Andreessen-Backed Teleport Launches A Mobile Search Engine For Globally Nomadic Tech Workers

teleport It’s a big hairy vision — knowledge-workers freely migrate around the globe, and city and federal governments compete to lure them. Teleport, a startup founded by early Skype employees around the idea of supporting an increasingly global mobile workforce, is taking another step toward that reality today. It’s launching a set of iOS and Android apps that help tech workers find… Read More


Microsoft Jumps Into The Mobile Device Management Fray

microsoft-data Microsoft made good on its promise to add mobile device management (MDM) to its Office 365 productivity service, making the feature generally available today to commercial customers. Microsoft has added, in English, the ability for large companies to manage Office on the phones of their employees, regardless of whether the handset in use is a personal or corporate device. Read More


Percona XtraDB Cluster 5.5.41-25.11 is now available

Percona XtraDB Cluster 5.5.41-25.11Percona is glad to announce the new release of Percona XtraDB Cluster 5.5 on March 30th 2015. Binaries are available from downloads area or from our software repositories.

Based on Percona Server 5.5.41-37.0 including all the bug fixes in it, Galera Replicator 2.11, and on Codership wsrep API 25.11, Percona XtraDB Cluster 5.5.41-25.11 is now the current 5.5 General Availability release. All of Percona‘s software is open-source and free, and all the details of the release can be found in the 5.5.41-25.11 milestone at Launchpad.

Bugs Fixed:

  • XtraBackup SST wouldn’t stop when MySQL was SIGKILLed. This would prevent MySQL to initiate a new transfer as port 4444 was already utilized. Bug fixed #1380697.
  • garbd was returning incorrect return code, ie. when garbd was already started, return code was 0. Bugs fixed #1308103 and #1422863.
  • wsrep_sst_xtrabackup-v2 script was causing innobackupex to print a false positive stack trace into the log. Bug fixed #1407599.
  • MyISAM DDL (CREATE TABLE only) isn’t replicated anymore when wsrep_replicate_myisam is OFF. Note, for older nodes in the cluster, wsrep_replicate_myisam should work since the TOI decision (for MyISAM DDL) is done on origin node. Mixing of non-MyISAM and MyISAM tables in the same DDL statement is not recommended with wsrep_replicate_myisam OFF since if any table in list is MyISAM, the whole DDL statement is not put under TOI (total order isolation). This also doesn’t work if default_storage_engine variable is set to MyISAM (which is not recommended for Percona XtraDB Cluster) and a table is created without the ENGINE option. Bug fixed #1402338.
  • Percona XtraDB Cluster now shows a warning in case additional utilities, like pv which may not affect critical path of SST, are not installed. Bug fixed #1248688.
  • wsrep_causal_reads variable was not honored when declared as global. Bug fixed #1361859.
  • garbd would not work when cluster address was specified without the port. Bug fixed #1365193.
  • garbd was running as root user on Debian. Bug fixed #1392388.
  • Errors in garbd init script stop/start functions have been fixed. Bug fixed #1367956.
  • If mysqld gets killed during the SST it will leave an unclean data directory behind. This would cause Percona XtraDB Cluster to fail when the server would be started next time because the data directory would be corrupted. This was fixed by resuming the startup in case wsrep-recover failed to recover due to corrupted data directory. The old behavior is still achievable through --exit-on-recover-fail command line parameter to mysqld_safe or exit-on-recover-fail under [mysqld_safe] in my.cnf. Bug fixed #1378578.
  • gvwstate.dat file was removed on joiner when XtraBackup SST method was used. Bug fixed #1388059.
  • xtrabackup-v2 SST did not clean the undo log directory. Bug fixed #1394836.
  • stderr of SST/Innobackupex is logged to syslog with appropriate tags if sst-syslog is in [sst] or [mysqld_safe] has syslog in my.cnf. This can be overridden by setting the sst-syslog to -1 in [sst]. Bug fixed #1399134.
  • clustercheck can now check if the node is PRIMARY or not, to allow for synced nodes which go out of PRIMARY not to take any writes/reads. Bug fixed #1403566.
  • Race condition between donor and joiner in Xtrabackup SST Configuration has been fixed. This caused XtraBackup SST to fail when joiner took longer to spawn the second listener for SST. Bug fixed #1405668.
  • SST will now fail early if the xtrabackup_checkpoints file is missing on the joiner side. Bug fixed #1405985.
  • socat utility was not properly terminated after a timeout. Bug fixed #1409710.
  • 10 seconds timeout in Xtrabackup SST Configuration script was not enough for the joiner to delete existing files before it started the socat receiver on systems with big datadir. Bug fixed #1413879.
  • Conflict between enforce_storage_engine and wsrep_replicate_myisam for CREATE TABLE has been fixed. Bug fixed #1435482.
  • SST processes are now spawned with fork/exec instead of posix_spawn to allow for better cleanup of child processes in event of non-graceful termination (SIGKILL or a crash etc.). Bug fixed #1382797.
  • Variable length arrays in WSREP code were causing debug builds to fail. Bug fixed #1409042.
  • Signal handling in mysqld has been fixed for SST processes. Bug fixed #1399175.
  • Inserts to a table with autoincrement primary key could result in duplicate key error if another node joined or dropped from the cluster during the insert processing. Bug fixed #1366997.

