Our community is talking about the new Dell Technologies. Join the discussion in the Dell EMC Community Network:
Dell: Can you introduce yourself? What are you doing and whom are you working for?Hendrik Volkmer: I am co-founder of a company called Cloudbau. We are based out of Berlin and we are doing OpenStack deployments, DevOps consulting and Chef training.
Q: You recently wrote a blog post series "There will be no reliable cloud". Can you tell about that?A: There is this notion that you can’t use the cloud right now because it is too unreliable and we should really wait until it becomes reliable and then we can put all these business related applications on it. My view is a little bit different. I think it won’t get any more reliable. As Amazon shows all kind of people really like to bang that it’s so unreliable. Let’s see what Google Cloud Service will provide. I think it’s just a matter of scale that you cannot provide a really reliable service at this scale and so that’s basically what the series is about. And I think people should just accept that and try to make the applications reliable in a way that they can accept failure and make use of the cloud anyway and build reliable services instead of waiting for reliable infrastructure underneath.
Q: Can you give us any real life examples for that sort of approach?A: I think the best example is probably Netflix. They are doing a lot of work to get their service reliable on the unreliable Amazon infrastructure. And they have really good presentations, blueprints on the architecture and how you could actually achieve this kind of resiliency.
Q: What is your view on OpenStack? Where do you think does OpenStack shine? - public cloud or private cloud or both?A: I think there are a few well-known public clouds based on OpenStack. Rackspace and HP Cloud are the most prominent examples. I think though that OpenStack currently is more in the private cloud space because people really like that they can actually deploy OpenStack themselves. And from my experience I think this will settle and there will be some kind of blueprints or known good configurations of OpenStack private cloud settings that you can just use and deploy and then have your own cloud. And even if you just want to have your own cloud infrastructure for your internal services or if you have a hybrid approach … this is where it really becomes interesting when there more public cloud services based on OpenStack because you can actually use public cloud with the same API as the private cloud.
Q: What is your definition of DevOps?A: It’s about making the dev and ops teams working together and not really creating a new team. There’s actually a really good blog post called “There’s no such thing as a DevOps team”. And I agree because I think it's really an approach of dev and ops guys working together and trying to understand what the limitations and the plans of each group are. Dev developers normally wouldn’t just really care about operations. They just want to get the stuff out. Today I think the focus is more on rapid delivery. This has to include some kind of understanding of how the operations will work. And the operations guys really want to have the stuff working, because they are there to make it work and if something fails they are the ones who get the call at night. If these groups work together, you can actually achieve fast delivery with a very high service quality.
Q: Your company Cloudbau is co-sponsoring devopsdays. What are your expectations?A: We hope to get more interesting contacts, we talk to a lot of people here at the event, share this mindset of getting to work together and actually using these new technologies to achieve better service delivery. And what I think is most interesting in all these conferences will always be case studies, experience reports, war stories of people who are actually doing stuff in a different way and explaining how they do it and we can all learn from them.
Q: Do you see any differences in meet-ups like these compared to other regions such as United States?A: I think generally in United States people are more progressive, whereas in Europe or especially in Germany, people are more conservative. They're waiting until this stuff actually works and then they're trying it. Obviously, in Berlin it’s a little bit different because there are a lot of start ups that are actually utilizing those technologies they would use in the U.S. But generally speaking, I think, the German IT space is very conservative and that’s why I think cloud technologies aren’t that established in Germany like they are in the U.S..
Q: Okay, Hendrik. Thank you very much.A: Thank you.
Hendrik’s blog, Twitter and Github.
Yesterday, at the BUILD conference, Microsoft announced the availability of new SkyDrive Pro apps for Windows 8 and Apple’s iOS. This follows Microsoft’s free release of the SkyDrive Pro desktop client, which was originally tied to Office 2013.
Windows 8 Client for SkyDrive Pro
iOS Client (from Microsoft.com)
For years, some users have tried to use a variety of offline configurations to bring their documents with them using some rudimentary offline access built into Outlook. But increasingly, Microsoft has recognized the future of collaboration is outside Outlook, and has been bring solutions into the market with SkyDrive and SkyDrive Pro to empower document centric collaboration beyond the boundaries of email. Because, let’s face it, the last thing any of us want is more time in email!
