Just in time for Monday’s release of Táve 4 and the 3-hour maintenance window needed for the upgrade, we’re pleased to make our private standby system available to all users.

Any time the main web application is unavailable, be it due to scheduled maintenance or an availability event in Virginia, you can visit the standby site and access a read-only version to get at your critical data.


 The Táve 4 maintenance page will include the link for your convenience.

The web servers and database are located in the AWS US-West Oregon region and pull their data from our primary AWS US-East Virginia region using database replication. In the event that the primary database failed, the standby database would simply stop updating but remain available.

Per our policy of scheduling maintenance at least 4 days in advance, we’re declaring a 3 hour maintenance window on Monday, November 10th, starting at 7AM Eastern, to perform certain highly demanded system upgrades.

We’re announcing a 90-minute downtime window this weekend starting at 11:30pm here in Boca Raton, Florida. This translates to the following times in our 5 most common timezones:

Eastern: Saturday, February 1st, at 11:30PM – Sunday, February 2nd, at 1:00AM
Central: Saturday, February 1st, from 10:30PM – Midnight
Pacific: Saturday, February 1, from 8:30PM – 10:00PM
Melbourne: Sunday, February 2nd,  from 3:30PM – 5:00PM
London: Sunday, February 2, from 4:30AM – 6:00AM

If you or your clients access a Táve link during this window, a “Maintenance in Progress” page will be displayed and will list the five timezones above.

We’ll be upgrading our primary database server software and increasing the speed and size of its storage in preparation for our major upcoming release planned for release prior to attending WPPI 2014.

Specifically, we’re upgrading the server software from MySQL 5.5 to 5.6 and increasing our storage to 100GB in order to take advantage of Amazon Web Service’s Provisioned IOPS. These improvements dramatically trim the time of the migration (and thus downtime) needed for the release mentioned above, which was 4×-6× slower and unacceptably long when tested with our current configuration.

Update: This maintenance event was performed successfully and completed ahead of schedule, allowing us to resume normal operation prior to the close of our 90-minute window. While it has been over a year since our last maintenance event, we’ll need to take another short window when rolling our our next major release, as there are substantial changes to the application and the database schema to support all the great new features in it. We’ll publish an article with the details later this month.

If you’re using Google Checkout with your Táve account, you’ll need to find an alternative online credit card processor before Google Checkout closes on November 20th. We recommend switching as soon as possible to avoid any interruption to your business.

We use Braintree for our subscription system and recommend them, as they don’t charge any setup or monthly fees and have a low 2.9% + 30¢ fixed rate (even for American Express cards). Be aware that they were acquired by PayPal last week, so their reputation for amazing support, speedy services, and data portability may well fade in time.

We currently support the following payment gateways:

  • Bank of America
  • Beanstream
  • Braintree
  • Merchant Warehouse
  • PayPal Standard
  • Stripe
  • WorldPay

Contact information and tips can be found on the Payment Gateways setup page, just click to add credit card support and select the gateway you’re interested in.

Táve 2.9 registration has been closed for the 3.0 rollout.

3.0 registration will begin on the evening of Wednesday, February 8th.

If you don’t have an existing account, please sign-up for our newsletter to receive an email when Táve 3.0 registration begins.

Updated (2/8): Registration has resumed, so register now and take your 30-day free trial for a spin!

We’ve received several reports of studios who use GMail not seeing new lead emails due to GMail marking the email as Spam. While unconfirmed, it’s possibile that emails sent from studios via Táve to GMail recipients may be affected by this issue as well. This issue affects both Táve 2 and 3.

Our outgoing mail has been independently verified to pass all anti-spam checks, including SPF, DKIM, and SpamAssassin, so we believe this to be an isolated issue with GMail.

Workaround: While we work to resolve the issue, GMail users should add a rule in their GMail settings to never mark email coming from @tave.com as spam (and we suggest having it mark the message as important). Also, check your Táve daily for new leads as they will always show up there even if the email notice doesn’t reach you.

RESOLVED (1/25):Our outgoing email is now using new IP addresses and our test emails have been delivered successfully. Please comment or post a Helpdesk ticket if a new lead message goes to your spam or isn’t delivered after 2PM Eastern today (Wednesday, January 25th).

