Within this one page you'll have all information on the Citrix and Intel relationship available at your fingertips...or mouse click so to speak.  

Interested in learning about Citrix's latest activities with Intel? We know your time is valuable and that's why we've consolidated all related news, blogs, articles, videos, etc... into the feeds section of this page. We've also created an area that highlights all related forum threads and postings. Look for answers to your burning questions, participate in an existing discussion, or post a new thread. Epitomizing the true nature of community... this page is here for you!  

Did you know that Citrix and Intel are working together to jointly develop a bare-metal Type-1 client hypervisor? Learn about and stay up to date on our joint collaboration of Citrix XenClient. View demos, read blogs and let us know what you think about this upcoming solution.

But it doesn't stop there! Get information and results on our joint product validation, benchmarking and scalability tests. You can even let us know how you're working with Citrix and Intel solutions together in our Community Verified area.

You're only one click away from all things Citrix and Intel. Stop by the Citrix Community page for Intel and check back often! And don't forget to follow us on Twitter!

To finish off with another Damone line, "isn't this great?"

"bnsmdf.sys" driver is a simple scsi miniport filter driver that is responsible for handling /reporting device power state change, device error or status change, etc.  Please check if it happens on some specific hardware or all machines. Also, are they using target-side cache for this?

 "bnsmdf.sys"ドライバは、シンプルなscsiミニポートフィルタドライバで、デバイスの電源状態の変更、デバイスのエラーなどを

処理し、レポートします。

特定のハードウェアに依存するかどうか、複数のマシンで発生するのかを確認します。

また、キャッシュの設定について確認します。

[質問]
キャッシュ方式をローカルHDDキャッシュにして起動したのですが
サーバサイドキャッシュになってしまいます。
[回答]
PVS のターゲットデバイスが起動する際に、ドライバの起動時
（起動の途中で画面背景が黒から青に切り替わる時）に
サーバのレスポンスが悪い場合に、Write Cache が "Local-HDDキャッシュ"
から　"サーバキャッシュ" へと切り替わることがございます。
これは現在のPVSの実装による、期待する動作となります。
サーバキャッシュへスイッチされるトリガーとなる条件は、
ローカルキャッシュのHDDがフルになる場合です。
この機能がない場合、PVS のクライアントソフトウェアは、
キャッシュへデータ書き込みがこれ以上できなくなった際に、
ターゲットデバイスが適切に機能し続けることができなくなるためです。
RAM キャッシュでは、BSODを起こさないように、
PVS のクライアントソフトウェアは、ライトキャッシュがサーバに
移動されることで、PVS のクライアントソフトウェアが復旧し、
機能し続けることができます。この移動がなければ、
ターゲットデバイスはハング、またはBSODを起こすことになります。
 
参考までに、英語の情報も添付いたします。
参照ください。
The condition that triggers the switch to Server cache is when the hard drive on the local cache becomes full. Without this feature the target device would discontinue to function properly when the PVS client software could not write any more data to the cache. Unlike Ram cache, the condition results in a BSOD, the PVS client software is able to recover and continue to function once the write cache is moved to the server. Without this move the target device may hang or cause a BSOD.
 Linux については、Windowsとは異なり、クライアントローカルDiskキャッシュからサーバキャッシュへスイッチする機能は、サポートされていません。
参考:
Linux client does not support cache failover from the local HDD to the server.

"Citrix Provisioning Services 5.1 Administrator's Guide"
"Citrix Provisioning Services 5.1 Installation and Configuration Guide"

 http://support.citrix.com/product/provsvr/psv5.1/#tab-doc

If you select Standard Image and Cache in target device RAM, select the cache
size in megabytes.
The max size of the RAM write cache is determined by the registry setting
WcMaxRamCacheMB in the BNIStack Parameters. This is a DWORD
parameter. If the registry entry does not exist, then the default value used is 3584
MB.

 Standard Image と Cache in target device RAM　を利用している場合、キャッシュサイズをMBで設定します。

RAMの書き込みキャッシュの最大値は、 BNIStackパラメータのWcMaxRamCacheMBレジストリで決定されます。

これは DWORD値です。もし設定されていない場合は、デフォルトで3584MBです。

This setting enables the bootstrap to work with newer Windows OS versions
and is enabled by default. Only disable this setting on older XP or Windows
Server OS 32 bit versions that do not support PAE, or if your target device is
hanging or behaving erratically in early boot phase.
 