It’s also important to note that these clients support access to Office 365 versions of SharePoint – not on premises editions.
Let that sink in. For a while, O365 has trailed on premises SharePoint in functionality. But with the “Wave 15” release, many of those gaps were closed. Last year, I noted in SharePoint Pro that someday, Microsoft would start releasing functions for the cloud first, making the on-premises/O365 debate far more challenging.
So, that happened.
Microsoft has more details on the SkyDrive Pro app for Windows 8 and SkyDrive Pro app for iOS on Office.com. Let’s hope the on premises versions don’t lag too far behind.
You know the chills you get when you're watching a good flick - the action is rising to a climax but our hero is down and out. The music spells the loss of good triumphing over evil. But you, yes, you, Mr. & Mrs. Movie Watcher just know that something has to turn. Movies don't end up with the bad guy winning, the zombies taking over or the world ending in a freak snow storm caused by ancient aliens.
And its at that moment when all hope is lost that something happens. A lone spectator begins to clap. The mad genius makes a mistake. The President of the United States gives a rousing speech on behalf of mankind to give it one last go against the alien invasion:
'"We will not go quietly into the night! We will not vanish without a fight! We're going to live on, we're going to survive.' Today we celebrate our independence day!"
YEAH! They will not take our wives or our freedom! They will not take me lucky charms! I will fight! I will innovate! I will go to SharePoint Conference 2014 in Vegas!
Wow - an awesome speech - much better than anything in real life because real life doesn't have moving musical scores rising in the background, rousing me to action! And Independence Day is so much better than the President's speech in Armageddon - a meteor barrelling towards earth is so less believable than Bill Pullman as President of the free world. Duh!
Its on this epic scale that Microsoft has updated its What is SharePoint video from boring old Monique of yore to this amazingly cool, chill inducing video:
You go Microsoft!
Microsoft recently announced public preview versions of Windows Server 2012 R2 and System Center 2012 R2 at the TechEd Europe 2013 Conference in Madrid this week. Here are a some resources you can use to get an early look at these preview releases. In addition to Windows Server 2012 R2 and System Center 2012 R2 Preview downloads, there are also evaluations for Windows Intune, SQL Server 2014 and the Windows Azure Pack. Check them out.
MS TechNet Resources:
Microsoft Preview Evaluation Releases
What's New in Windows Server 2012 R2
What's New in Windows Server 2012 R2 Essentials
What's New in System Center 2012 R2
Evaluation Guide for System Center 2012 R2 and the Windows Azure Pack
MS Virtual Academy Jump Start Events:
What's New in Windows Server 2012 R2 Jump Start
What's New in System Center 2012 R2 Jump Start
Lately my 15 months old daughter has picked up a bad habit. Here’s how it goes.
Before she came along, my minor verbal infractions were relegated to private moments when something unexpectedly bad happened:
You get the picture. But now, with a child, those private “Oh Ships” are caught, consumed and repeated – and not just in the car – in church, on walks, in an airplane. Now the whole world knows I secretly have the mouth of a sailor.
In SharePoint, we all have those “Oh ship” moments. It happens when we delete an important document or get click happy on the site settings page and blow the site to smithereens before we realize what we just did. And lost SharePoint data isn’t just a private pain administrator’s experience. Just like my daughter exposing my “Shut the Front Door” moments, so to do Users who interact with that data daily. It happens when a user comes running to our desks frantic, yelling and pleading all at once for the sales presentation he had on SharePoint that is just now gone and which he needs in a few hours.
But the pain doesn’t stop there. We feel the full effects of my daughter’s favorite word when we try to restore the site because it’s not sitting neatly in our recycle bin. And speaking of recycle bins, ours are so overloaded we can’t easily find the salesman’s presentation. Restoring a document with native tools takes hours and scales up to days and weeks when you’re talking about a farm restore. Everything else in between takes just as long and is just as much of a bear.
