrastructures.

You can have real-time monitoring and alerting for the Veeam Agent jobs managed by Veeam Backup & Replication servers that you monitor in Veeam ONE. To do so, go to Data Protection view in Veeam ONE Monitor and open the Agent Jobs tab of the desired backup infrastructure node. For each job, you will gain information such as the status of a backup job (Success, Warning, Failed, Running, or jobs with no status), backup job name, type, transferred data and more.

Veeam ONE 09

Veeam Agent jobs in Veeam ONE Monitor

In the Veeam Backup Agents report you can choose to include business groups (defined in Veeam ONE Business View) or Veeam Backup & Replication servers and protection groups. Likewise, you can choose to include in the report either specific Veeam backup agents or backup jobs and policies.

Veeam ONE 10

Selecting the report’s parameters

Veeam ONE 11

Veeam ONE 12

In my example, I chose to have a report for Windows-only machines with the RPO (Recovery Point Objective) of 1 week, meaning I will get a list of computers protected weekly. On the second page, you’ll find a detailed list of protected and unprotected computers, including information like IP address, Backup Job/Policy, Last Backup Date and more.

Conclusion

Monitoring your entire IT infrastructure can often be overlooked. This is a mistake that can not only cost your business money, but also its reputation. Minor breaches, due to their complexity, can cause performance issues in virtual and physical environments. However, this can be avoided with a good monitoring tool that alerts you when things are not working as they should. Veeam ONE provides a comprehensive set of alerts and a very user-friendly interface that facilitates visibility, troubleshooting and resolving issues.


This article was provided by our service partner Veeam

Re-Thinking ‘Patch and Pray’

When WannaCry ransomware spread throughout the world last year by exploiting vulnerabilities for which there were patches, we security “pundits” stepped up the call to patch, as we always do. In a post on LinkedIn Greg Thompson, Vice President of Global Operational Risk & Governance at Scotiabank expressed his frustration with the status quo.

Greg isn’t wrong. Deploying patches in an enterprise department requires extensive testing prior to roll out. However, most of us can patch pretty quickly after an announced patch is made available. And we should do it!

There is a much larger issue here, though. A vulnerability can be known to attackers but not to the general public. Managing and controlling vulnerabilities means that we need to prevent the successful exploitation of a vulnerability from doing serious harm. We also need to prevent exploits from arriving at a victim’s machine as a layer of defense. We need a layered approach that does not include a single point of failure–patching.

A Layered Approach

First off, implementing a security awareness training program can help prevent successful phishing attacks from occurring in the first place. The 2017 Verizon Data Breach Investigations Report indicated that 66% of data breaches started with a malicious attachment in an email—i.e. phishing. Properly trained employees are far less likely to open attachments or click on links from phishing email. I like to say that the most effective antimalware product is the one used by the best educated employees.

In order to help prevent malware from getting to the users to begin with, we use reputation systems. If almost everything coming from http://www.yyy.zzz is malicious, we can block the entire domain. If much of everything coming from an IP address in a legitimate domain is bad, then we can block the IP address. URLs can be blocked based upon a number of attributes, including the actual structure of the URL. Some malware will make it past any reputation system, and past users. This is where controlling and managing vulnerabilities comes into play.

The vulnerability itself does no damage. The exploit does no damage. It is the payload that causes all of the harm. If we can contain the effects of the payload then we are rethinking how we control and manage vulnerabilities. We no longer have to allow patches (still essential) to be a single point of failure.

Outside of offering detection and blocking of malicious files, it is important to stop execution of malware at runtime by monitoring what it’s trying to do. We also log each action the malware performs. When a piece of malware does get past runtime blocking, we can roll back all of the systems changes. This is important. Simply removing malware can result in system instability. Precision rollback can be the difference between business continuity and costly downtime.

Some malware will nevertheless make it onto a system and successfully execute. It’s at this point we observe what the payload is about to do. For example, malware that tries to steal usernames and passwords is identified by the Webroot ID shield. There are behaviors that virtually all keyloggers use, and Webroot ID Shield is able to intercept the request for credentials and returns no data at all. Webroot needn’t have seen the file previously to be able to protect against it. Even when the user is tricked into entering their credentials, the trojan will not receive them.