ターゲットデバイスの起動初期のフェーズでターゲットデバイスがハングする、または予期しない動作をする場合、

Jetデータベース（.mdb）の破壊の原因と対処KB一覧

http://www.users.gr.jp/blogs/fullmoon/archive/2004/11/25/6267.aspx

http://forums.citrix.com/thread.jspa?threadID=242484&tstart=0

http://forums.citrix.com/message.jspa?messageID=1366389

Currently, using "Network service account" instead of "Specified user" is the only workaround.

「ドメインに接続できません。ドメインコントローラがダウンしているか利用
できない状態になっている。またはコンピュータアカウントがみつからなかった
ことが原因と考えられます。後からもう一度やり直してください。」
The security database on the server does not have a computer account for this workstation trust relationship.

DC 上のイベントログには、下記のイベントが記録されている。

-------------------------------------------------
「イベントの種類: エラー
イベント ソース: NETLOGON
イベント カテゴリ: なし
イベント ID: 5722
日付: xxxx
時刻: xxxx
ユーザー: N/A
コンピュータ: xxxx01
説明:
コンピュータ xxxx2 からのセッション設定を認証できませんでした。 セキュリティ データベースで参照されたアカウント名は xxxx2$ です。次のエラーが発生しました:
アクセスが拒否されました。

Just in case you haven't seen the posts by Ed York and Michael Bogobowicz on Workflow Studio recently I wanted to call them out. In addition to providing some excellent activities and workflows for you, there are posts that explain the process of creating activities and workflows around XenApp, Provisioning Server, and SQL.

Michael wrote a workflow that automates the update of vDisks in Provisioning Server using the CLI tools:
http://community.citrix.com/display/wf/PVS+Automatic+vDisk+Update+Script

Ed has written up two excellent blog series on Workflow Studio:

• Workflow Studio SDK - This first series of posts detail Ed's experiences using the Workflow Studio SDK to build and debug custom activities and are an excellent place to start if you want to build your own activities:
  http://community.citrix.com/x/YIUJB
  Ed followed this series up with a post on building an activity to get a list of applications from XenApp using the MFCOM SDK and this activity and source code are available for download:
  http://community.citrix.com/x/KgE-B
  Ed also wrote a couple of activities that work with SQL Server:
  http://community.citrix.com/x/2IJiB

• Automating Workflow Studio - The second series of posts is on automating workflows. Ed has posts that explain how to use the PowerShell interface to start a workflow and pass parameters to it. He then goes on to explain how to build your own front-end for starting these workflows:
  http://community.citrix.com/x/ToNiB

As a reminder, all community posts on activity libraries and workflows will show up right in the Workflow Studio product under the Community tab so you don't even have to go out to the site to look. RSS feeds are also available on the Community site if you want to get notified in your favorite RSS reader.

PVS 3.x 以前では、ソケットにblocking I/Oのみ実装し利用していました。
PVS 4.0 を設計する際に、検証に利用したハードウェアにおいて、
non-blocking I/Oがblocking I/Oより30%以上の速度の向上が確認されました。
この結果により、non-blocking I/Oをデフォルトに設定いたしました。
また、検証することができないハードウェアにおける予期しない影響による重大な
問題を避けるために、blocking I/Oを有効にできるようにオプションを残しました。

Blocking I/O = Synchronous（同期） I/O.  
つまり、すべてのソケットコールは、返信の前に完了を待たなければなりません。

Non-blocking I/O = Asynchronous（非同期） I/O.  
つまり、すべてのソケットコールはすぐに返信します。
I/Oが完了したかどうかは、イベントを待つかステータスを確認しなければなりません。

一般的にはNon-blocking I/O (非同期 I/O) がより高速になります。
しかし、速度は、ハードウェア、OSのバージョンに大きく依存します。
したがって、ある特定のハードウェアの組み合わせにおいては、
blocking (Synchronous) I/Oがより高速に動作することもあります。

It's really there for legacy reasons.  PVS 3.X and previous used blocking I/O on sockets.   When we designed 4.0 our testing showed that non-blocking I/O was about 30% faster on the hardware we had to test with.  So we used non-blocking I/O as the default.  But in order to avoid any major problems with unforeseen consequences on hardware we could not test with, we left blocking I/O in as an option the customer could enable. 