I don’t really need to expound upon the recovery process. We’ve all experienced it. We all know how painful it is (if you have yet to then check out these processes for various native restore scenarios here).
Basically a lot more "Oh ship" moments are in store for your native SharePoint restores, but Recovery Manager for SharePoint makes recovery fast and simple. Just take a look at the screenshot below to see one example of how simple it is to restore a single item:
Recovery Manager gives you an easy to use interface where you can search all backups made for a particular content database and across Recycle Bins for a file name, partial file name, file extension and so forth. Who wants to try to restore a single item natively when you can do the SharePoint recovery with Recovery Manager?
Recovery Manager also restores full farms, web applications, site collections, sites, and on down to the item and document level.
For those “Oh Ship” moment, remember to Download the Free Trial of Recovery Manager so you can be the hero!
Dell: Can you introduce yourself and tell us about your company that you just founded.Hui Cheng: I am an early OpenStack evangelist in China and once led a OpenStack dev team at Sina. Now I am a new entrepreneur, I just founded a OpenStack company UnitedStack, it is a new start up targeting the Chinese and the Asia market, since the U.S. market is so crowded. We are the first professional OpenStack start-up in China and even in Asia, we gather a group of outstanding OpenStack developers as well as some top engineers in networking, storage and DevOps.
Q: What exactly is UnitedStack doing?A: We are building a cloud OS, UnitedStack OS, which is completely different from all the so-called cloud OSes in the market. We want to build a commoditized and self-serviced cloud OS which will be as easy as Windows or MacOS. Everybody who has interest in our cloud OS but does not have any professional knowledge on OpenStack can play it without external service from others. Specifically, we have a unified storage solution which manages local storage and distributed block storage in one single storage system, we have simplified OpenStack Networking to make networking easy to use and understand, we have world-class UI designers and front-end engineers to design and develop a more intuitive and elegant management console etc.
Q: What is different about a distribution targeted at an Asian market compared to those distributions that are already available?A: For example, our management console not only implements lots of powerful functionalities and features but also, I believe, is the most human and artistic work. Besides the UI works, we invest much resources to design the core backend services, such as the unified storage system, which combines the advantage of high-performance local disks and flexibility of distributed block storage.
Q: Do you already sell your distribution to customers?A: It is still under heavy development, but we will release our first distro this summer. Our goal is to make UnitedStack OS the best cloud OS (including the OpenStack distributions) in the world. So that’s very ambitious, eh? (laughs)
Q: What use cases are you targeting with your distribution?A: Good quesiton. Another big difference from other OpenStack distributions is that we are build a general purpose cloud OS including a sophisticated cloud app store, not a domain specific cloud. The customization and business related requirements are fulfilled through the app store. Enterprise app developers, ISVs can easily put their software on our app store, and distribute to their customers directly.
That’s to say we aim to build a fundamental platform for any types of companies that want to build private or public clouds or whatever type of technologies.
Q: What is currently the state of affairs with OpenStack in China?A: China has the second biggest OpenStack user base in the world after the U.S. Almost all major players in the Chinese IT industry are interested in OpenStack and most of them are at least doing some research. Some of them have already deployed OpenStack in their production environment, amongst them are the biggest Chinese internet companies. As for developers, I know of some of our top 10 universities that are engaging with OpenStack, their students are doing research and writing papers about OpenStack. The Chinese OpenStack user group has more than 3,000 members.
But we also have some problems in the Chinese OpenStack community. One is that most Chinese companies are just using OpenStack instead of contributing to upstream. This is a big problem.
Q: What is the reason?A: OpenStack community is basically centralized in U.S., whereas the Chinese community is decentralized. Also, the language barrier is a significant one for Chinese developers. And few companies are truly understanding open source spirit and culture, so they do not have much desire to deeply involve in the open source community, never say to contribute back. I have an open source community background, and as the founder of a tech company, I want UnitedStack to be a open source company, and I want it to be as huge and successful as Red Hat.
Q: Do you see the spirit and culture change somehow?A: Yes, as the international communication becomes more and more frequent among Chinese developers, there are more engineers are realizing that involving into open source projects is the best way to develop themselves and propel their career. We also have done a lot of promotion activities such as meet ups around OpenStack and open source projects, I think it's changing now and things are getting better.