There is one essential final step. You need to have offline data backups. The damage ransomware does is no different than the damage done by a hard drive crash. Typically, cloud storage is the easiest way to automate and maintain secure backups of your data.

Greg is right. We can no longer allow patches to be a single point of failure. But patching is still a critical part of your defensive strategy. New technology augments patching, it does not replace it and will not for the foreseeable future.


This article was provided by our service partner Webroot.

meltdown spectre

Spectre, Meltdown, & the CLIMB Exploit: A Primer on Vulnerabilities, Exploits, & Payloads

In light of the publicity, panic, and lingering despair around Spectre and Meltdown, I thought this might be a good time to clear up the differences between vulnerabilities, exploits, and malware. Neither Spectre nor Meltdown are exploits or malware. They are vulnerabilities. Vulnerabilities don’t hurt people, exploits and malware do. To understand this distinction, witness the CLIMB exploit:

The Vulnerability
Frequently, when a vulnerability is exploited, the payload is malware. But the payload can be benign, or there may be no payload delivered at all. I once discovered a windows vulnerability, exploited the vulnerability, and was then able to deliver the payload. Here’s how that story goes:

It’s kind of embarrassing to admit, but one evening my wife and I went out to dinner, and upon returning, realized we had a problem. It wasn’t food poisoning. We were locked out of our house. The solution was to find a vulnerability, exploit it, and get into the house. The vulnerability I found was an insecure window on the ground floor.

With care I was able to push the window inward and sideways to open it. From the outside, I was able to bypass the clasp that should have held the window closed. Of course, the window was vulnerable for years, but nothing bad came of it. As long as nobody used (exploited) the vulnerability to gain unauthorized access to my home, there was no harm done. The vulnerability itself was not stealing things from my home. It was just there, inert. It’s not the vulnerability itself that hurts you. It’s the payload. Granted, the vulnerability is the enabler.

The window was vulnerable for years, but nothing bad happened. Nobody attacked me, and while the potential for attack was present, an attack (exploit) is not a vulnerability. The same can be true of vulnerabilities in software. Opening the window is where the exploit comes in.

The Exploit
My actual exploit occurred in two stages. First, there was proof of concept (POC). After multiple attempts, I was able to prove that the vulnerable window could be opened, even when a security device was present. Next, I needed to execute the Covert Lift Intrusion Motivated Breach (CLIMB) exploit. Yeah, that means I climbed into the open window, a neat little exploit with no coding required. I suppose I could have broken the window, but I really didn’t want to brick my own house (another vulnerability?).

The Payload
Now we come to the payload. In this case, the payload was opening the door for my wife. You see, not all payloads are malicious. If a burglar had used the CLIMB exploit, they could have delivered a much more harmful payload. They could have washed the dishes (they wouldn’t, unless they were Sheldon Cooper), they could have stolen electronic items, or they could have planted incriminating evidence. The roof is the limit.

Not all vulnerabilities are as easy to exploit as others. All of my second-floor windows had the same vulnerability, but exploiting them would have been more difficult. I am sure happy that I found the vulnerability before a criminal did. Because I was forgetful that fateful night, I’m also happy the vulnerability was there when I found it. As I said, I really didn’t want to break my own window. By the way, I “patched” my windows vulnerability by placing a wooden dowel between the window and the wall.

There you have it. Vulnerabilities, exploits, and payloads explained through the lens of the classic CLIMB exploit.


This article was provided by our service partner : Webroot

MySQL

Veeam – Linux VM: A place to back up MySQL

What does it take to back up MySQL on a Linux VM? This is a riddle we sometimes hear at Veeam: When running on a Linux VM, how does one quiesce MySQL databases? Unfortunately, there are not many new ways to answer this riddle, and the answers we currently have are already tried and tested!

The answers can be found in our popular white paper Consistent protection of MySQL/MariaDB with Veeam, written by Solutions Architect Pascal Di Marco. The paper is available for download on our website and describes three different methods for backing up MySQL/MariaDB on a Linux VM. Two hot backup methods running pre- and post-snapshot scripts, and cold backup using database shutdown. This makes use of VMware tools installed on the Linux VM. It is not straightforward like a Microsoft SQL quiescence, because Linux doesn’t have a VSS mechanism like Windows does.