Blocking I/O = Synchronous I/O.  i.e. all socket calls must wait for completion before returning.

Non-blocking I/O = Asynchronous I/O.  i.e. all socket calls return immediately.  We must wait for an event or check a status to determine when I/O completes.

Non-blocking I/O (Asynchronous I/O) is generally faster.  But much depends on hardware, OS version, etc.  So both versions are included "just in case" some particular hardware combination works better with blocking (Synchronous) I/O. 
 

Microsoft SQL Server Management Studio Express
http://www.microsoft.com/DOWNLOADS/details.aspx?familyid=C243A5AE-4BD1-4E3D-94B8-5A0F62BF7796&displaylang=ja

SQL Expressを運用の場合、PVSインストール時にインストールされる
"SQL Server 2005 Surface Area Configuration" において、
Local and remote connections, Using both TCP/IP and named pipes を
確認します。

PVS で利用されているSQL のデータモデル、テーブルなどの情報については、
一般に公開された情報はございません。予めご理解いただければと存じます。

Select PAE Mode option when the target device's operating system is enabled with PAE (such as Windows Vista), and the Cache Type in the vDisk properties is set to Cache in device RAM.

Citrix Provisioning Server 5.0 (SP1, SP1a) Administrator's Guide
http://support.citrix.com/article/CTX117916
-----

その他、参考情報です。
PAE
http://msdn.microsoft.com/en-us/library/ms791485.aspx
BCDEdit /set
http://msdn.microsoft.com/en-us/library/aa906211.aspx

○Socket Communicationsについて
Socket Communications — ネットワーク通信のnon-blocking I/Oを有効・無効し
します。一般的に、Non-blockingソケットはblockingソケットよりも（より少ない
CPUサイクルを消費する）より効果的です。したがって、non-blockingソケットは
ネットワークのスループットを増加することになります。しかし、non-blockingソケットは
パケットが欠落することも考慮しているので、デバイスへのリトライも増加傾向となります。
つまり、Provisioning ServerのOS、バージョン、およびサーバーのNICドライバー
・ハードウェアを考慮し、この設定を行います。

PVS 4.5.1 では、non-blocking I/Oを設定する場合、多くのリトライを発生します。
PVS 4.5.2 では、クライアント側でいくつかの重要なアーキテクチャの改善がされ、
リトライへの改善がされています。

Socket Communications — Enable/disable non-blocking I/O for network
communications. Non-blocking sockets generally work more efficiently
(consume fewer CPU cycles) than blocking sockets. Non-blocking sockets
therefore usually increase network throughput. However, non-blocking
sockets also increase the likelihood of dropped packets and therefore tend
to increase device retries. Set to whichever setting works best with your
version of the Provisioning Server OS and server NIC drivers/hardware

PVS 4.5.1 may be causing a lot of retries when using non-blocking I/O. Some major architecture changes on the client in the 4.5.2 release may have made some of these retries go away.

So to clarify:

PVS 4.5.1
Non-Blocking I/O set is causing network utilization to go to 30% (during boot?)
Blocking I/O uses 10% ?

Is this what the ticket is saying, Non-Blocking I/O is meant to decrease the CPU utilization of the PVS server and increase the Network utilization. What are their boot times and retry counts in both scenarios and they precisely explain the testing results and methods of testing?

suggesion at this time is to upgrade 4.5.2

  NEC Express5800/51lc x 1
  http://www.express.nec.co.jp/teci/tecbook-pdf/50tb050210/51Lc.html
  NEC Express5800/51le x 2
  http://support.express.nec.co.jp/teci/tecbook-pdf/50tb070420/51Le.pdf

NIC:
Intel 82547GI
http://www.express.nec.co.jp/teci/tecbook-pdf/50tb050210/51Lc.html
Driver: Intel(R) PRO/1000 CT Network Connection
version 8.10.3.0

