The Jungle Disk Blog
I’m back. Some of you may already know me as JungleMarty. I drifted away from the Jungle Disk product to work on other Rackspace projects, but now I’m back with Jungle Disk full swing. The decision was entirely mine.
For those of you who don’t know me, I was born and raised in upstate NY. I earned a degree in electronics and worked in the audio electronics industry for a few years. Around that time, computers started to evolve and they really caught my attention. I decided to focus on software full time and earned a computer science degree from RIT. My final computer graphics assignment is responsible for a few jokes in the office: http://www.youtube.com/watch?v=Z0tMP921ub4
After writing code for many companies, including Xerox and SAIC, I found myself working at Jungle Disk as their first full time .Net developer. I started with the company shortly after it was purchased by Rackspace. Some of the projects that I worked on include the integration with Chase Payments, the web access website, the public file sharing feature and most recently, the move to our PCI compliant environment.
A discussion about me would not be complete without mentioning guitars.
I’ve been playing guitar since the 6th grade and I’ve taken it to an entirely different level; collecting, playing and meeting with some of the best guitar players in the world (Guthrie Govan, Paul Gilbert and Tony McAlpine). My favorite amps are from the UK and my favorite guitars are custom made in California.
I’m in the process of moving to Austin, TX so that I can fully commit to Jungle Disk and Rackspace. I’m currently making fundamental changes with the methods used to develop and test code to our backend systems. Once these processes and systems are in place, we will be able to improve the Jungle Disk product and make it even better.
Developer, Jungle Disk
You’ve asked, we’ve listened, and now we’re ready to start talking. Jungle Disk phone support beta is live starting now.
To initiate a call during the first phase of our beta we’re asking you to begin by opening a chat through our Jungle Disk Support Site using the “Chat with us!” pop-up (pictured to the left). The team is excited and looking forward to speaking with each of you.
During beta the team may need to say they are unavailable for a phone call. We do not know how many of you will want to talk to us and how those interactions will impact our staffing levels. The team is also unsure about how long the beta will need to last or if it will need to be shut down while we add staff to handle the increased interactions with you.
The beta will allow us to learn based on real world, real customer feedback. Your feedback to the beta is critical to our learning and what we will offer to serve your needs. You can provide feedback through a comment on this blog, an email to email@example.com, or a chat session with our team.
We’re not posting a 1-800 number on the site today or giving one out for inbound calls at all times. Our goal in the future is to understand if we should add an inbound phone number so you can call us without having to pop into chat first. Another idea we’ve discussed is scheduled call backs — if you like this or other ideas share through the feedback channels.
“Hello, this is Jungle Disk” — starting now!
The Jungle Disk Team
Today’s post is written by Jonathan Robertson, one of Jungle Disk’s customer support technicians. Jonathan has been a Racker for a little over 2 years. He began as a support tech for Email before moving over to the team in October 2012.
There are two versions of Automatic Backups available to Desktop and Workgroup Editions of Jungle Disk. Even though Backup Vault has some clear advantages over Legacy Backup, there are still some situations in which Legacy would be the better choice.
Note: Since this blog post is all about how things work, I’ll be adding some extra details that will apply to all versions of Jungle Disk – but keep in mind that this post is primarily directed at Desktop and Workgroup Editions.
Backup Vault (All versions of Jungle Disk)
When a Backup Vault Job is run, a lot happens in the background.
- Backup Search
- 1. All files in your backup job that are on your computer are quickly analyzed to check their “Modified Date” attributes. This information is used to identify which files appear to have been updated since the last time the Backup Job ran.
- 2. Each of these files that appear to have been modified are then compressed and split into 1MB pieces that we call “Chunks”.
- 3. The data inside each Chunk is then analyzed with a mathematical formula to produce a series of numbers (this is an MD5 hash… but one way to think of it is a fingerprint or an ID number).
- 4. With this hash (ID), we’re able to identify which parts of the file have changed. Each piece of the file that has a different hash than the ones we have on our end are queued to be uploaded (since these are identified as ‘new parts’ of the file).
- Backup Upload
- • At some point in the Backup Search, enough data is processed for the Backup Upload to begin. This process involves streaming your chunks from your computer to your Online Disk for safekeeping.