Other bugs fixed: #1391634 and #1396757.

Release notes for Percona XtraDB Cluster 5.5.41-25.11 are available in our online documentation along with the installation instructions.

Help us improve our software quality by reporting any bugs you encounter using our bug tracking system. As always, thanks for your continued support of Percona!

Please also note that Percona XtraDB Cluster 5.6 series is the latest General Availability series and current GA release is 5.6.22-25.8.

The post Percona XtraDB Cluster 5.5.41-25.11 is now available appeared first on MySQL Performance Blog.


Box CEO Aaron Levie Talks IPO Delay And Growth Strategies

Aaron Levie, Box CEO, speaking at TechCrunch Disrupt Berlin in 2013. Regardless, Levie didn’t have much time to savor the moment. He told TechCrunch in a wide-ranging interview last month that Box has no intention of standing still and he discussed his plans to keep the company growing, while trying to find ways to get more efficient. He also talked about the long road to IPO and what was behind Box’s thinking in delaying it for so long. One of the… Read More


‘Woz on your mind?’ Share your questions for Steve Wozniak during his Percona Live keynote!

Here’s your chance to get on stage with Woz! Sort of. Apple co-founder and Silicon Valley icon and philanthropist Steve Wozniak will participate in a moderated Q&A on creativity and innovation April 14 during the Percona Live MySQL Conference and Expo in Santa Clara, California.

Woz once said that he never intended to change the world. That was the other Steve, Steve Jobs.

“I didn’t want to start this company,” Woz told the Seattle Times of Apple’s beginnings in a 2006 interview. “My goal wasn’t to make a ton of money. It was to build good computers. I only started the company when I realized I could be an engineer forever.”

What would you ask Woz if given the opportunity?

“Woz, what first sparked your interest in engineering?”
“Hey Woz, how did you come up with the design for the first Apple?”
“Woz, what do you see as the next big thing in personal computers?”
“Hi Woz, what’s the deal with your giant vacuum tube watch?”

Now it’s your turn! Ask a question in the comments below and be sure to include your Twitter handle – or your Facebook page or LinkedIn profile. If we use your question, then your profile and question will be displayed on the giant screen behind Woz on stage as it’s being asked during his big keynote! How cool is that?

Want to be there in person? See Woz speak for just $5! That’s $70 off the regular admission price! Just use the promo code “KEY” at registration under the “Expo Hall and Keynote Pass” selection. Following Woz’s keynote, be sure to stop by the Percona booth, say “hello, Tom,” and I’ll give you a limited-edition Percona t-shirt. :)

In the meantime, help spread the word! Please share this tweet:

“Woz on your mind?” Tweet @Percona your questions for Apple’s Steve Wozniak who speaks April 14 at #PerconaLive!

Do that, then follow @Percona and I’ll send a DM for your address and will ship a t-shirt right to your door. See you at the conference!

The post ‘Woz on your mind?’ Share your questions for Steve Wozniak during his Percona Live keynote! appeared first on MySQL Performance Blog.


FoundationDB is acquired by Apple: My thoughts

TechCrunch reported yesterday that Apple has acquired FoundationDB. And while I didn’t see any mention if this news on the FoundationDB website, they do have an announcement saying: “We have made the decision to evolve our company mission and, as of today, we will no longer offer downloads.”

This is an unfortunate development – I have been watching FoundationDB technology for years and was always impressed in terms of its performance and features. I was particularly impressed by their demo at last year’s Percona Live MySQL and Expo. Using their Intel NUC-based Cluster, I remember Ori Herrnstadt showing me how FoundationDB handles single-node failure as well as recovery from complete power-down – very quickly and seamlessly. We have borrowed a lot of ideas from this setup for our Percona XtraDB Cluster Demos.

I think it was a great design to build a distributed, shared-nothing transaction aware key value store, and then have an SQL Layer built on top of it. I did not have a chance to test it hands-on, though. Such a test would have revealed the capabilities of the SQL optimizer – the biggest challenge for distributed relational database systems.