Provisioning Serverの推奨ネットワーク設定

http://support.citrix.com/article/CTX118706

http://buffalo.jp/products/catalog/network/lsw3-gt-5ep/

http://qa.buffalo.jp/eservice/esupport/consumer/esupport.asp?searchtype=normal&id=BUF4479

 http://buffalo.jp/products/new/2008/000648.html

●高速ネットワーク「Gigabit Ethernet（1000BASE-T）」と効率的な転送を実現する「Jumbo Frame」に対応

　全16ポートGiga(1000M)対応のスイッチングHub。高速なGigaネットワークを引き出す「Jumbo Frame　16,000Byte※(ヘッダ14Byte+FCS 4Byte含む)」で　Ethernetの送単位を大きくして、Gigabitの高速性能を引き出します。

※ワイヤーレートは12,000Byteまで

●EAPOL（IEEE802.1X）及びBPDUフレーム透過機能に対応

　MACベースのIEEE802.1X認証環境において必要とされる『EAPOL（IEEE802.1X）フレーム）透過機能』に対応。セキュアなネットワーク構築に最適です。また、STP利用時に用いられるBPDUフレーム透過にも対応しています。

It has come to my attention that the PVS product guide does not clearly state that it is only the server that fails over and not the vDisk. For example, if the vDisk is on a shared store and the HA server currently providing that vDisk goes offline, another HA server (previously configured; not necessarily the server used to login) provides that same vDisk.

I edited the Managing HA chapter in the guide and removed the plural use of vDisk, which I believe may have caused that confusion. I have attached the updated version. Please read through the first few pages and let me know if this helps clarify this point.

PVSトラブルシューティングの確認事項

PVS のstream.logログ、および Target Device のVirtual Disk Status のStatistic の
取得をお願いいたします。

○ stream_logについて
注意：PVS 4.5 SP2 または PVS 5.0 SP1a 以降のみ詳細なログが取得可能です。
それ以前のバージョンではログが取得できませんので、ご注意ください。
PVS 4.5 SP2の場合

　C:\Program Files\Citrix\Provisioning Server\stream_log.config 、
PVS 5.0 SP1a以降の場合
　C:\Documents and Settings\All Users\Application Data\Citrix\Provisioning Server\stream_log.config

または　PVS 5.1以降の場合(Provisioning Services Consoleから設定することもできます)
　C:\Documents and Settings\All Users\Application Data\Citrix\Provisioning Services\stream_log.config
において、ログのレベルをより詳細なTRACEへ変更し、クライアント起動時のログを取得ください。
例：
　<level value="ERROR" />
を
　<level value="TRACE" />
へ変更します。

○ 起動のタイミングの詳細
起動時のどの段階で発生しているのかわかりましたら、詳細をお教えください。
DHCPサーバからのクライアントのアドレス取得、
PVSとの接続（PXE Bootstrap よりVerbose Modeを選択することで詳細アドレスを表示）、
または、vDiskよりのOSの起動。
起動のタイミング
下記の起動時の流れの中で、どの時点で発生しますでしょうか。
(例：Vista SP1 の場合）
 電源On
 　↓
 DHCPサーバからのクライアントのアドレス取得
 　↓
 Citrixのロゴ表示
 　↓
 PVSに接続し以下のメッセージが表示される。
 　Connecting to the Provisioning Server.
 　virtual disk found.
 　
 Verbose Modeの場合は　IPアドレスが表示
 　↓
 真っ黒画面表示
 　↓
 Vistaのスプラッシュ画面表示(起動バーが流れる画面)

たとえば、まず切り分けの判断基準
Power ONからDHCP, DHCP Optionをゲットするまで→DHCPサーバー
NBPファイルを落としてくるまで→TFTPサーバー
NBPファイルを読み込み、PVSにLogin Request, MACからvDisk Assignmentをゲットするまで→NBPファイル
StreamServiceからvDiskをストリームしてOS軌道まで→StreamService

○ ネットワークの構成
単独の閉じられたネットワークでも事象は発生、確認されていますでしょうか。
また、以下の点も確認いただけますでしょうか。
1. DNSサーバーの設定
 vDiskのネットワークプロパティの設定において、
 DNSサーバーのアドレスを設定する（動的に取得しない）場合の動作の違い。