- Backup Database Upload
- • During the Backup Search phase, changes that were being identified in files were being recorded to your Backup Database.
- • This is a special file that includes all of the Folder and File Names, as well as their locations in your folder structure.
- • This also includes the ‘Assembly Instructions’, so to speak, for how each of your Chunks relate to each other in each file.
- • Without this Backup Database, your Chunks cannot be put back together again.
- • Because this database is so important, it is also uploaded during your Backup Job once your Backup Search completes. We do this so that you can download it from your Online Disk in case your hard drive becomes inaccessible and you need to restore your data to a different computer.
- • During the Backup Search phase, changes that were being identified in files were being recorded to your Backup Database.
Where Backup Vault Would Work Well
- • If you have an individual file that’s larger than 5GB, a Backup Vault is the only way you could upload this to our system.
- • Only changed parts of files are uploaded (in increments of 1MB each), which means that even if you do wind up with a file that’s multiple gigabytes in size (like a large SQL export), Jungle Disk can still backup the changed portions of the file without having to re-upload the entire file.
- • Since backup data is compressed, it saves you time when needing to download/restore.
- • Another benefit of ‘Chunking’ and compression is that you usually wind up paying much less due to each of these features resulting in a smaller amount of data needing to be stored on your Online Disk.
Where Backup Vault is Not the Best Choice
- • If you need to quickly reference or browse certain backup files in Web Access or in the Network Drive (particularly for situations where you would want to listen to a song or view a picture), this would not be possible without initiating a Restore job.
- • In this case, it would be better to consider Legacy Backup since the viewing of files/data is still possible without having to Restore due to it not being compressed or broken up into Chunks.
- • Since your file assembly instructions and file list are all contained within the Backup Database, this database must be up to date in order to initiate a restore. If it’s not (which is mostly descriptive of situations where you’re needing to restore a group of files to a different computer, i.e. your computer crashed), Jungle Disk will need to download the Backup Database from our end (from your most recent successful backup) and verify its contents before a restore is possible.
- • Restoring data from Legacy Backup requires no such download, which can help to make the restore process more smooth if you’re initiating the restore from a new computer
- • A Backup Vault can only be connected to a single computer at a time.
- • If you’re wanting to backup multiple sets of files to a common directory on your Network Drive on a regular basis, Legacy Backup would be more suited to this kind of configuration.
Legacy Backup (Desktop and Workgroup Editions only)
When a Legacy Backup Job is run, the files identified in the backup are analyzed and their Modified Dates are compared to the Modified Dates of the previous backup. Files that have a more recent Modified Date are treated as “updated”, so they are queued to be uploaded. When one of these files is uploaded, a full copy of the old file is treated as a previous version, which means it then becomes subject to your Retention Policy.
Where Legacy Backup Would Work Well
- • Even in your Initial Backup, you can stop the Backup Job halfway through and then continue it later. All files uploaded at that point will still be there when you come back and start another backup when it’s more convenient for you.
- • The data is still viewable in Web Access and your Network Drive (via [Online Disk]/backups/[folders] by default) for quick reference.
- • If you need to store media data, like photos, music, or older documents that no longer need to be updated (but you might like to view from time to time), Legacy Backup is usually the perfect solution.
Where Legacy Backup is Not the Best Choice
- • If you need to keep lots of previous versions, your retained data would grow much more quickly with this method than it would in a Backup Vault.
- • So if you need to hold onto more than 2 previous versions of each backup file, we would recommend considering a Backup Vault instead.
- • Since Legacy Backup stores its data directly on the Network Drive, it’s also subject to the same limitations. The most commonly encountered limitation in this case would be that you cannot upload individual files that are more than 5GB each.
- • If this is something you need to do, a Backup Vault would be the only way to upload a file that exceeds 5GB in Jungle Disk.
- • Even though the Legacy Backup data is visible in Web Access and your Network Drive, you should not make any modifications here. This Legacy Backup data is meant for Reading/Viewing only. Modifying this data is not recommended since it could produce unwanted results with what is or isn’t backed up.
- • If you need to view and modify data regularly, consider uploading the data directly to the Network Drive instead of using an Automatic Backup job.