Described are methods of backing up MySQL/MariaDB on a Linux VM using activated scripts local to the database. VMware can run a script to act before the snapshot is created, known as the pre-freeze script and can run a script to do things after the snapshot is created, known as the post-thaw script.

Here’s a quick summary:

Option 1: Hot backup — Database online dump

The mysqldump command copies a database to storage accessible from the MySQL server, taking an online dump of each database without disrupting the MySQL service. This method lets you take a transaction consistent backup of databases but more steps are needed to perform a restore. As with Option 1, the pre-freeze-script will only run if you have the VMware tools running.

Advantage: This allows for 100% uptime; the MySQL service does not stop and the dumped databases are in a transaction-consistent state.

Disadvantage: Depending on the size of your databases, the process may take a considerable amount of time to achieve. A second copy of the database means extra storage space is required to maintain it.

Option 2: Hot backup — Database freezing

Stop the MySQL service for a few moments while the snapshot is created, then start it again. The post-thaw script will not run until the snapshot is created. The pre-freeze script and post-thaw script will only be able to run if you have the VMware tools running in your MySQL server.

Advantage: This is quick and simple, allowing you to take a transaction of all databases with no additional disk usage local to the MySQL server.

Disadvantage: Databases running on the MySQL server will briefly be unavailable, and applications that need 100% uptime may not find this suitable.

Option 3: Cold Backup — Database shutdown

In this method, the application service will be stopped during snapshot creation and restarted once the VM snapshot has been created. It requires permission to start and stop application services but does not require MySQL user permissions. You can authenticate by either using the MySQL default configuration file or hardcoding the username and password in the script.

Advantage: This is easy to set up and doesn’t take extra space. It provides a short RTO, since no further action is required aside from booting the restored guest.

Disadvantage: The databases will be totally unavailable while the guest snapshot is created.

Recovery

Guest recovery: The cold backup and freeze method will leave the database consistent and able to start up without additional operation, so restoring the VM from the backup files is the only operation to perform. The guest recovery may benefit from Veeam’s Instant VM Recovery feature, which lets you boot up the guest directly from the Veeam Backup Repository in minutes.

Additional dump restoration: The extra task of injecting the dump file into the database using file redirection is necessary if the following is true: The issue is not limited to a database outage, the entire VM must be recovered from the Veeam Backup file and the database dump method has been used.

Veeam U-AIR database restoration: Whether it is a granular or a full database restoration, Veeam U-AIR wizard can be used in conjunction with any relevant database management tool such as MySQL Workbench to recover a database item.

 

meltdown spectre

Microsoft Releases More Patches for Meltdown & Spectre

Microsoft informed users on Tuesday that it released additional patches for the CPU vulnerabilities known as Meltdown and Spectre, and removed antivirus compatibility checks in Windows 10.

Meltdown and Spectre allow malicious applications to bypass memory isolation and access sensitive data. Meltdown attacks are possible due to CVE-2017-5754, while Spectre attacks are possible due to CVE-2017-5753 (Variant 1) and CVE-2017-5715 (Variant 2). Meltdown and Spectre Variant 1 can be resolved with software updates, but Spectre Variant 2 requires microcode patches.

In addition to software mitigations, Microsoft recently started providing microcode patches as well. It initially delivered Intel’s microcode updates to devices running Windows 10 Fall Creators Update and Windows Server 2016 (1709) with Skylake processors.

Now that Intel has developed and tested patches for many of its products, Microsoft has also expanded the list of processors covered by its Windows 10 and Windows Server 2016 updates. Devices with Skylake, Coffee Lake and Kaby Lake CPUs can now receive the microcode updates from Intel via the Microsoft Update Catalog.

Microsoft also informed customers on Tuesday that software patches for the Meltdown vulnerability are now available for x86 editions of Windows 7 and Windows 8.1.

The company has also decided to remove the antivirus compatibility checks in Windows 10. The decision to introduce these checks came after the tech giant noticed that some security products had created compatibility issues with the Meltdown patches. This resulted in users not receiving security updates unless their AV vendor made some changes.