Q: How can the global OpenStack community support peers in China?A: As far as I observed, the global OpenStack players such as Rackspae, Red Hat, Dell and etc. have little presence in Chinese market. Maybe they do not know how to enter such a special market with an ambiguous government policy and market, and they do not have much experience in dealing with local companies, but they should prepare and try it ASAP … build local teams, find some local partners, give much support to local OpenStack communities.
Q: Thank you very much.A: Thank you.
Dell OpenManage System Administrator (OMSA) 7.3 for Ubuntu and Debian is now published. When we recently published OMSA 7.2, we switched to a new apt repository format to better work with both Ubuntu and Debian and to allow packages for multiple OS releases in the same repository. OMSA 7.3 continues that. Furthermore, OMSA 7.3 is now built on both Ubuntu 12.04 and Debian Wheezy to increase compatibility with Debian. All packages that are in the Ubuntu distribution but are not in Debian are rebuilt for Debian and included in OMSA Wheezy repository for convenience.
(Please note that OMSA 7.2 and above are not built for Ubuntu 10.04 and Debian Squeeze. The last OMSA release tested with Ubuntu 10.04 and Debian Squeeze is OMSA 7.1, which is also provided in the new apt repository for convenience.)
Additionally, OMSA's Integrated Tunnel Provider (srvadmin-itunnel) is now built for Ubuntu and Debian. This brings Ubuntu and Debian closer to parity with RHEL and SLES in terms of the System Administrator functionality in OMSA.
Also included in this release is version 4.3 of the Dell Deployment Toolkit (DTK) 4.3 for Ubuntu and Debian. DTK is lighter-weight than OMSA and is meant to assist with system deployment. The packages are named syscfg, raidcfg, and dtk-scripts. syscfg is a tool to configure server BIOS, BMC/iDRAC settings, DTK state settings, and to do PCI device detection. raidcfg, as the name suggests, is a tool to configure RAID on Dell PowerEdge servers. dtk-scripts contains sample DTK scripts and tools to build a bootable Dell utility partition for DOS-based firmware updates.
More information on where to download these packages is at http://linux.dell.com/repo/community/ubuntu/.
Please join us on the email@example.com mailing list for support and feedback. You can sign up at <https://lists.us.dell.com/mailman/listinfo/linux-poweredge>.
Dell Server Deployment pack (DSDP) configures the server's Dell Remote Access Controller (DRAC), integrated Dell Remote Access Controller (iDRAC), RAID, and
BIOS using INI files and Command Line Interface (CLI) options. You can also configure RAID using the Array Builder Wizard. The Array Builder Wizard also creates
a Dell specific boot image, which you can use for deploying the operating systems.
Dell Server Deployment Pack v2.1 includes the following high level features:
What’s new in version 2.1
Key features and functionality
Dell Server Deployment Pack v2.1 download is available here
Dell TechCenter Page
This blog is about the silent installation of Red Hat Enterprise Virtualization - Hypervisor 6.4 (RHEV Hypervisor) using DTK. It covers silent installation in both BIOS and UEFI modes.