2.DHCPサーバ上のスコープOption66,67について
 PXE Serviceを使用とのことですが、ブートサーバーホスト名および
 ブートファイル名を設定する場合の動作の違い。

Microsoft DHCPサーバーのDHCP PXEオプションを追加する方法
http://support.citrix.com/article/CTX118513
How to Configure Option 60 on a DHCP Server
http://support.citrix.com/article/CTX115212

3. Provisioning Serverの推奨ネットワーク設定
 http://support.citrix.com/article/CTX118706

 Large Send Offloadオプションについて
 Provisioning Server Target Devices Lock up Shortly Before or After Logging on to Windows
 http://support.citrix.com/article/ctx116814
 Registry Settings to Improve Failover Times
 http://support.citrix.com/article/ctx119223
 The Image Build Process Takes a Long Time to Complete
 http://support.citrix.com/article/ctx115658

○ PVS のネットワークトレースの取得
Wireshark (http://www.wireshark.org/) をPVS サーバーにインストールし、
クライアント起動時のPVS サーバーからのネットワークのすべてのパケットを
すべてキャプチャーします。
その際のログを保存いただき、返信いただければと存じます。
ネットワークの状況がクライアントの起動に影響していないかどうか確認できることを
見込んでいます。

○ NICのドライバーおよびROM
マシン製造元またはNIC製造元にもお問い合わせいただき、
NIC の最新のROM、および NIC の最新のドライバーへアップグレードし、
動作を確認ください。

Citrix Provisioning Server 5.0 (SP1, SP1a) Installation and Configuration Guide
http://support.citrix.com/article/CTX117917
Network Card　PXE 0.99j　以降

○ Citrix Knowledge Center
http://support.citrix.com/product/provsvr/psv5.0/topic/troubleshoot/
http://support.citrix.com/product/provsvr/psv4.5/topic/troubleshoot/

○ローカルHDDキャッシュ

PVSクライアントをVdisk（Standardモード、EncryptedローカルHDDキャッシュ）から起動し、

何らかの処理により、おそらくOut of Virtual Memory が発生し、ハングアップしてしまう。

おそらくは、Virtual Memory の管理系かと思われるが、開発元では原因究明できていないようです。

回避策は、 ページファイルの配置場所を、デフォルトのCドライブからDドライブ（ローカルHDD）に変更。

This was posted some time ago, but I suspect that there are many that may have missed it since there are so many new virtual desktop projects springing up all the time. So I'm bringing it back up to the top for more exposure!

This blog has been invaluable for our field, for our technical marketing staff and for our event demo preparations. It's called "Windows XP Performance Optimizations for XenDesktop and Provisioning Server vDisks."

There's more to be discussed on how to successfully make that P2V conversion of a desktop, and we are working on additional white papers that will outline some of the real world best practices we've uncovered - so stay tuned!

New courseware has been released from Citrix Education - Get up to speed on Citrix XenApp 5.0, XenDesktop 3, NetScaler 9 and more! Click on the course/certification title below for more information on these hot releases.

Instructor-led Courses

• CXA-100-1I Providing Help Desk Support for Citrix XenApp
  Prepare to troubleshoot common XenApp issues and identify whether complex issues are related to the XenApp product, the network or directory services.
• CXD-200-1I Implementing Citrix XenDesktop 3
  Learn to install, configure and manage Citrix XenDesktop 3.
• CMB-201-1I Implementing Citrix XenApp Platinum Edition Components
  Learn to properly design, implement, and support Citrix XenApp Platinum Edition components, including EdgeSight 5.0, Access Gateway 9.0 Enterprise Edition, Password Manager 4.5 and WANScaler 4.3. For experienced IT professionals.
  

Self-paced Online Courses

• CNS-200-1W Basic Administration for Citrix NetScaler 9.0
  Gain an understanding of NetScaler features such as load balancing, SSL offload, classic and advanced expressions, compression, auditing and logging and monitoring.
• CPV-200-1W Implementing Citrix Provisioning Server 5.0
  Learn to install, configure and administer Citrix Provisioning Server 5.0 farms.
  

