Today (06/30/2015) we will have a second more, what could be the consequences for our systems?

Question

In case anyone doesn’t know, today 30/06/2015 We’ll have an extra second. That is, today after the 23h59m59s we shall have 23h59m60s!

This to make the correction of International Atomic Time.

I know that there are several services that are synchronized with atomic clocks, like the one in the Observatório Nacional, who may have problems with this second more.

What kinds of problems can we expect in our systems/servers and applications when an event like this occurs?

In case you don’t know anything, see some stories talking about this:

National Radio Agency - Watches will have a second extra for gravitational changes

Tecmundo - June 30 will have 1 second more, but what this can cause?

Answer 1

Getting around the problem

Watches do not normally account for the second of number 60, so some alternatives should be taken in this regard. Some possibilities are:

• Some Linux kernels implement a jump behind 1s, repeating the 59°second. For more information: Resolve Leap Second Issues in Red Hat Enterprise Linux (in English);
• Windows servers ignore the second 60, making it synchronize again with the atomic clocks soon after it passes. This means that they count twice the second 0 of the day 1° of July. For more information: How the Windows Time service Treats a Leap Second (in English);
• Some organizations, including Amazon Web Services, plan to divide and spread the extra second for several hours by making each second a little longer (the English term is "Leap Smear");
• If the clock does not connect to a synchronization system, it simply does not implement any kind of adjustment to it.

Source: Look Before You Leap - The Coming Leap Second and AWS

Possible complications

Many technological devices synchronize their watches with an atomic clock. However, many of them were not programmed to consider the possibility of the second extra happening, so when the system identifies it it presents an unforeseen result, which may result in crashes of servers and consequently in decline of their services.

In 2012, Mozilla, Reddit, Foursquare, Yelp, Linkedin and Stumbleupon presented crashes system when the second extra was added. Already Google, which used the tactic of "Leap Smear", escaped unscathed from this situation.

This year it is expected that some servers present this problem again.

Source: Daily News - 'Leap Second' Coming up June 30 may cause computer system problems

Update: What were the damage of the second extra 2015?

Although AWS said it was not the fault of the second extra, their services were off the air for just over 40 minutes, but no ace 00:00 UTC and yes of 00:25 at 01:07 UTC, leaving off-air services like Slack, Netflix, Pinterest and thousands of other websites and services.

The news:

Between 5:25 PM and 6:07 PM PDT we Experienced an Internet Connectivity Issue with a Provider Outside of our network which affected Traffic from some end-user Networks. The Issue has been resolved and the service is Operating normally.

The root cause of this Issue was an External Internet service Provider incorrectly Accepting a set of Routes for some AWS Addresses from a third-party who inadvertently advertised These Routes. Providers should normally Reject These Routes by policy, but in this case the Routes Were accepted and propagated to other Isps affecting some end-user’s Ability to access AWS Resources. Once we identified the Provider and third-party network, we Took action to route Traffic Around this incorrect routing Configuration. We have worked with this External Internet service Provider to ensure that this does not reoccur.

Source: AWS Service Health Dashboard

According to them, the fault was external servers who incorrectly accepted a set of routes to some AWS addresses that were inadvertently announced by third parties... Is that not clear to you? Not to me either. The fact is that many people suspect that the problem was yes of the second extra, although AWS states that.

Source: Mashable - Slack, Netflix, Pinterest crash and you can’t Blame the Leap Second

Answer 2

This question is related to one of the themes of one of the most famous answers from Soen, by the renowned Jon Skeet, in the Soen.

There will be a time discontinuity.

A discontinuity can happen in two ways:

• Clockwork
• Clock delay

Both can cause problems if your code is not handled correctly.

In this case, there will be a clock delay, on time 23:59:59 in the order of 1 second.

---

Despite being treated differently, it is worth noting that in summer time also occurs exactly this same problem of discontinuity, and for even longer than 1 second: 1 hour!

As I have worked on collection systems, I can say that the following cases happen:

• Clockwork (00:00:00 pula to 01:00:00)
  • A connection is started in 23:59:00, and finalised in 01:00:00: the bond lasted only 00:01:00.
• Clock delay (00:00:00 back to 23:00:00)
  • A connection is started in 23:30:00, and finalised in 23:31:00: the connection may have lasted as long as 00:01:00how much also 01:01:00.

One solution can be to use (underneath the cloths) a neutral time form (that disregards these discontinuities), another would be to treat them.

Answer 3

What will happen is that on the clock it will hold a second during the turn of the night.

Note: The information below varies from S.O. and Programming Language.

Tip: Forward your computer clock, and test the application.

Example that will occur 23h 59m and 59s: 50,51,52,53,54,55,56,57,58,59,59,00

This second will be presented twice the impacts for financial systems and that use time as a financial base will be the processing of the same second "23:59:59" twice.

But each application has its function, and the developer must know what will be the behavior that your code will present in this situation.

Congratulations on the question!

Answer 4

This extra second can cause several problems in the systems we develop and use, from a simple exception until the system is completely out of the air.

This is because most systems are not prepared to deal with this second too much, for these systems the second 60 does not exist.

In 2012, for example, the last time that a second extra needed to be included in the atomic clocks, Mozilla, Reddit, Foursquare, Yelp, Linkedin, Stumbleupon, the Linux operating system, programs developed in Java, for example; showed failures and many even went off the air.

The reason for this problem is due to problems in the code of the operating systems or the support applications that upon seeing this second additional appear end up freezing due to bugs or programming failure.

So far only google has spoken out about the case, and claimed that several techniques are being developed to deal with the problem and not affect users.

Sources: Facts Unknown, Pplware

Answer 5

According to the very link you passed, there is the following information:

The correction of the "Second Intercalated" is determined by the International Service of Earth Reference and Rotation Systems. The last adjustment took place on December 31, 2012.

These one-second adjustments have been made since 1972 and as of July this year, the difference between International Atomic Time and Coordinated Universal Time (UTC) will be 36 seconds.

Probably the servers will advance their schedules and even if there was the time 23:59:60 technologies are provided for this, in this case the 23:59:60 would probably be converted to 00:00:00, also note computer systems operating in UTC/GMT.

An example with the javascript engine to test:

var test = document.getElementById("test");

document.getElementById("run").onclick = function() {
    //new Date(anom, mês, dia, horas, minutos, segundos, milisegundos)
    var date = new Date(2015, 05, 30, 23, 59, 60, 0);
    test.innerHTML = date;
};

<div id="test"></div>
<button id="run">Testar</button>