Monthly Archives: March 2016
Doughnuts! I told you I don’t like ethnic food! – Mr Burns, The Simpsons
This edition of Friday Shorts isn’t really going to be that short at all – It’s been a while since I’ve done one and there is a ton of content out there that I wanted to re-share for people so buckle up, here it is!
2016 Toronto VMUG UserCon !
Thursday, February 25th marked the date for the 2016 Toronto VMUG Full Day UserCon and I couldn’t be happier as to how it turned out! Honestly, as a leader I was interested the whole day about how many attendees we had and as it turned out we had our best year ever with 800+ registrations and 605 brains that walked through the doors! Last year we had 601 and I’m sure it would have been more for this year but you know, Canadian winters eh! We had Nick Marshall do the morning keynote and his talk was awesome – I heard nothing but good things from the attendees in regards to the content Nick had to talk about (BTW it was dubbed Building your IT Career) If you get a chance to have Nick out for your VMUG do so! He’s awesome! For our lunchtime keynote we had a VCDX panel – Tim Antonowitz, James Wirth, Joe Silvagi and Eiad Al-Aqqad graciously volunteered to sit up on stage as we had some great discussion with the attendees at the conference from everything dealing with certification through to NSX integration! All in all it was a great day and if you were there I hope you enjoyed it – if you weren’t, come next year!
What does Docker, vExpert, Ravello and Netflix have in common?
@h0bbel that’s what! More explanation – well, Christian is a vExpert, vExperts get some free CPU hours on Ravello Systems (Ravello is awesome BTW – be nice Oracle!), and he has a great post on his blog on how to setup Dockerflix within Ravello on Photon in order to test (get around) those geo-blocking Netflix firewall “you can’t watch this” setups! So if you are say a vExpert blogger that lives in Canada that really fancies yourself a couple of hours of Legally Blonde I suggest you set the Maple Syrup ladle down and head over to Christians post!.
Speaking of Ravello – How’s about some community cmdlets?
And when you think of community in the sense of this blurb just think of Luc Dekens being a one-man community! If you are Ravello user and fancy yourself some PowerShell cmdlets Luc has an updated module on his blog available for download. Luc has certainly put a lot of effort into this module which contains hooks into almost every single API Ravello offers! I certainly find this module very useful for my work with Ravello and just want to give Luc a big Thank You for this!
Running commands on the VCSA without SSH
If I was being shipped off to a deserted island and could only take with me the RSS feed of one virtualization blog I have to think that as of today that blog would be William Lams virtuallyGhetto! William just seems really good at figuring things out – is that a line that can go on a resume? I don’t know, either way, his recent post on how to run commands on the VCSA remotely without enabling ssh is pretty awesome! He utilizes the Guest Operations API through a vSphere SDK to do so! Just go read it, it’s gold Jerry, it’s gold! goes here it is
#ChrisWahlFacts – He doesn’t mess around when it comes to dropping knowledge about PowerShell and REST APIs
Over yonder at WahlNetwork.com Chris Wahl has a great series going on dubbed Automation for Operations – Honestly the whole series is great but its the last 4 parts that have really helped me out…a lot! Chris has done a great job at explaining a bunch of concepts around utilizing PowerShell to connect to RESTful API’s, including Authentication, Processing GET Requests, Sending data with POST/PUT/PATCH, and his latest, creating hashtables for JSON payloads!
Veeam Vanguard nominations are open!
Ever wake up and think “Hey, Why isn’t so and so a Veeam Vanguard?” or “Why am I not a Veeam Vanguard?” Well, so long as you wake up wondering about that before March 30th you have a chance to throw your name or someone whom you think is worthy into the mix! You can check out the official Veeam post here.
We all know that being a vExpert isn’t about what you get, but more about what you give – buuuuutttt, the fact of the matter is you do get stuff, sometimes lots of stuff, and it’s hard to keep track of it all! Thankfully for the vExpert community Andrea Mauro is doing a great job at keeping track of it all for you! Without posts like this there is no way I’d be able to keep track of it! So, Thanks Andrea!
Veeam v9 is here and if you have already performed the upgrade you might be a bit anxious to start using some of the new features that came along with it. In my case I’ve already gone ahead and done my due dilligence by enabling and configuring some of the new features on a few test backup/replication jobs and I’m ready to duplicate this to the rest of the environment – problem being I have A LOT of jobs to apply these to. As always I look to automation to solve this issue for me. One, it is way faster and two, it provides a consistent set of configuration (or errors) accross my jobs – making it far more easier to troubleshoot and change if need be. Thanfully Veeam provides a set of PowerShell cmdlets that allows me to automate the configuration of some of these features. So, if you are ready to go let’s have a look at a few of the new features within Veeam v9 and their corresponding PowerShell cmdlets.
Just a note – for each of these examples I’ve just posted the code to handle one object – but you could easily surround the blocks of code with a foreach() if you are looking to apply the configurations to many objects. <- this is what I have done however it’s easier and much easier to read if I just insert code dealing with individual objects.
Enabling Per-VM file chains
First up is the Per-VM Backup File Chain introduced in v9. In previous version of Veeam all of the VMs contained within a single job were also contained within single backup files – In the end we were left with some massive backup files sitting on our repositories. Having a massive file laying around isn’t such a big deal, but when the time came where we were required to manage or move that file in any way it presented a few problems – it took a long time to move and activity surrounding that file need be disabled until we were done. In the end we were left with a lot of waiting and no backups. The v9 Per-VM Backup File chain fixes this – it allows us to store our backup files on a per-vm basis, leaving them much easier to manage, not too mention the headaches that are saved if corruption on our backup files occur. Either way I wanted to enable this on a dozen or so of my repositories…
I say repository since that is where the per-VM Backup Chain is enabled – not on the job, not on the VM, on the actual Veeam Repository. The process of doing so is pretty simple, get our repository, set a flag to true, and call back the saveOptions() function – as follows…
$repo = Get-VBRBackupRepository -Name "Name of repository" $repo.Options.OneBackupFilePerVm = $true $repo.saveOptions()
New Mount Server
In previous versions of Veeam before v9 certain restore operations required mounting backups to a Veeam backup server, which when dealing with remote sites could of resulted in increased bandwidth usage depending on how you had configured your environment. v9 gives us the ability to designated any windows machine as a mount server. The mount server can then be used as a mount point to perform file-level recovery operations, allowing the bandwidth to stay local at the remote site.