Microsoft has determined that this is no longer an issue on Windows 10 so the checks have been removed. On other versions of the operating system, users will still not receive updates if their antivirus is incompatible.

Microsoft’s Patch Tuesday updates for March 2018 fix over 70 flaws, including more than a dozen critical bugs affecting the company’s Edge and Internet Explorer web browsers.

veeam

Getting started with Veeam Explorer for Microsoft SQL Server

Believe it or not, I used to work a lot with Microsoft SQL Server. While I did not call myself a database administrator (DBA), I did know my way around a database or two. Since I’ve been at Veeam, I have always enjoyed telling the Veeam story around using SQL Server as a critical application that needs the best Availability options.

That’s why I took particular interest in Veeam Explorer for Microsoft SQL Server that came in Veeam Backup & Replication. Veeam Explorer for Microsoft SQL Server allows application-specific restores of SQL databases, and also contents of tables, objects such as stored procedures, views and more. Additionally, you can also restore the databases to a specific transaction.

This is a great combination of functionality from the established application-aware image processing with a dedicated tool for database restores in Veeam Explorer for Microsoft SQL Server. Additionally, Veeam Backup & Replication and the Veeam Agent for Microsoft Windows also provide an image backup of the entire system.

For those who are not a DBA, sometimes dealing with low-level SQL Server topics can be a bit overwhelming. To help this process, I created a few scripts to help individuals learn this type of interaction with SQL Server. I put three (and a deleted script) up on the Veeam Github site. To use this script, only an S:\ drive is needed (the path can be changed) to create the sample database and put in a SQL Server Agent job to automatically run a few stored procedures that will insert and delete random data. This creates a database called SQLGREENDREAM.

After running the three scripts to create the database, implement the random number function and set the schedule to create the random data (2 records) and delete 1 record. The SQL Server Transaction Log Backup will show the new database being backed up after the next incremental backup:

Veeam Explorer for Microsoft SQL Server

Once the interval of the SQL Server Agent job runs (12 minutes in the GitHub script) and the Veeam Backup Job interval passes, the most selective restore point option can be selected in Veeam Explorer for Microsoft SQL Server. This selective option, to restore to a specific transaction, is shown in the figure below:

Veeam Explorer for Microsoft SQL Server

Once the interval of the SQL Server Agent job runs (12 minutes in the GitHub script) and the Veeam Backup copy interval process through a time when the test data has been run, the restore to a specific transaction option can be visible to the controlled scripting for the SQLGREENDREAM database in the GitHub repository. Then you can see the records in question being just as scripted, 2 records added then one record deleted. Those entries are done by the SQL Server Agent:

From there, the restores can be done with confidence to see how the SQL databases are restored with Veeam. With the sample scripts in the GitHub repository, one can become more comfortable with these restore situations when venturing out of normal comfort zones! If you are using Veeam Backup Free Edition and the SQL Server is a VM being backed up, you can still use Veeam Explorer for Microsoft SQL Server to restore the database to the time of the image-based backup; just no transaction rollback. You can use the NFR program for a fully functional installation also.


This article was reposted from Veeam.com

veeam

Best practices from Veeam support on using tape

When speaking about backup storage strategies, Veeam’s recommendation is to keep the initial backup chain short (7 – 14 points) and use general purpose disk that will allow you to recover data in the shortest amount of time. The long-term retention should come from secondary and tertiary storage, which typically boasts a much lower storage cost per TB, but at a trade-off, the RTO when restoring from such storage can take much longer time. Here’s the graphics, which illustrates this scenario:

veeam support

Additionally, with many new features of Veeam, the tape support now includes putting vSphere, Hyper-V, and Veeam Agents for Microsoft Windows and Linux backups on tape.

One of the most popular choices for backup archival is tape. It is cheap, reliable and offers protection against crypto viruses and hacker attacks. Additionally, it’s offline when not in a tape loader.

With Veeam, IT administrators can use flexible options to create copies of backups and store them on a different media, following the 3-2-1 Rule for the backup and disaster recovery. This blog post provides advice and considerations that will help you create a robust tape archival infrastructure.

