Feed aggregator

A Practical Example of Using Cloud Storage with XenServer in the Citrix Demo Center: Chapter 2

Citrix Blogs - Thu, 01/18/2018 - 11:00
In this second installment, I’ll discuss why I chose a shell (Bash) to script the Object Storage libraries, and I’ll discuss some of the more complex issues when handling errors in Bash.

Language wars: Some may say that Python would …

  Related Stories

Accelerate ISV Business Transformation with the Power of Citrix Cloud

Citrix Blogs - Thu, 01/18/2018 - 09:00

Following on Calvin Hsu’s recent blog post “Project Chrysalis: Transforming Business for Application ISVs,” today, we are going to dive into the mechanics of how Citrix Cloud can help an application ISV address the unique opportunity and deliver …

  Related Stories

Knowledge Center Highlights – January 2018

Citrix Blogs - Thu, 01/18/2018 - 08:30
Hello, Citrix community! Welcome to the January 2018 edition of Citrix Knowledge Center Highlights. This is a collection of new and updated articles, trending articles, and other updates from the Citrix Knowledge Center team.

We have organized these highlights into …

  Related Stories

Permanently Clear Previous Mailbox Info

Exchange Team Blog - Wed, 01/17/2018 - 14:11

We are introducing a new parameter that can be called by using the Set-User cmdlet in Exchange Online PowerShell. The feature is focused for customers doing migration of on-premises mailboxes to the cloud and you will be able to use it within few days.

Customers who have Hybrid or on-premises environments with AAD Connect / Dir Sync may have faced the following scenario:

  1. User Jon@contoso.com has a mailbox on-premises. Jon is represented as a Mail User in the cloud.
  2. You are synchronizing the on-premises directory to the cloud in preparation to migrate to Exchange Online.
  3. Due to issues with the on-premises sync or due to a configuration problem, the user Jon@contoso.com does not get the ExchangeGUID synchronized from on-premises to the cloud.
  4. If the Exchange GUID is missing from the object in the cloud, assigning an Exchange license to Jon@contoso.com will cause Exchange Online to give the user a mailbox, converting the object from a Mail User to a User Mailbox. (Adding the license is a step required for the migration of the mailbox from on-premises to the cloud.)
  5. The end result is the user that has 2 mailboxes: one on-premises and one in the cloud. This is not good. Mail flow issues will follow.

Those doing these types of migrations will know that the ExchangeGUID value is very important as it helps Exchange Online identify that the user has a mailbox on-premises, and if an Exchange license is assigned in the cloud, a new mailbox should not be created.

The immediate fix for this situation is to remove the Exchange License from Jon@contoso.com. This will convert the cloud object for Jon back to a Mail User. Mail flow should be restored at this point.

The problem now is that you have an “unclean” cloud object for Jon. This is because Exchange online keeps pointers that indicate that there used to be a mailbox in the cloud for this user:

PS C:\WINDOWS\system32> Get-User Jon@contoso.com | Select name,*Recipient*
Name PreviousRecipientTypeDetails RecipientType RecipientTypeDetails
---- ---------------------------- ------------- --------------------
Jon UserMailbox MailUser MailUser

Re-assigning the license after that will always err on the side of caution and Exchange Online will try to re-connect the (duplicate, temporary) mailbox in the cloud (and mailboxes can be reconnected for 30 days). Therefore Jon’s account in the cloud can’t be licensed in preparation for migration.

Up to now, one of the few options to fix this problem was to delete *only in the cloud* Jon’s object and re-sync it from on-premises. This would delete jon@contoso.com from the cloud – but from all workloads, not only Exchange. This is problematic because Jon could have his OneDrive or SharePoint data in the cloud only and deleting his account means that this will be deleted too. If the account is then re-created, Jon and the tenant admin would have to work to recover to his new account all the data he used to have in OneDrive or SharePoint just because Exchange data needed to be “cleaned up”.

The new parameter in the user cmdlet will allow tenant admin to clean up Exchange Online Jon’s object without having to delete it.

To clean the object, you can run the following command:

PS C:\> Set-User Jon@contoso.com -PermanentlyClearPreviousMailboxInfo
Are you sure you want to perform this action?
Delete all existing information about user “Jon@contoso.com"?. This operation will clear existing values from Previous home MDB and Previous Mailbox GUID of the user. After deletion, reconnecting to the previous mailbox that existed in the cloud will not be possible and any content it had will be unrecoverable PERMANENTLY. Do you want to continue?
[Y] Yes  [A] Yes to All  [N] No  [L] No to All  [?] Help (default is "Y"): Y

Executing this leaves you with a clean object that can be re-licensed without causing the 2-mailbox problem. Now you can on-board Jon’s on-premises mailbox following the usual steps. An alternative – a call to support to do the clean-up for you - is also not needed.