Certification

• Citrix Certified Administrator (CCA) for Citrix NetScaler 9
  Validate expertise in the implementation and administration of NetScaler 9.
  

With users scattered across the globe, the need to tighten IT management is ever-increasing.
Citrix XenDesktop working alongside Intel vPro-based devices provides IT centralized desktop management with complete control over the PC.

Citrix XenDesktop 3, deployed in conjunction with Intel® vPro™ technology, delivers an uncompromised IT centralized manageability and end user experience. This combination provides IT with simple, secure remote desktop management and hardware-based security, resulting in reduced TCO and providing end users with a full desktop PC experience. Together, Citrix and Intel provide a robust desktop and PC management solution.

Join us on May 20th for a XenDesktop Tech Talk session, featuring Intel. This webinar will provide a deep-dive introduction on how these solutions work together and the value they provide IT organizations.

Register Today!

Date: 5/20/2009
Time: 1:00pm - 2:00pm
https://www1.gotomeeting.com/register/211934248

 Keith McLaughlin, Escalation Engineer
 
Keith McLaughlin is a Provisioning Server expert on the Citrix Technical Support Escalation team, joining the team when Citrix acquired Ardence about two years ago. Keith filled us in on the two sessions that he'll present at Citrix TechEdge during Citrix Summit and Synergy 2009: End-to-end virtualization with Citrix Delivery Center, with a focus on Active Directory integration with Provisioning Server, and then his in-depth session will be on Planning and implementing a Provisioning Server high availability (HA) solution.

Q. How has Provisioning Server improved from a support perspective over the past year?
Keith: The biggest improvement this year is the addition of the Streaming Service Logs.  These logs, which came out as part of 5.0 SP1 are extremely helpful in narrowing down the issue.

 Q. What Provisioning Server and Citrix Delivery Center tips will attendees learn at your session this year?
Keith: This year's session is focused on High Availability. In the session we are going to go over troubleshooting procedures and explain in depth what happens when a Target Device fails over and how to track that failover through the logs files. For the Citrix Delivery Center session, I'll focus on Active Directory integration with Provisioning Server Standard Image.

Q: What Provisioning Server Tech Tip can you give people now?

Keith: When planning your Provisioning Server deployment, give the Target Devices unique names in the Provisioning Server Console. Do not use the hostname of the machine that is being imaged as the name of the Target Device. This avoids conflicts when booting the Target Device from the Vdisk.

Q. What new tools or techniques are you using to troubleshoot Provisioning Server?

Keith: A year of working with the 5.0 is probably the biggest factor.  5.0 had many improvements over the previous version and many architecture changes.  Also seeing where customers and end users were running into problems and being able to identify the symptoms because of past experiences greatly cuts down on troubleshooting time.

 Q. What types of cases have you worked on this past year? Why?

Keith: As part of the Provisioning Server Escalation group, I have covered a lot of different issues ranging from Active Directory integration to tracking down possible bottlenecks on customers networks that could be causing timeouts on the provisioning server.

 About Keith McLaughlin

Keith's been with Citrix Technical Support for two years.  He holds certifications in Citrix Certified Administrator, CCA, for Provisioning Server and XenServer. During his free time Keith loves playing the guitar, and his favorite artist is Stevie Ray Vaughan.

Do you have a Provisioning Server troubleshooting area that you would like Keith to focus on during his presentation? Leave a comment.

 Want to learn more about TechEdge 2009, www.citrix.com/techedge. Stay tuned for our weekly close-up interviews on the TechEdge presenters.

Posts in this series:

• Citrix TechEdge - Troubleshooting Tips and Tricks

• Interview 1: Close-up with Jamie Baker

• Interview 2: Close-up with Keith McLaughlin  

Jamie Baker, Sr. Escalation Engineer 

Without a doubt one of Citrix Technical Support's top engineers is Jamie Baker. Jamie works with the Strategic Managed Accounts Resolution team, SMART, which handles 24 hour coverage on critical and high priority issues, and he also is a Subject Matter Expert, SME, in various areas such as Printing, Application Streaming and currently the XenDesktop, SME, for the SMART team. Jamie took a few moments out of his busy day of solving customer issues to answer some questions on the sessions that he'll be presenting at Citrix TechEdge.

