One Firm's Wild Ride to the Cloud

My IBM Connect presentation is now posted on Slideshare.

Thank you everyone for coming and your questions.

One Firm's Wild Ride to The Cloud from Keith Brooks

Mission Tools: Lotus Quickr to IBM Connections Cloud

Welcome to the 3rd installment of my migration project.

I am happy to say that we moved everything over, now we are verifying everything came over and cleaning up all the orphaned and misc. files that were found and cross linked at one time.

So, how did we get it all moved I bet you are asking.

There is an automated way, via IBM and their partner in Europe T-systems, which is part of T-Mobile it seems, they can help and their pricing is really for the bigger companies. They have a process and methodology and will then get working on it....but you still have limits of bandwidth and time although they claimed they could get this done in a week.

Our problem was, we could not have any down time, but also not have any duplication of data which they made it sound like they offload the data first then do a second synch afterwards.

If you want to go with them, contact me and I will connect you.

After weighing the pros and cons we decided to do it in house.

These are the tools I used, you may have other preferences.

IBM: IBM Connections Cloud Plugins for Windows and the plugins for Notes clients. You will also need the Quickr Places Monitor.

RPR Wyatt: Quickr Essential File Extractor and Essential Place Catalog. There are some fees with these and you need to ask for it as it no longer sits on the website. I previously worked with RPR Wyatt and was loaned the tools. I owe you drinks Jim! 

Microsoft: Windows Explorer

Lialis: MAC people this is for you, File Jockey, to have a plugin that doesn't cost you anything. As they state on the site:

"We also supply File Jockey for all IBM Connections Cloud users. It's 100% free, no limitations and no signup. Feel free to use it now by pressing the button above to read the how to guide."

And there you go. Nice and simple right? Off you go.

Not so fast. You will find a few things along the way that you also need this cute tool I found called  Bulk Rename Utility. It does exactly what it says and makes changing, say 800 folders that had an added _Folders added to them, magically disappear in seconds. No command line knowledge required. Ask me if you need help.

Why do we need this tool? Because the RPR Wyatt tool extracts all the attachments and maintains the folders layout but adds _Folder to everything, My main gripe, but a small one. The developer did not return my messages, hey it is free without support, I just wanted to make it a better tool. But since these posts are capping out at 200 views I figure no more than 200 Quickr shops exist and maybe it isn't a big enough group to warrant the effort. So if you still have Quickr and are thinking of migrating let me or RPR Wyatt know. There is safety in numbers.

The steps of my process was, after backing it all up someplace first:

1. Use the RPR Wyatt Place Catalog tool first and review what we had and how much of it existed.
2. Use the RPR Wyatt Extractor tool to extract the files from one place at a time. Why not all at once? 100's of GB and you know there will be issues, you want to progress, even if it is slowly.
3. Once extracted and you installed all your plugins, login to IBM Connections Cloud and manually start making all of your Communities. If you read my post on automating this process you will think, cool I can load them all at once. Sorry to burst your bubble, but no. IBM in a PMR I raised outlined that you can only create 1 community at a time and only with one owner. Thus, the actual time it took me to create a community and configure it the way we needed and have all the proper owners and members was approximately 2-3 minutes each. Not so bad, but I really wish IBM had fixed their API.
4. Next step is to add the Communities to the Windows Explorer plugin.
5. Now copy and paste from one Windows Explorer window from the drive on teh server where you extracted the files to the other which has the Community list.
6. Once done, delete all non admin members from the Place within a browser so users do not accidentally add or delete files that are needed.
7. Continue until done.

The Lialis File Jockey was for the MAC people to start using Connections files the way they needed to for every day work. So technically I did not use it to do the migration, but I give them kudos for being there and were it NOT for them, the project might not have started as the office is moving over to all MACs.

Sounds easy, but what we found was rather dismaying.

1. Cross linked files were everywhere. This was from various crashes over the years and while we knew we had some, I was not expecting the 100's of folders linked.
2. Orphan files. The extractor tool was great about files and folders but sometimes random files just in a place did not get migrated. Always verify what you see is what you got.
3. Hidden folders and other anomalies. Probably some bad drag and drop by someone left some folders in the oddest places.
4. Size matters. What the Quickr server and the RPR WYatt tool showed us was not what we found when we extracted the data, In fact, we found we had about 60% of the data once extracted. As an example if we were told there were 10GB in a place we usually had about 6GB after extraction. SWEET! But left me wondering what was wrong. See cross linked files and you start to understand.