Remember, cleaning up the user means that the older associated disconnected (duplicate) cloud mailbox is not recoverable. If you want to keep it or be able to check it’s content, we recommend using Soft Deletion or Inactive Mailboxes to keep the mailbox.

Mario Trigueros Solorio

Coretek Services: Taking Innovation for Citrix Partners to Infinity… and Beyond!

Citrix Blogs - Wed, 01/17/2018 - 09:00
Every year, #CitrixSummit provides the perfect showcase for the year’s big reveal: The winner of the Innovation Award for Partners! And this year was no different.

Nominees for the Citrix Innovation Award exemplify how Citrix partners across the globe are …

  Related Stories

How to Win with Citrix Partner Services

Citrix Blogs - Wed, 01/17/2018 - 07:00
In 2018, one of Citrix’s key priorities is helping our customers move into the Cloud. In order for this transition to be seamless and successful, we know aligning with our partners to deliver cohesive, effective services will play a pivotal …   Related Stories

Ny undersökning: Dålig förberedelse inför GDPR

Citrix Blogs - Wed, 01/17/2018 - 01:00

GDPR är på mångas läppar och som de flesta vet träder den i kraft om ganska exakt 5 månader. För att ta reda på hur väl förberedda företagen är inför införandet av den nya dataskyddslagen så genomförde vi tillsammans med …

  Related Stories

Exchange Server guidance to protect against speculative execution side-channel vulnerabilities

Exchange Team Blog - Tue, 01/16/2018 - 17:46

We wanted to make you aware that we have now published a KB article on this subject, KB4074871. Please read it here:

Exchange Server guidance to protect against speculative execution side-channel vulnerabilities

The Exchange Team

Yellow Jackets Sport HDX 3D Pro and NVIDIA GRID in the College of Engineering

Citrix Blogs - Mon, 01/15/2018 - 07:00

Being a lifelong Wahoo — a graduate of Mr. Jefferson’s University in Charlottesville — writing about technology innovation at another Atlantic Coast Conference (ACC) school pains me. Georgia Institute of Technology, otherwise known as Georgia Tech (or in the …

  Related Stories

Citrix TIPs Series: Making the Case for XenMobile in the Cloud

Citrix Blogs - Mon, 01/15/2018 - 07:00
Have a XenMobile deployment on premises today that you are thinking about consuming as a service? Or maybe you are planning your brand-new XenMobile deployment and are torn about whether to deploy on-prem or consume XenMobile Service as part of …   Related Stories

SD-WAN: The Backbone of Hybrid Cloud for the Enterprise

Citrix Blogs - Thu, 01/11/2018 - 13:00
Software-Defined WAN (SD-WAN) is, quite simply, the backbone of hybrid cloud for enterprise applications. It’s the backbone of a secure digital perimeter. Simply said, it’s the foundation of modern networking requirements.

2018 will be the year that SD-WAN is recognized …

  Related Stories

3 Things to Look For in an Awesome Content Collaboration Product

Citrix Blogs - Thu, 01/11/2018 - 09:00
As we move into 2018, many IT leaders will be looking for new content collaboration solutions that improve worker productivity and keep important files secure. If you’re looking for a new file sync-and-share solution, consider finding a product with these …   Related Stories

Citrix Announces 2017 Americas Award Winners at Summit 2018

Citrix Blogs - Wed, 01/10/2018 - 11:00
Citrix Summit 2018 in Anaheim, CA, was the perfect setting to spotlight the outstanding contributions made by our Americas partners in 2017. These annual awards recognize categories such as Partner of the Year, by segment, product and area, as well …   Related Stories

Preview the New XenApp & XenDesktop Advanced Design Course!

Citrix Blogs - Wed, 01/10/2018 - 09:00

Did you know that site design can impact many aspects of your Citrix solutions, including their day to day manageability? We’re excited to announce that Citrix Education will soon release a new advanced course CXD-410: XenApp and XenDesktop 7.1x Assessment,

  Related Stories

Middle America Banks on Citrix with Citrix Workspace, NetScaler MAS & ShareFile

Citrix Blogs - Wed, 01/10/2018 - 07:00
Two decades ago, I worked for a Fortune 50 technology company that applied spatial analysis to the challenge of bank branch placement. The results revealed that if drive time to a branch exceeded twenty minutes, a customer wouldn’t engage.