The sessions he'll be covering are End-to-end virtualization with Citrix Delivery Center with a focus on XenDesktop, and then presenting an in-depth session on XenDesktop 3 architecture and design. TechEdge is the new name for the Citrix Support and Engineering Institute of Technology, CSEIT. The event is hosted by top Citrix Technical Support engineers at Citrix Summit and Synergy, and is the event for troubleshooting your Citrix Application Delivery environment

 Q. How has XenDesktop improved from a support perspective over the past year?

Jamie: The release of XenDesktop 3.0 has closed a lot of issues, added USB device support which was a big need for a lot of customers and provided pool management failover when using XenServer as the desktop hosting infrastructure. The new failover allows the administrator to configure multiple pool masters for the same XenServer pool in the XenDesktop Management Console and if the original XenServer pool master fails, the Pool Management service will fail over to the secondary pool master. This makes the product much more resilient by eliminating a single point of failure, which is always more supportable

Q. What XenDesktop and Citrix Delivery Center tips will attendees learn at your session this year?  

Jamie: We will provide tips on both how to design your XenDesktop infrastructure to take advantage of the new features in XenDesktop 3.0 and how to implement those new features, including USB device support. We will also provide tips on how to configure XenDesktop to take advantage of the robust hosting infrastructure provided by XenServer.

The Citrix Delivery Center Session will highlight how each of the components of the Citrix Deliver Center can integrate to provide secure and robust access to desktops and applications. We'll focus on each product and the features it brings specifically to complement the entire Citrix Delivery Center.

Q. What new tools or techniques are you using to troubleshoot XenDesktop?

Jamie: Besides from having a year's more experience working with customer issues and the new products, we've developed our internal tools to more quickly identify XenDesktop issues. We've updated the CDFControl utility to allow for remote CDF trace on VDA and DDC machines. This allows us to more easily set up a diagnostic trace and resolve issues more quickly.

This year we've also stood up a public symbol server. This allows us to access more diagnostic data through system dumps and memory dumps without having to bring those dump files in house. In time sensitive situations, this can save us hours of time. It also allows customers and other vendors to diagnose issues from dump files more quickly.

Q. What types of cases have you worked on this past year? Why?

Jamie: For me, being on the critical situation team, I've focused on identifying single points of failure and issues that have the potential for wide impact. The pool management service's ability to connect to a hosting infrastructure and to recover quickly in case of an issue with the infrastructure is the biggest issue we tackled this year. As mentioned above, we were able to build in failover when using XenServer infrastructure. The addition of HDX technology has improved multi-monitor display handling as well as USB device remoting and multi-media display.
 

About Jamie Baker

Jamie's been with Citrix Systems for six years, and all of those with Citrix Technical Support. He started out as a first level frontline support engineer, supporting MetaFrameXP. After two years on the phones, he moved to the Escalation team, and a year later moved to the SMART team. Jamie is currently a Sr. Escalation Engineer, and holds certifications as a CCA in XenDesktop and XenApp, as well as a Windows MCSE.

When Jamie isn't working, he loves to spend time with his wife and three year old twin daughters, exploring all the playgrounds that North Georgia has to offer. On Fall Sundays, you'll find him watching the Philadelphia Eagles and trying desperately to control his emotions so he doesn't scare the dog.

Do you have a XenDesktop troubleshooting area that you would like Jamie to focus on during his presentation? Leave a comment.

TechEdge 2009 sessions:

• End-to-end virtualization with Citrix Delivery Center
• XenDesktop 3 architecture and design
• Planning and implementing a Provisioning Server high availability (HA) solution
• Integrating and troubleshooting Citrix Access Gateway
• Enterprise Edition Advanced troubleshooting of Citrix NetScaler
• XenServer disaster recovery and automation
• Troubleshooting tools and methodology for Citrix XenApp 5 environments

Want to learn more about TechEdge 2009, www.citrix.com/techedge. Stay tuned for our weekly close-up interviews on the TechEdge presenters.

Posts in this series:

• Citrix TechEdge - Troubleshooting Tips and Tricks

• Interview 1: Close-up with Jamie Baker