Hi there,

Let’s face it! From time to time, you woke up, grabbed your iPhone 4S and asked Siri: “What does my day look like?”. Then, the answer may have been: “Damned! There is a scalarizr update planned today. Don’t forget your stress pills”. For all the awesomeness that scalarizr brings to infrastructures, managing the agent has never been a walk in the park. Like every Rockstar, it had its whims and offered you tight choices. A Shakespearean dilemma: to update or not to update.

For instance, last Thursday, we released scalarizr version 0.7.226. We ran tests for various scenarios and farms types. Regrettably, scalarizr post-scripts deleted important runtime informations -sometimes leading to scalarizr’s crash. This kind of situation is exactly WHY the engineering team worked that hard to bring you today’s process update. Well, Scalr is thrilled to announce an era of seamless updates.

Like a Muffin, the revolution is built on three ingredients: the scone, the chocolate and the cream.

The scone

At first, a much more explicative interface has been shipped: on the server’s extended information window (Servers > View All > Options) you can display the current scalarizr status, updated details and informations on the new versions, check the last update errors, and update schedule info. Even better, you may also restart the scalarizr or upgrade it from this panel. This helps you to keep a record of the previous scalarizr updates and to know precisely what is under the hood.

The chocolate

The chocolate makes the muffin’s taste: what a particularly true statement here!

You can now cook scalarizr’s updates up in whatever sauce you like. We implemented a double repo feature based on a simple principle: choice is better. If you are an “early-adopter” you should have the right to try the newest features. If you don’t like to dive in first, something proven must be at your disposal.

So, two drivers are actionnable:

• You can select a “stable” (or “prudent administrator”) or latest (or “go-getter”) repository
• The update’s frequency (for instance: not Saint Valentine’s day)

The repositories are completely isolated. Stable repo packages have been rock-solidly tested in real-life condition for at least one month. Unlike the newly introduced stocks of a famous social network, these won’t be downgraded after two days. ONLY the packages passing these hard tests will be marked as stable. You can sleep in peace.

The repository named “latest” straightforwardly contains the latest scalarizr version. It has been of course tested but may have been occasionally marked as non-stable. Theoretically, an issue is still possible, while actually rare. Therefore we recommend to use this repository for your stagging/development farms. If you do so, that’s ideal to test Scalr’s latest improvements at no risk.

The cream on the top

By default the auto-update-client will use the stable repository and check for new updates every hour. The frequency can be customized along with the repository settings. It has a classic cron job format without minute and month to define. Time intervals can take a value anywhere between one hour and one month.

Here are a few examples to illustrate those settings. For example if you let all the columns empty, that will keep the default choice: the cron job will be activated every hour. If you put 11 (or 23) under Hour the job will run everyday at 11AM (or 11PM). Eventually, you can choose 6,7 as the Day of the Week column to install potential updates only during the WE.

We hope that you will enjoy this major step towards a more personal Scalr (dare I say: “fun”?). As usual, we’re waiting for your feedback.

Enjoy!

The Scalr “Make a wish” team

Hey Cloudsters,

Paul, our amazing Sales Engineer, has worked very hard these last days to write tutorials on the new documentation.

In case you missed them, here are some of them:

• How to install Magento with Scalr
• How to setup a Master-Master MySQL Replication in different regions
• How to install a WordPress Blog with Scalr
• How to install Apache Tomcat with Scalr and Chef

We are currently working on others:

• How to setup Varnish
• How to migrate to the Cloud

Which tutorials would you like to see on the documentation? Is there something in particular you struggle to install when you used Scalr for the first time? Your feedback would be really appreciated! If you want to contribute, you have our word: the most deserving tutorials editors will get a tee-shirt (or a hoodie).

The Paul Chapotet Fan Club

Great news for all Redis and PostgreSQL users: you can now choose the storage engine of your database between EBS volumes and Ephemeral devices. For those who do not remember, only EBS volumes were available before. But you may wonder about the differences between these two types of storage:

• EBS volumes provides block level storage volumes for use with Amazon EC2 instances. The data persists independently from the life of an instance. So when you terminate your database, the EBS volume attached to it will still be available and you could attach it to another instance. As a reminder, you can manage all your EBS volumes in Scalr going to Tools > Amazon Web Services > Amazon EC2 > Elastic Block Storage > Manage volumes
• Ephemeral devices are ideal for temporary storage of information that changes frequently, such as buffers, caches, scratch data, and other temporary content, or for data that is replicated across a fleet of instances. The data persists only during the life of the associated Amazon EC2 instance.

You can learn more about storage on Amazon EC2 on our fresh new wiki.

To select your storage engine, click on your Redis or PostgreSQL role, and then on Database settings. Please note that you will not be able to change the storage settings once the farm has been launched (you’ll have to remove the role and then add it again). Ephemeral devices are available for Redis and Postgres databases only, but MySQL support should come soon.

You might remember a previous post we did on our quest to go realtime. Today, we take this another step further by giving you realtime updates on the progress towards taking your backups. As you know, taking manual backups of your data with Scalr is easy. Simply open the status window of your database (Options > [database] Status) and select the backup method that you want to apply. Please keep in mind that backups and data bundles are different:

• Data bundles are snapshots of the EBS volume containing the data (binary storage).
• Backups are snapshot of the EBS volume containing the data by running a database dump (object storage).

You can now see, as it happens, all the Scalr magic. Take a look below. This new feature is currently for MySQL, Postgres and Redis and will be extended soon to Postgres and Redis. .

Oh my, this one sure was overdue.

For the past two weeks we’ve worked on revamping the documentation site, this time on Confluence instead of DekiWiki, and I think we have a good foundation for future updates.

You can head on over to the new documentation site to take a look. We want this platform to become the ultimate knowledge resource on Scalr. Just let us know if you would like other tutorials to be added – this can be done. We’re waiting for your feedback.

Enjoy,
The Scalr “RTFM” team