Today, …

  Related Stories

Making Medics Mobile

Citrix Blogs - Wed, 01/10/2018 - 01:00
In our last look at Citrix in the NHS and other healthcare organisations, we focused on the enormous value gained from simply fixing business-as-usual issues. Just improving the performance of desktop terminals means that busy clinicians spend more time …   Related Stories

Citrix Summit 2018 Day 2: Giving Back, and Looking Ahead

Citrix Blogs - Tue, 01/09/2018 - 19:38

The past two days have been a whirlwind of networking receptions, interactive sessions, breakouts and demos. I’ve had the chance to meet so many of you, and I hope all of this nonstop activity has been energizing—and that you’ve had …

  Related Stories

QNAP to add pfSense to its products

pfSense Blog - Tue, 01/09/2018 - 18:00

You may have noticed this press release today from QNAP. Here’s the short story.

The case of Reply Log Manager not letting lagged copy lag

Exchange Team Blog - Tue, 01/09/2018 - 15:24

In a previous blog post Ross Smith IV had explained what the Replay Lag Manager is and what it does. It's a great feature that's somewhat underappreciated. We've seen a few support cases that seemed to have been opened out of the misunderstanding of what the Replay Lag Manager is doing. I wanted to cover a real world scenario I had dealt with recently with a customer that I believe will clarify some things.

What is a Replay Lag Manager?

In a nutshell, Replay Lag Manager provides higher availability for Exchange through the automatic invocation of a lagged database copy. To further explain, a lagged database copy is a database that Exchange delays committing changes to for a specified period of time. The Replay Lag Manager was first introduced in Exchange 2013 and is actually enabled by default beginning with Exchange 2016 CU1.

To understand what it is let's look at the Preferred Architecture (PA) in regards to a database layout. The PA uses 4 database copies like the following:

As you can see the 4th copy is a lagged copy. Even though we're showing it in a secondary site, it can exist in any site where a node in the same DAG resides.

The Replay Lag Manager will constantly watch for any of the three things to happen to the copies of DB1. Ross Smith's post does a wonderful job of explaining them and how Exchange will take other factors (i.e. disk IO) into consideration before invoking the lagged copy. In general, a log play down will occur:

  • When a low disk space threshold (10,000MB) is reached
  • When the lagged DB copy has physical corruption and needs to be page patched
  • When there are fewer than three available healthy HA copies for more than 24 hours

A log "play down" essentially means that Replay Lag Manager is going to force that lagged database copy to catch up on all of the changes to make that copy current. By doing this it ensures that Exchange maintains at least 3 copies of each database.

When things are less than perfect…

In the real world we don't always see Exchange setup according to our Preferred Architecture because of environment constraints or business requirements. There was a recent case that was the best example of Lag Replay Manager working in the real world. The customer had over 100 DB's, all with 6 copies each. There were 3 copies in the main site and 3 copies in the Disaster Recovery site with one of those copies at each site being lagged. The DB copies were configured like this for all databases.

As you can see in this particular instance the lagged copy at Site A was being forced to play down while the other copy showed a Replay Queue Length (RQL) of 4919. This case was opened due to the fact that the lagged DB copy at Site A was not lagging.

The customer stated that the DB was lagging fine until recently. However, after a quick check of the Replay Queue Length counter in the Daily Performance Logs it didn't appear to have ever lagged successfully for this copy.

So, what we're seeing is the database has 6 copies, 2 lagged but 1 of those lagged copies isn't lagging. Naturally, you may try removing the lag by setting the -ReplayLagTime to 0 then changing back to 7 (or what it was before). You may even try recreating the database copy thinking something was wrong with it. These still don't cause Exchange to lag this copy.

The next step is to check if it's actually the Replay Lag Manager causing the log play down. You can quickly see this by running the following command specifying the lagged DB\Server Name. On this example will use SERVER3 as the server hosting the lagged copy of DB1.

Get-MailboxDatabaseCopyStatus DB1\SERVER3 | Select Id,ReplayLagStatus
Id                                      : DB1\SERVER3
ReplayLagStatus                         : Enabled:False; PlayDownReason:LagDisabled; ReplaySuspendReason:None;
Percentage:0; Configured:7.00:00:00; MaxDelay:1.00:00:00; Actual:00:01:22

What we see is that the ReplayLagStatus is actually disabled and the PlayDownReason is LagDisabled. That tells us it's disabled but it doesn't really give us more detail as to why..

We can dig further by looking at the Microsoft-Exchange/HighAvailability log and we see a pattern of 3 events. The first event we encounter is the 708 but it doesn't give us any more information than the previous command does.