I hope this helps to clarify the differences between our two backup methods for you and where each method really shines.
As with any topic, please feel free to submit a support ticket to our team at https://support.jungledisk.com if you have any questions or need further clarification.
Jonathan Robertson (“JungleJonathan”)
Jungle Disk Support
UPDATE (2013-05-19 10:25am CDT): Scheduled software release was completed successfully with no observed customer impact.
We are pleased to announce that our infrastructure upgrade that took place on Sunday the 12th of May was a success with negligible customer disruption. We are already seeing significant improvements to performance and the customer experience.
As part of our continuing improvements we will be releasing a software update to our server-side infrastructure on Sunday, 19th May at 9.00am CDT.
We have scheduled this release for a 2-hour window. Customers may experience short periods of instability where the Jungle Disk client will be unable to connect to the infrastructure, however we expect the actual time that customers are impacted to be minimal. If severe customer disruption exceeds 10 minutes we’ll initiate rollback procedures. As part of our plan, we will do everything possible to minimize disruption to your services.
We will provide status updates throughout the release on our Support Site and via Twitter. If you have any support questions please email firstname.lastname@example.org or go to our Support Site to Chat or submit a ticket. Please direct any feedback to email@example.com.
We apologize in advance for any inconvenience this release may cause and appreciate your patience as we work to improve Jungle Disk’s services.
The Jungle Disk Team
P.S. In addition, the Jungle Disk team is working on updates to the client software that we are looking to release in the coming months. Stay tuned.
P.P.S Don’t forget, we’re available via Live Chat. Login to the Support Site to start chatting now!
Update (10:00am CDT 2013-05-12): No additional issues have arisen so maintenance is complete. Thank you for your patience. Please let us know if you experience any subsequent issues via Live Chat (http://support.jungledisk.com/) or ticket (firstname.lastname@example.org)
Update (9:00am CDT 2013-05-12): All services performing as expected. Hard downtime was negligible during the DNS cutover to our new infrastructure. No customer issues experienced at this point.
Update (8:20am CDT 2013-05-12): New signups (via www.jungledisk.com) are now available.
Update (7:20am CDT 2013-05-12): Control panel access (via www.jungledisk.com) is now available.
Update (7:00am CDT 2013-05-12): Maintenance has begun. Control panel access and signups (via www.jungledisk.com) are temporarily unavailable.
On Sunday the 12th of May at 7:00 am CDT, Jungle Disk will be performing maintenance on our infrastructure in order to upgrade the security, performance and resiliency of our environment.
We have scheduled this maintenance for a 4-hour window. We expect the actual time that customers are impacted to be significantly less than this. During the maintenance window the Jungle Disk website will be unavailable for control panel access and new signups or subscriptions. Customers should also expect short periods of instability where the Jungle Disk client will be unable to connect to the infrastructure. If severe customer disruption exceeds 20 minutes we’ll initiate rollback procedures. As part of our plan, we will do everything possible to minimize disruption to your services.
We apologize in advance for any inconvenience this may cause and appreciate your patience as we work to improve Jungle Disk’s services.
The Jungle Disk team
Don’t forget, we’re available via Live Chat. Login into our Support Site to start chatting now!
Our first week of Live Chat with our invited list of beta testers has gone well and we’re excited to roll out the beta program for Live Chat Support to all customers! We reserve the right to pause the beta program at any time if we are unable to fanatically support the live chat volume. Please bear with us as we get to grips with the anticipated load.
You can access Live Chat by logging into your support account at http://support.jungledisk.com.
Currently, Live Chat is available Monday-Friday 8am-4pm CST.
Note: If we are outside chat hours or are already taking the maximum number of chats, your support request will automatically be created into a ticket for you and we will address your ticket as soon as possible.
As mentioned in our earlier post this is a premium service that we will be charging extra for but given this is a new support channel we are offering a free beta in order to get an understanding of how to support customers well.
We are continually working to improve the support experience for customers including increasing our staffing levels and adjusting our support coverage to better manage support requests. We aren’t going to get it right first time so we appreciate your patience and we encourage you to share your feedback, good and bad, with us at email@example.com.
General Manager, Jungle Disk