Thus while the project was completed by the time everyone started work today, I still have many folders and files to clean up, VERY manually eye balling everything. And I am sure I will get some panic call about a missing file or two, but at least to the employees, everything is moved.

There will be one more post in this series about what I learned about IBM Connections Cloud along the way, stay tuned.

So now that I am done with this project, anyone else need me to help migrate them to IBM Connections Cloud?

Mission Worries: Lotus Quickr to IBM Connections Cloud

Who knew so many people still had Lotus Quickr? See IBM I told you people love it still.

There are many things to think about during a migration of data, most of which you as the client will not think about, or even know about, until you have moved and find it out. Do you know why? Because not very many companies, people, or Business Partners have done these migrations like I have over 20+ years. 

These "Mission" posts are meant to help you in this process.

Last post I gave some basic information, now we can dig into some details.

The client uses a plain and out of the box Lotus Quickr so we did not have to worry about meta data or odd forms or whatever else dreamed up by some developers or over eager Admins. I know because I am their Admin.

We did worry about some things and here are some of those items:

• My name for the most part is tagged to all the files and folders NOW which is how IBM Connections and IBM Connections Cloud works since I am the one uploading the files.That is the down side, however, once you get into the folders and find the files, you can see all the document details and it shows all the past information. Not a bad thing, but not perfect. Client accepted this as a good monetary trade off.

• MACs lack of a Notes sidebar plugin. No comment. I am not a MAC user so I am not justified in arguing this point but the Windows Notes plugin is helpful....BUT not for the migration itself. Afterwards it acts like the Quickr Places sidebar app and drag and drop and such is fine plus you get Files, Activities and Status Updates in your sidebar to better leverage the rest of Connections.

• Training users. Actual training is minimal, updating the Notes clients on machines and then the plugin installs and configurations takes about an hour a person on average. If you leverage SmartUpgrade or my friends at Panagenda's Marvel Client you may just need the plugins updated and that is quick.

• Where is the ROI? Mobility! iPad access in front of a client meeting or potential client meeting is crucial and even better, IBM handles the backup of our data. 

• Which brings me to backups. IBM doesn't really backup the data, this is an issue to be addressed by me going forward, but IBM does synch data between their data centers and in IBM we trust, so again a trade off. Not to worry the Quickr server running on VM will just be sitting here if we need data prior to January 1, 2017. Just like the Domino.Doc VM server sits waiting for something we lost from the previous migration. For those keeping score, I have NEVER run that VM to find any files that went "missing". 

• Multiple Folder levels. It is this one item which has prevented the client from moving until now. Promised by IBM over a year ago, and it was delivered, just not the way the client wanted it to work(it also only now lets you edit and create folders from the plugins and mobile which also did not come out last year). We had patience in IBM and since I am an IBM Champion and part of various beta teams for IBM and advisory councils, I pushed for it all to get done, as did quite a few fellow Champions. So if this is what you have been waiting for, you have no reason not to move forward now. January is free for me, let me know when you need your migration done by and I am here for you.

• Timing. Start testing as early as possible as you need a lot of testing scenarios. Keep in mind that uploading speeds are usually MUCH slower than downloading speeds. So that Terabyte of data will not get uploaded overnight, I don't care who you use or what they promise you.

• What if they hate using the Cloud? Said, no. one. ever...who knows what they are doing.

Are there more issues? Sure, every environment is different and I am happy to discuss them if you comment or tweet, call, skype or email me.

I forgot to mention this was a Windows server customer, and I could complete the migration no matter what operating system platform you are running Lotus Quickr.

Mission Accepted: Lotus Quickr to IBM Connections Cloud

My last Lotus Quickr customer is moving off of Quickr by December 31st.