How to deploy a tape library and use it with Veeam

When planning and implementing your deployment project, follow the recommendations below:

  1. It is recommended to configure the tape library for use exclusively by Veeam Backup & Replication. Using it together with any third-party tape-recording software (for example, in your evaluation lab) may prevent other software from recording.
  2. To streamline the workflow, use tapes with barcodes. Please check the integrity of the barcodes before you start using the tapes, and make sure the barcode reader is turned on. If you have multiple libraries, ensure that the barcodes are unique throughout the infrastructure.
  3. For increased capacity, use the latest LTO. Since Veeam Backup & Replication 9.5 Update 3, Veeam now supports the LTO-8 format.
  4. If you plan to use encryption for archived data, consider using hardware encryption (implemented in LTO-4 and later). Software encryption can decrease performance.
  5. Do not use hardware compression with compressed Veeam backups. Double compression will not give any benefits and can even increase the size of the file on tape.
  6. Install and check the following:
  • The latest drivers for your tape library. Remember that only the original (OEM) drivers are supported. Drivers supplied with Microsoft Windows are not recommended.
  • The latest changer and controller firmware. Changers working via SCSI are supported.

You will need a tape server that will perform most data transfer tasks during archiving to tape. Check the following prerequisites:

  • This should be a physical machine or a VM connected through iSCSI, since direct pass-through is not supported.
  • Using a Windows 2008 R2 machine for a tape server is not recommended due to the possibility of performance degradation. Instead, use Windows Server 2012 or later to achieve better performance and seamless operation.
  • Best practice is to provide a direct connection from tape server to the repository to improve the performance and specify this preferred repo in tape server connections.
  • If you plan to create synthetic backups, using a deduplication storage is not recommended.

One thing to also consider is to use GFS media pools with the tape support in Veeam. This feature allows longer-term retention to be set easily for tape backups as shown in the picture below:

veeam support

If you plan to perform file-to-tape archiving for a large number of files (more than 500,000 per job), consider using any commercial edition of SQL Server for Veeam configuration database to support these operations. Configuration database stores information about all files backed up by Veeam Backup & Replication, and using SQL Server Express edition (with its 10 GB limit for a database size) may lead to significant performance degradation. If database size reaches 10 GB, all Veeam operations will stop.

To load or get the tapes from the library, use the import-export slots. If you need to perform these operations manually, remember to stop tape jobs, stop tape server, perform manual operation, start server, rescan or run inventory for the library (to recognize the uploaded tapes) and then restart the tape job.

  • If the tapes have barcodes, then you can perform the rescan.
  • If the tapes do not have barcodes, then you should perform the inventory.

What to consider before starting the upgrade

If you are upgrading your Veeam deployment, then you should first upgrade the Veeam backup server.

The tape server will be upgraded after that, using the automated steps of the Upgrade wizard that opens after the first launch of Veeam Backup & Replication console. However, you can choose to upgrade it manually by starting the Upgrade wizard at any time from the main Veeam Backup & Replication menu.

If you are upgrading your tape library, consider the following:

  1. To streamline the process and skip the catalogue step, you can add the new library to the existing media pools, and after the old library is switched off, remove it from the media pools.
  2. After connecting the new library to Veeam server, you should load the existing tapes with their barcodes to that new library and perform the rescan. Then you can switch the old library to the offline state (detach it from Veeam server) and then delete it from Veeam Backup & Replication configuration.

What to consider when planning for tape jobs

Before you start configuring Veeam jobs for tape archiving, consider the following factors:

  1. What entities will you need to archive? Will these be files and folders, or VM backups? Do you need to archive full backups only, or both full and incremental backups?
  2. What is the estimated data size?
  3. How often will you need to archive data?
  4. What will be the retention policy for your data?
  5. How often will the tapes be changed? Will they be exported?
  6. What is the tape capacity?
  7. What tape device will be used for archiving?

After these considerations, it is recommended that you double your estimated number of tapes when planning for the resources.

Conclusion

In this blog post, we’ve talked mainly about tape infrastructure. We recognize that when setting up tape jobs, the learning curve can be quite steep. Instead of explaining all the concepts, we chose a different approach. We’ve prepared a list of settings and a well-defined result that will be achieved. You can choose to use them as they are or as a basis for your personal setup. Check out this Secondary Copy Best Practices Veeam guide for more details.