Time:     11/31/2017 3:32:55 PM
ID:       708
Level:    Information
Source: Microsoft-Exchange-HighAvailability
Machine:  server3.domain.com
Message:  Log Replay for database 'DB1' is replaying logs in the replay lag range. Reason: Replay lag has been disabled. (LogFileAge=00:06:00.8929066, ReasonCode=LagDisabled)

The second event we see has a little more information. At this point we know for sure it's the Replay Lag Manger because of its FastLagPlaydownDesired status.

Time:     11/31/2017 3:32:55 PM
ID:       2001
Level:    Warning
Source: Microsoft-Exchange-HighAvailability
Machine:  server3.domain.com
Message:  Database scanning during passive replay is disabled on 'DB1'. Explanation: FastLagPlaydownDesired.

On the third event we see the 738 which actually explains what's going on here.

Time:     11/30/2017 1:50:15 PM
ID:       738
Level:    Information
Source: Microsoft-Exchange-HighAvailability
Machine:  server3.domain.com
Message:  Replay Lag Manager suppressed a request to disable replay lag for database copy 'DB1\SERVER3' after a suppression interval of 1.00:00:00. Disable Reason: There were database availability check failures for database 'DB1' that may be lowering its availability. Availability Count: 3. Expected Availability Count: 3. Detailed error(s):
Server 'server4.domain.com' has database copy auto activation policy configuration of 'Blocked'.
Server 'server5.domain.com' has database copy auto activation policy configuration of 'Blocked'.
Server 'server6.domain.com' has database copy auto activation policy configuration of 'Blocked'.

The "Availability Count: 3. Expected Availability Count: 3." is a tad confusing but the heart the issue is in the detailed errors below that…

It's Replay Lag Manager doing it… but why?

The entire reason for this blog post comes out of the fact that we've seen the Replay Lag Manager blamed for not letting a lagged copy lag. So, the next step someone will do is to disable it. Please don't do that! It only wants to help!

Let's look at how we can resolve the our above example. The logs are showing that it's expecting 3 copies but there aren't 3 available.  How can that be? They have at least 4 copies of this database available?!? If we run the following command we see a hint at culprit.

Get-mailboxdatabasecopystatus  DB1 | Select Identity,AutoActivationPolicy
Identity          AutoActivationPolicy
--------          --------------------
DB1\SERVER1 Unrestricted
DB1\SERVER2 Unrestricted
DB1\SERVER3 Unrestricted - Lagged Copy (Not lagging)
DB1\SERVER4 Blocked
DB1\SERVER5 Blocked
DB1\SERVER6 Blocked - Lagged Copy (Working)

There it is! There are 6 database copies, however, the copies in Site B are all blocked due to the AutoActivationPolicy. Now things are starting to make sense. In the eyes of the Replay Lag Manager those copies in Site B are not available because Exchange cannot activate them automatically. So, what's happening is the Replay Lag Manager only sees the 2 copies (in the green square below) as available. Therefore, it forces a play down of the logs on the lagged copy to maintain it's 3 available copies.

That explains why the lagged copy at Site A isn't lagging but why is the lagged copy at Site B working fine? This is because from the perspective of that database there are 3 available copies in Site A once that lagged copy was played down.

That's cool… how do I fix it?

There are essentially two ways to resolve this example and allow that lagged copy at Site A to properly lag.

The first way is to revisit the decision to block Auto Activation at Site B. The mindset in this particular instance was that their other site was actually for Disaster Recovery. They wanted some manual intervention if databases needed to fail over to the DR site. That's all well and good but it doesn't allow for a lagged copy at Site A to work properly due to the Replay Lag Manager. The customer did actually end up allowing 1 copy at the DR site (site B in our example) for Auto Activation. To do this you can run the following command:

Set-Mailboxdatabasecopystatus SERVER4\DB1 -AutoActivationPolicy Allowed

The other option here would be to create another database copy at Site A. Obviously, that's going to require a lot more effort and storage. However, doing this would allow for the Replay Lag Manager to resume lagging on the lagged database copy.

I hope this post clarifies some things in regards to the Replay Lag Manager. It's a great feature that will provide some automation in keeping your Exchange databases highly available.

Michael Schatte

Congratulations Citrix Summit 2018 Strategic Partner WINNERS!

Citrix Blogs - Tue, 01/09/2018 - 11:00
Now in their second year, the Global Strategic Partner awards honor the high achieving members of our global partner ecosystem.

The Strategic Partner Awards highlight the real and ongoing innovation in workspace mobility, networking, security and cloud from our partner …

  Related Stories


Subscribe to Fix the Exchange! aggregator