My hope was to see, over time, this technology becoming available as open source (fully or partially), which would have dramatically increased adoption by the masses. It will be interesting to see Apple’s long-terms plans for this technology.

In any case it looks like FoundationDB software is off limits. If you are an existing FoundationDB customer looking for alternatives, we here at Percona would be happy to help evaluate options and develop a migration strategy if necessary.

The post FoundationDB is acquired by Apple: My thoughts appeared first on MySQL Performance Blog.


FinancialForce Hauls In $110M Investment

Pile of US bills; packet of hundreds on top of loose one dollar bills. FinancialForce, the cloud ERP system (enterprise resource planning) built on top of the Salesforce1 platform, announced a new round of funding tonight, landing $110M in additional investment. This round follows a $50M round just last April. Today’s investment is led by Technology Crossover Ventures with additional money coming from Salesforce Ventures. It brings the total investment… Read More


Yelp IT! A talk with 3 Yelp MySQL DBAs on Percona Live & more

elp IT! A talk with 3 Yelp MySQL DBAs heading to Percona Live 2015Founded in 2004 to help people find great local businesses, Yelp has some 135 million monthly unique visitors. With those traffic volumes Yelp’s 300+ engineers are constantly working to keep things moving smoothly – and when you move that fast you learn many things.

Fortunately for the global MySQL community, three Yelp DBAs will be sharing what they’ve learned at the annual Percona Live MySQL Conference and Expo this April 13-16 in Santa Clara, California.

Say “hello” to Susanne Lehmann, Jenni Snyder and Josh Snyder! I chatted with them over email about their presentations, on how MySQL is used at Yelp, and about the shortage of women in MySQL.


Tom: Jenni, you and Josh will be co-presenting “Next generation monitoring: moving beyond Nagios ” on April 14.

You mentioned that Yelp’s databases scale dynamically, and so does your monitoring of those databases. And to minimize human intervention, you’ve created a Puppet and Sensu monitoring ensemble… because “if it’s not monitored, it’s not in production.” Talk to me more about Yelp’s philosophy of “opt-out monitoring.” What does that entail? How does that help Yelp?

Jenni: Before we moved to Sensu, our Nagios dashboards were a sea of red, muted, acknowledged, or disabled service checks. In fact, we even had a cluster check to make sure that we never accidentally put a host into use that was muted or marked for downtime. It was possible for a well-meaning operator to acknowledge checks on a host and forget about it, and I certainly perpetrated a couple of instances of disks filling up after acknowledging a 3am “warning” page that I’d rather forget about. With Sensu, hosts and services come out of the downtime/acknowledgement state automatically after a number of days, ensuring that we’re kept honest and stay on top of issues that need to be addressed.

Also, monitoring is deployed with a node, not separate monitoring configuration. Outside of a grace period we employ when a host is first provisioned or rebooted, if a host is up, it’s being monitored and alerting. Also, alerting doesn’t always mean paging. We also use IRC and file tickets directly into our tracking system when we don’t need eyes on a problem right away.

Tom: Susanne, in your presentation, titled “insert cassandra into prod where use_case=?;” you’ll discuss the situations you’ve encountered where MySQL just wasn’t the right tool for the job.

What led up to that discovery and how did you come up with finding the right tools (and what were they) to run alongside and support MySQL?

Susanne: Our main force behind exploring other datastores alongside MySQL was that Yelp is growing outside the US market a lot. Therefore we wanted the data to be nearer to the customer and needed multi-master writes.

Also, we saw use cases where our application data was organized very key-value like and not relational, which made them a better fit for a NoSQL solution.

We decided to use Cassandra as a datastore and I plan to go more into detail why during my talk. Now we offer developers more choices on how to store our application data, but we also believe in the “right tool for the job” philosophy and might add more solutions to the mix in the future.

Tom: Jenni, you’ll also be presenting “Schema changes multiple times a day? OK!” I know that you and your fellow MySQL DBAs are always improving and also finding better ways of supporting new and existing features for Yelp users like me. Delivering on such a scale must entail some unique processes and tools. Does this involve a particular mindset among your fellow DBAs? Also, what are some of those key tools – and processes and how are they used?

Jenni: Yelp prizes the productivity of our developers and our ability to iterate and develop new features quickly. In order to do that, we need to be able to not only create new database tables, but also modify existing ones, many of which are larger than MySQL can alter without causing considerable replication delay. The first step is to foster a culture of automated testing, monitoring, code reviews, and partnership between developers and DBAs to ensure that we can quickly & safely roll out schema changes. In my talk, I’ll be describing tools that we’ve talked about before, like our Gross Query Checker, as well as the way the DBA team works with developers while still getting the rest of our work done. The second, easy part is using a tool like pt-online-schema-change to run schema changes online without causing replication delay or degrading performance :)

Tom: Josh, you’ll also be speaking on “Bootstrapping databases in a single command: elastic provisioning for the win.” What is “elastic provisioning” and how are you using it for Yelp’s tooling?