I love Quickr, still do, but time marches on and maybe you found yourself looking at Quickr wondering:

• How do I secure this box? You have options, and hopefully have used them, but the server is trapped on an 8.5 server which is basically an unsafe server to run at t his point especially if it has access to/from the outside world.
• Accessing your data is getting harder as browsers advance but the code on the server does not recognize all the updates of course.
• Mobile needs, never really got a working Mobile client for Quickr
• Java, yeah well it was good while it lasted

The thing is, if you are solely using it internally it can just sit there for years on end, some of you out there know exactly what i am talking about too.

IBM does not provide a tool to migrate data from Quickr to Connections. IBM Global Services relies on a tool from a Business Partner and you can go that route. It is priced accordingly.

However, in this case my client is not that large and only has about 300GB of data in Quickr so we decided to move it using the tools available to us.

I did many tests before it was agreed to move to IBM Connections Cloud, the client already owns and uses Mail and Sametime there so the move to it would not involve new licensing. 

In fact, we saved money by not needing to renew our Lotus Quickr, now called IBM Connections licenses and support.

Once it was shown we had viable functionality, which basically meant many tiers of FOLDERS that could be created and edited via Notes, Plugins to Windows and MACs, we were ready to go. 

Sorry taxonomy and search people, but when you need many files quickly, search is not your friend especially if you have no idea what the file is called. This is a religious argument to have at the next LUG or IBM Connect event.

So if I didn't go with IBM, how did I do it? Same way I have done migrations for over 20 years, ask around and build and leverage what you have to work with and it should save the client about 40% of the Global services price. 

Your mileage may vary, but if you are in need of this, my January is free now so call, email, tweet, skype me to help move you to IBM Connections Cloud from Quickr.

Next post will be about the process involved, for those interested let me know so I can provide more details to help you.

How to Create Multiple Communities in IBM Connections or Connections Cloud?

As a non developer, this was easier than I imagined it would be, once help arrived from my friends.

I have a Lotus IBM Quickr customer, my last one, so any Quickr sites, contact me I am available! Anyway, I am moving them from Quickr to the IBM Connections Cloud environment.

This post is not about the move itself, but about how one can create over 200 new communities(200+ Places and 220GB of data) without having to hire an intern to do it manually.

I posted to the chat for IBM Connections people and my friends Christoph and Ben pointed me to the Connections API documentation. Yes, it looks and reads like gibberish so I asked if there was an easier way and one of them found a blog post by Julius.

Julius, as it turned out, was also in the chat and I asked if this post would work for Connections Cloud, no one knew. So we tried. And failed. Christoph said we needed a header, which as it turns out was the correct answer, an atom+xml header.

Confused yet? Well you could review Julius's blog post where he covers almost everything, but here is what you need to know.

Using Firefox and the RESTClient plugin you get a nice little UI that lets you create your request in pretty simple XML terms. Thanks Julius for coding something simple enough for me to figure out how to swap in the information to make it work.

These are the steps:
1) Set the header. I did not even see the little menu on top of the screen with a drop down to create headers.
Click on Headers to see the choices and select Custom Header.










When the next window comes up, enter the text as in my screen shot. To remember this for future need, click on  the Save to favorite box then say ok.








2) For the Method drop down box, select POST
3) For the URL, you need to use: https://apps.na.collabserv.com/communities/service/atom/communities/my
4) The body field, this is where the fun starts.

This is what my code looked like in basic form.
<?xml version="1.0" encoding="UTF-8"?>

<entry
xmlns="http://www.w3.org/2005/Atom"
xmlns:app="http://www.w3.org/2007/app"
xmlns:snx="http://www.ibm.com/xmlns/prod/sn">
<id>ignored</id>

<title type="text">Jefferson Airplane</title>

<summary type="text">ignored</summary>
<content type="html">Music group in san francisco</content>
<published>ignored</published>

<author>
<name>Keith Brooks</name>
<email>keith@keithbrooks.com</email>
<snx:userid>18675309</snx:userid>
<snx:userState>active</snx:userState>
</author>

<updated>ignored</updated>

<category term="community" scheme="http://www.ibm.com/xmlns/prod/sn/type"></category>
<snx:membercount>ignored</snx:membercount>
<snx:communityType>private</snx:communityType>
</entry>

 5) Once run you should get a 201 Created reply and when you go to the communities you will see it.


Next step is to include members and format the communities. Will post an edit or a new post with that code for those in need.

Thank you again Julius, Christoph and Ben!