As with the Per-VM backup chains, mount servers are enabled on a repository level. In my cases I wanted my repositories and mount servers to be one of the same – in order to do that I simply get the remote repository, then call Set-VBRBackupRepository passing it my mount host name and turning on the vPowerNFS flag as shown below…
$repo = Get-VBRBackupRepository -Name "Name of repository" $repo | Set-VBRBackupRepository -MountHost (Get-VBRServer "Name of desired Mount Host") -EnableVPowerNFS
Guest Interaction Proxy
Another new ROBO enhancing feature in v9 is the ability to specify a guest interaction proxy. Previously the Veeam Backup and Replication server handled deploying runtime processes into the VMs to facilitate different parts of the backup and replication jobs – in v9, we can now designate servers that may be onsite to do this – This helps in a couple of ways – first, this helps reduce traffic traversing our WAN and secondly, sometimes backup servers were isolated from the VMs they were backing up, prevening certain actions from even being able to take place. Anyways, the Guest Interaction Proxy is a per-job setting and is setup within the VSS settings of the job. In my cases I just needed to flip the AutoDetect to $true in order to get Veeam to select the proper GIP.
$job = Get-VBRJob -Name "Job Name" $vssoptions = $job.GetVssOptions() $vssoptions.GuestProxyAutoDetect = $True $job.setVssOptions($vssoptions)
Enable deleted file blocks
Veeam v9 has introduced many data reduction technologies within their application in order to help us save space and more efficiently manage all of our backup capacity. The first technique we will look at is the ability to not backup deleted file blocks. This can be enabled on your existing backup jobs by setting the DirtyBlocksNullingEnabled flag as follows.
$job = Get-VBRJob -Name "Job Name" $joboptions = $job.getOptions() $joboptions.ViSourceOptions.DirtyBlocksNullingEnabled = $True $job.setOptions($jobOptions)
Exluding certain folders/files
Another space saving feature inside of v9 is the ability to exclude or include certain files or folders contained with the VMs – think about Temp directories – under normal circumstances we don’t need them so why take up all that capacity backing them up. We can set this up by first setting the BackupScope property – this can be set to exclude folders (ExcludeSpecifiedFolders), only include folders(IncludeSpecifiedFolders) or simply backup everything(Everything). Depending on the setting of the BackupScope we then set the GuestFSExcludeOptions or the GuestFSIncludeOptions with an array of strings pointing to the desired folders – finally, saving our job options as follows…
$job = Get-VBRJob -Name "Job Name" $jobobject = Get-VBRJobObject -Job $job -Name "VM Name" $vssoptions = Get-VBRJobObjectVssOptions -ObjectInJob $jobobject $vssoptions.GuestFSExcludeOptions.BackupScope = "ExcludeSpecifiedFolders" $vssoptions.GuestFSExcludeOptions.ExcludeList = "C:\folder","D:\folder","c:\test\folder" $jobobject.SetVssOptions($vssoptions)
Storage-Level Corruption Guard on Production Backup Jobs (not just backup copy)
SureBackup does a great job at ensuring our VMs will boot however they may be certain portions of our data that can become corrupt that can actually pass a SureBackup test. To help alleviate this, Veeam has introduced something called Storage-Level Corruption Guard (SLCG) to periodically identify and fix certain storage issues. SLCG has actually been around in previous versions, but only available for Backup Copy jobs. In v9 it can now be enabled on our production backup jobs, giving us one more piece of mind when the need to restore comes along. This is enabled by first enabling the EnableRechek (yes, it’s spelled like that) flag, then setting a schedule (Daily/Monthly) and few other options and finally saving our options… Below we’ve set a job up to perform SLCG on Fridays.
$job = Get-VBRJob -Name "Job Name" $joboptions = $job.getOptions() $joboptions.GenerationPolicy.EnableRechek = $True $joboptions.GenerationPolicy.RecheckScheduleKind = "Daily" $joboptions.GenerationPolicy.RecheckDays = "Friday" $job.setOptions($jobOptions)
Defragment and compact full backup file – on production backups not just backup copy
Over time our full backup files can become bloated and heavily fragmented – when we delete a VM for example, the full backup might still be holding onto certain data that was in that VM. Normally we could take an active full backup in order to help purge this data, but as we all know that requires us to affect production and use up valuable resources. To help alleviate this v9 has introduced the ability to defragment and compact our full backups on a schedule. This is done very similar to that of SLGC, getting the VSS options of a job and setting the schedule. Below we enable our defrag to run on Fridays.
$job = Get-VBRJob -Name "Job Name" $joboptions = $job.getOptions() $joboptions.GenerationPolicy.EnableCompactFull = $True $joboptions.GenerationPolicy.CompactFullBackuScheduleKind = "Daily" $joboptions.GenerationPolicy.CompactFullBackupDays= "Friday" $job.setOptions($jobOptions)
So there you have it – a little bit of automation for those that may have to update numerious jobs to fully take advantage of some of the features Veeam v9 has introduced. As always please feel free to reach out if any of this isn’t working, or you have any comments, questions, concerns, rants, etc. Thanks for reading!