Josh: When I say that we use elastic provisioning, I mean that we can reliably and consistently build a database server from scratch, with minimal human involvement. The goal is to encompass every aspect of the provisioning task, including configuration, monitoring, and even load balancing, in a single thoroughly automated process. With this process in place, we’ve found ourselves able to quickly allocate and reallocate resources, both in our datacenters and in the cloud. Our tools for implementing the above goals give us greater confidence in our infrastructure, while avoiding single-points of failure and achieving the maximum possible level of performance. We had a lot of fun building this system, and we think that many of the components involved are relevant to others in the field.

Tom: Susanne and Jenni, last year at Percona Live there was a BoF session titled “MySQL and Women (or where are all the women?).” The idea was to discuss why there are “just not enough women working on the technology side of tech.” In a nutshell, the conversation focused on why there are not more women in MySQL and why so relatively few attend MySQL conferences like Percona Live.

The relative scarcity of women in technical roles was also the subject of an article published in the August 2014 issue of Forbes, citing a recent industry report.

Why, in your (respective) views, do you (or don’t) think that there are so few women in MySQL? And how can this trend be reversed?

Susanne: I think there are few women in MySQL and the reasons are manifold. Of course there is the pipeline problem. Then there is the problem, widely discussed right now, that women who are entering STEM jobs are less likely staying in there. These are reasons not specific for MySQL jobs, but rather for STEM in general. What is more specific for database/MySQL jobs is, in my opinion, that often times DBAs need to be on call, they need to stay in the office if things go sideways. Database problems tend often to be problems that can’t wait till the next morning. That makes it more demanding when you have a family for example (which is true for men as well of course, but seems still to be more of a problem for women).

As for how to reverse the trend, I liked this Guardian article because it covers a lot of important points. There is no easy solution.

I like that more industry leaders and technology companies are discussing what they can do to improve diversity these days. In general, it really helps to have a great professional (female) support system. At Yelp, we have AWE, the Awesome Women in Engineering group, in which Jenni and I are both active. We participate in welcoming women to Yelp engineering, speaking at external events and workshops to help other women present their work, mentoring, and a book club.

Jenni: I’m sorry that I missed Percona Live and this BoF last year; I was out on maternity leave. I believe that tech/startup culture is a huge reason that fewer women are entering and staying these days, but a quick web search will lead you to any number of articles debating the subject. I run into quite a few women working with MySQL; it’s large, open community and generally collaborative and supportive nature is very welcoming. As the article you linked to suggests, MySQL has a broad audience. It’s easy to get started with and pull into any project, and as a result, most software professionals have worked with it at some time or another.

On another note, I’m happy to see that Percona Live has a Code of Conduct. I hope that Percona and/or MySQL will consider adopting a Community Code of Conduct like Python, Puppet, and Ubuntu. Doing so raises the bar for all participants, without hampering collaboration and creativity!

* * *

Thanks very much, Susanne, Jenni and Josh! I look forward to seeing you next month at the conference. And readers, if you’d like to attend Percona Live, use the promo code Yelp15 for 15% off! Just enter that during registration. If you’re already attending, be sure to tweet about your favorite sessions using the hashtag #PerconaLive. And if you need to find a great place to eat while attending Percona Live, click here for excellent Yelp recommendations. :)

The post Yelp IT! A talk with 3 Yelp MySQL DBAs on Percona Live & more appeared first on MySQL Performance Blog.


LogMeIn’s Rescue Lens Lets Customer Support Reps See The Problem Through Your Smartphone’s Camera

Screen Shot 2015-03-24 at 12.15.05 PM LogMeIn, the makers of a variety of remote access solutions for business and personal use, is today launching a new service that will allow any company to provide video-assisted customer support, using only the camera on a customer’s iOS or Android mobile device. The idea is that a customer support representative would be able to tap into a live feed in order to remotely help with… Read More


Kabbage’s SMB Loan Platform Moves Into Asia Pacific As A White-Label Service

white cabbage Kabbage has taken root in the U.S. and UK markets as a platform where small businesses can quickly apply for and receive working capital loans — with Kabbage making fast decisions about eligibility through a mix of smart algorithms and online and offline data sources. Now the startup is growing its business on two fronts. It’s kicking off a new white-label offering where… Read More

Powered by WordPress | Theme: Aeros 2.0 by