This article was provided by Veeam.com

Disaster Recovery Planning

Disaster Recovery Planning

It seems like it’s almost every day that the news reports another major company outage and as a result, the massive operational, financial and reputational consequences experienced, both short- and long-term. Widespread systemic outages first come to mind when considering disasters and threats to business and IT service continuity. But oftentimes, it’s the overlooked, “smaller” threats that regularly occur. Human error, equipment failure, power outages, malicious threats and data corruption can too bring an entire organization to a complete standstill.

It’s hard to imagine that these organizations suffering the effects of an outage don’t have a disaster recovery plan in place — they most certainly do. But, why do we hear of and experience failure so often?

Challenges with disaster recovery planning

Documenting

At the heart of any successful disaster recovery plan is comprehensive, up-to-date documentation. But with digital transformation placing more reliance on IT, environments are growing larger and more complex, with constant configuration changes. Manually capturing and documenting every facet of IT critical to business continuity is neither efficient or scalable, sending to us our first downfall.

Testing

Frequent, full-scale testing is also critical to the success of a thorough disaster recovery plan, again considering the aforementioned scale and complexity of modern environments — especially those that are multi-site. Paired with the resources required and potential end-user impact of regular testing, the disaster recovery plan’s viability is often untested.

Executing

The likelihood of a successful failover — planned or unplanned — is slim if the disaster recovery plan continues to be underinvested in, becoming out-of-date and untested quickly. Mismatched dependencies, uncaptured changes, improper processes, unverified services and applications and incorrect startup sequences are among the many difficulties when committing to a failover, whether it’s a single application or the entire data center.

Compliance

While it is the effects of an IT outage that first come to mind when considering disaster recovery, one aspect tends to be overlooked — compliance.

Disaster recovery has massive compliance implications — laws, regulations and standards set in place to ensure an organization’s responsibility to the reliability, integrity and Availability of its data. While what constitutes compliance varies from industry to industry, one thing holds true — non-compliance is not an option and brings with it significant financial and reputational risks.

 

meltdown spectre

Meltdown & Spectre: Where Are We at Now?

Meltdown and Spectre still continue to dominate the security news and the more we delve into it, we are starting to understand the depth and breadth of what this now means for the future of the security landscape.

Turns out the three variants of side-channel attacks, Meltdown and two different for Spectre, were discovered back in June of last year [2017] by researchers using speculative execution, which is where processors execute on code and then fetch and store the speculative results in cache. It’s a technique used to optimize and improve the performance of a device. What is important to note with Spectre is that it puts users at risk for information disclosure by exposing the weakness in the architecture of most processors in the market, and the breadth is vast: Intel, AMD, ARM, IBM (Power, Mainframe Z series) and Fujitsu/Oracle SPARC implementations across PCs, physical and virtual servers, smartphones, tablets, networking equipment and possibly IoT devices.

Currently there are no reported exploits in the wild.

Of the two, Meltdown is the easier one to mitigate with operating system updates. AMD processors are not affected by Meltdown. Spectre is a bit more complex to resolve because it is a new class of attack. The two variants of Spectre both can potentially do harm like stealing logins and other user data residing on the affected device. Intel, ARM, and AMD processors are affected by Spectre. Recently, Microsoft released another emergency update to disable Intel’s microcode fix. This original update was meant to patch for variant 2 of Spectre. Unfortunately, that update had adverse effects as there were numerous reports of reboots and instability, so Microsoft issued an out of band update to disable.

Things are still evolving around Spectre and while operating system updates and browser updates are helping to patch for Spectre, it is being reported by some sources that a true fix may be an update to the hardware (processor) itself.

The following is a chart* to clarify each vulnerability:

meltdown-spectre-chart

*Chart is courtesy of SANS/Rendition Infosec. See full presentation here.

It will be important over the next few weeks to stay on top of any breaking news around Meltdown and Spectre. Mitigation efforts should be underway in your IT organization to prevent a future zero-day attack.


This article was provided by our service partner : Connectwise