1. Modify partcfg.sh (in /opt/dell/toolkit/template/scripts) a. Utility partition is not required, so set it to OFF i.e., in BIOS section, set DT_PARTN_UP=OFF b. set DT_DP_SIZE=3002. Execute modified partcfg.sh which will create a temporary MBR partition. This partition can be used to store RHEV-H contents required during installation3. Mount the RHEV-H NFS share to /tmp/os_src4. Export the following variables which are used by ‘lininst.sh’ a. DT_HD : The variable which denotes the drive where OS will be installed Example: #export DT_HD = /dev/sdX b. DT_OS_SRC : The variable which denotes the mount point of RHEV-H source Example: #export DT_OS_SRC=/tmp/os_src5. Modify lininst.sh (inside /opt/dell/toolkit/template/scripts)6. Modify the bios section as follows: a. Comment kernel, initrd and ks.cfg check block b. Comment kernel, initrd and ks.cfg copy block c. Modify the grub.conf block as follows
default=0timeout=10title RHEV Hypervisor install kernel /vmlinuz0 root=live:/XYZ.iso rootfstype=auto :linux storage_init=/dev/sdX storage_vol=::::: local_boot BOOTIF=eth0 adminpw=<hash value of password> initrd /initrd0.img
7. Mount /dev/sdX1 (partition created after partcfg.sh) and copy the vmlinuz0 and initrd0.img from NFS share to this mount. mount command example: #mount /dev/sdX1 /mnt8. Execute modified lininst.sh, which will install grub in /dev/sdX1 and reboots the system.9. After reboot, grub screen prompt is displayed and it chooses the ‘RHEV Hypervisor install’ option10. RHEV-H installation will proceed on /dev/sdX without any manual intervention11. After installation, login as admin and change the admin password
1. Set DT_DP_SIZE_GPT=3002. Execute modified partcfg.sh which will create a temporary GPT partition to store rhev-h image contents present in NFS share.3. Mount the RHEV-H NFS share to /tmp/os_src. 4. Export the following variables used by ‘lininst.sh’ a. DT_HD : The variable which denotes the drive where OS will be installed Example: export DT_HD = /dev/sdX b. DT_OS_SRC : The variable which denotes the mount point of RHEV-H source Example: export DT_OS_SRC=/tmp/os_src5. Mount /dev/sdX1 and copy the vmlinuz0, initrd0.img and EFI directory present in NFS share to this mount point.6. Create a file named BOOTX64.conf with the following text and copy to /dev/sdX1 in the EFI/BOOT/ directory.
default=0timeout=10title RHEV Hypervisor install kernel /vmlinuz0 root=live:/XYZ.iso rootfstype=auto :linux storage_init=/dev/sdX storage_vol=::::: local_boot BOOTIF=eth0 adminpw=<hash value of password> initrd /initrd0.img
7. Create an entry in UEFI firmware using the following command: # efibootmgr --create --gpt --disk /dev/sdX --part 1 --write-signature --label "RHEV Hypervisor" --loader "\\EFI\\BOOT\\BOOTX64.efi"
8. After reboot, grub screen prompt is displayed and it chooses the ‘RHEV Hypervisor install’ option9. RHEV-H installation will proceed on /dev/sdX without any manual intervention (the temporary partition will be removed during installation process)10. After installation, login as admin and change the admin password
NFS share creation on a host Linux machine (as a root):
1. Create a folder say /rhev_nfs2. Open /etc/exports/3. Add the below line in /etc/exports/ and save it /rhev_nfs *(rw,async,no_root_squash)4. Restart the nfs share by using below command #service nfs restart
This blog post is written by Kiran Poluri & Kiran D from Dell Hypervisor engineering team.
PowerEdge VRTX is a shared infrastructure solution designed for small or branch offices as it simplifies IT by combining all of this infrastructure in a single unit, eliminating the need for a lot of cabling and complexity, and allowing for management of these resources from a GUI single interface. The detailed specification of VRTX is available at Dell website.
VRTX integrates servers, storage, networking and management into a single office-optimized chassis in the size of a tower server (or 5U rack).
VMware ESXi 5.1 Update 1 is the supported VMware ESXi version for the sever nodes on PowerEdge VRTX. The RAID controller used in VRTX is "Shared PERC8" for the shared storage management. The driver for Shared PERC8 (the low-latency internal shared storage) is included from the Dell customized VMware ESXi 5.1 Update 1 Image (A01 Revision) which can be downloaded from support.dell.com.
Note: Earlier revisions (A00) of the Dell Customized VMware ESXi 5.1 Update1 Image does not support Shared PERC8 feature.
The server nodes Dell PowerEdge M520 & M620 are supported on Dell PowerEdge VRTX with BIOS Version 1.7.6 and above.
Dell starts shipping VMware ESXi 5.1 Update 1 (Dell Customized A01 revision) image from factory on all supported server nodes from 6/25/2013.
Refer to blog to know more about Dell Customization of VMware ESXi images.