UPDATE (1/30): The issue appears to still exist for some studios, with some users now receiving the messages and others still having them marked as spam or undelivered. We continue to investigate and are working with our email service provider for a solution. Please see the workaround in the meantime.

RESOLVED (2/6): We’ve changed email providers and are now using the same provider as companies such as Foursquare and Hoot Suite.

Update: The migration was completed successfully.  We were down for maintenance for just over an hour, about one third of our scheduled window. If you encounter any issues, please file a helpdesk ticket.

We’ll be moving from our servers based in Dallas, Texas, to Amazon’s AWS cloud starting at 11pm EST (Click the link to view the time in major cities around the globe). This will be a 3-hour window, but we’re aiming to have the system back online within an hour.

This blog and files uploaded by users in Táve 2.9 are already in the AWS cloud and every piece of the Táve 3.0 beta has been in AWS from the start as well. Wednesday’s migration completes our transition by moving the remaining 2.9 infrastructure. We’d wait until after the 3.0 release, but having 2.9 on AWS before then will dramatically speed up the studio-initiated upgrade system and also makes it easier for us to maintain both systems, as they’ll now have similar infrastructure behind them.

If this migration announcement seems familiar, you may recall we had planned on migrating back in March but postponed it due to an issue with Amazon’s email services. That has been resolved and in fact the new email service provider means we can now send emails from your studio without listing Táve as the Sender in Táve 3.0.

We’ll be moving from our servers based in Dallas, Texas, to Amazon’s AWS cloud starting at 3am EDT (that’s one minute after 1:59am EST, as the U.S. goes to Daylight Savings Time at the start of the maintenance window. Click the link to view the time in major cities around the globe). This will be a 3-hour window, but we’re aiming to have the system back online within an hour.

This blog and the Táve 3.0 beta have both joined the likes of Netflix and Yelp by living entirely within the Amazon cloud. We recently transitioned all user-uploaded files in Táve 2 over to Amazon S3 as well. These tests have been so positive that we’re ready to move the entire infrastructure over to AWS, as providing a fast and rock-solid experience in Táve 2 is our top priority.

With our current setup, adding new servers is a time consuming process that generally takes at least a week before the new servers are racked, stacked, and in the pool. Using AWS, we can add new server instances to the pool in just minutes. This also minimizes the need for downtime windows for server maintenance, as we can simply create new servers with updated software and just replace the outdated instances. Another exciting feature, besides the incredible global bandwidth Amazon provides, is the ability to automatically scale the number of servers to meet increased demand as it happens without any action on our part.

Update: Due to complications integrating with the Amazon email services, we are unable to perform the migration as scheduled and have canceled this maintenance window. A workaround is in place and we plan to reschedule the migration once we’re settled into our new office in Fort Lauderdale.

We just published a small Táve 2.9 maintenance release that we felt needed a brief mention as it may affect or pique the interest of a few of our users.

Release 2.9.5 moves all user uploaded files and email attachments into Amazon’s S3 service. When you or a client accesses one of these files, they’ll be automatically redirected to a the new S3 location. For security reasons, the URL expires after a few minutes, so if you link to the file from your website or other location, you’ll want to continue to use the URL provided inside the application prior to the redirect.

While the rest of our migration to the Amazon Web Services will wait for the launch of Táve 3.0, applying this change to the current Táve 2.9 release dramatically decreases the size of our database, speeds up our backups, and makes deploying new Táve 3.0 betas and the final release much faster. Update: We’ve decided to move the entire system over to Amazon prior to the T3 release, see the blog post for details.

It’s also worth noting that the tave.net blog itself is already completely inside Amazon’s cloud, using their EC2, RDS, S3, SES, Route 53, and CloudFront services (which are just a subset of the AWS services the Táve 3.0 beta is currently using).

This is a 48-hour notice for a planned 2-hour downtime event on the Wednesday night (Thursday morning) starting at 1AM Eastern (10PM Pacific, 0500 GMT).

Jason and I will be performing system maintenance and server upgrades in preparation for our big summer releases and all systems are affected.

While these upgrades are a necessary precursor to the Táve 3.0 release, no software is being released during this maintenance event. If you’re longing for news on Táve 3.0, please be sure to sign up for our Táve 3.0 announcement list (as we’re not quite ready to announce anything more on that front just yet).

Status: Maintenance complete.