Food safety is a major concern after a power outage.

According to FoodSafety.gov, your refrigerator will keep food safe for up to four hours during a power outage.

A full freezer will keep food safe for roughly 48 hours.

People should not taste the food after a power outage to determine its safety.

According to the United States Department of Agriculture, refrigerators should be set to maintain a temperature of 40 degrees Fahrenheit or below.

Some refrigerators have built-in thermometers to measure their internal temperature. If your fridge does not, keep an appliance thermometer in the refrigerator to monitor the temperature.

The USDA says this can be critical in the event of a power outage. When the power goes back on, if the refrigerator is still 40 degrees Fahrenheit, the food is safe.

Foods held at temperatures above 40 degrees Fahrenheit for more than two hours should not be consumed, according to the USDA.

For tips on when to throw out certain refrigerated and frozen foods, click here.

Amazon Web Services suffered a major outage Tuesday, the company said, limiting service at many key and popular sites.The company provides cloud computing services to many governments, universities and companies, including The Associated Press.Amazon said in a post an hour after the outage began that it had identified the root cause and was “actively working towards recovery.” It did not disclose more about the cause.The outage also affected Amazon’s ability to provide updates, it said.The outage began midmorning on the U.S. East Coast, said Doug Madory, director of internet analysis at Kentik Inc, a network intelligence firm. “AWS is the biggest cloud provider and us-east-1 is their biggest data center, so any disruption there has big impacts to many popular websites and other internet services,” he said.Madory said he did not believe the outage was anything nefarious. He said a recent cluster of outages at providers that host major websites reflects how the networking industry has evolved. “More and more these outages end up being the product of automation and centralization of administration,” he said. “This ends up leading to outages that are hard to completely avoid due to operational complexity, but are very impactful when they happen.”Kentik was seeing a 26% drop in traffic to Netflix, among major web-based services affected by the outage, Madory said. According to Down Detector, a clearinghouse for user reports on outages, Delta and Southwest have been affected, but not American, United, Alaska or JetBlue.People trying to use Instacart, Venmo, Kindle, Roku, and Disney+ have reported issues. The McDonald’s app is also down.Check back for updates on this story.

The web pages of several major companies including Amazon, AT&T, UPS and US Bank were briefly impacted by an apparent outage Thursday.

According to Downdetector.com, a website that tracks internet outages, the problems began for several websites late Thursday morning.

The cause of the outages is not known at this time. Internet company Akamai acknowledged the outage on its website saying they "are aware of an emerging issue with the Edge DNS service and that they are "actively investigating the issue."

This content is imported from Twitter.
You may be able to find the same content in another format, or you may be able to find more information, at their web site.

This content is imported from Twitter.
You may be able to find the same content in another format, or you may be able to find more information, at their web site.

As services began to get back online, Akamai tweeted, "We have implemented a fix for this issue, and based on current observations, the service is resuming normal operations. We will continue to monitor to ensure that the impact has been fully mitigated."

This content is imported from Twitter.
You may be able to find the same content in another format, or you may be able to find more information, at their web site.

BOSTON —

The single biggest global ransomware attack yet continued to bite Monday as details emerged on how the Russia-linked gang responsible breached the company whose software was the conduit. In essence, the criminals used a tool that helps protect against malware to spread it widely.

An affiliate of the notorious REvil gang, best known for extorting $11 million from the meat-processor JBS after a Memorial Day attack, infected thousands of victims in at least 17 countries on Friday, largely through firms that remotely manage IT infrastructure for multiple customers, cybersecurity researchers said.

REvil was demanding ransoms of up to $5 million. But late Sunday it offered in a posting on its dark web site a universal decryptor software key that would unscramble all affected machines in exchange for $70 million in cryptocurrency. It wasn't clear who they expected might pay that amount.

Sweden may have been hardest hit by the attack — or at least most transparent about it. Its defense minister, Peter Hultqvist, bemoaned on Monday "a serious attack on basic functions in Swedish society."

"It shows how fragile the system is when it comes to IT security and that you must constantly work to develop your ability to defend yourself," he said in a TV interview. Most of the Swedish grocery chain Coop's 800 stores were closed all weekend because their cash register software supplier was crippled. They remained closed Monday. A Swedish pharmacy chain, gas station chain, the state railway and public broadcaster SVT were also hit.

A broad array of businesses and public agencies were affected, including in financial services, travel and leisure and the public sector — though few large companies, the cybersecurity firm Sophos reported. The cybersecurity firm ESET identified victims in countries including the United Kingdom, South Africa, Canada, Argentina, Mexico, Indonesia, New Zealand and Kenya.

Ransomware criminals infiltrate networks and sow malware that cripples them by scrambling all their data. Victims get a decoder key when they pay up.

In Germany, an unnamed IT services company told authorities several thousand of its customers were compromised, the news agency dpa reported. Also among reported victims were two big Dutch IT services companies — VelzArt and Hoppenbrouwer Techniek. Most ransomware victims don't publicly report attacks or disclose if they've paid ransoms.

On Sunday, the FBI said in a statement that while it was investigating the attack, its scale "may make it so that we are unable to respond to each victim individually." Deputy National Security Advisor Anne Neuberger later issued a statement saying President Joe Biden had "directed the full resources of the government to investigate this incident" and urged all who believed they were compromised to alert the FBI.

Biden suggested Saturday the U.S. would respond if it was determined that the Kremlin is at all involved. Less than a month ago, Biden pressed Russian President Vladimir Putin to stop giving safe haven to REvil and other ransomware gangs whose unrelenting extortionary attacks the U.S. deems a national security threat.

On Monday, Putin spokesman Dmitry Peskov was asked if Russia was aware of the attack or had looked into it. He said no, but suggested it could be discussed by the U.S. and Russia in consultations on cybersecurity issues for which no timeline has been specified.

Experts say it was no coincidence that REvil launched the attack at the start of the Fourth of July holiday weekend, knowing U.S. offices would be lightly staffed and many victims might not learn of it until back at work Monday or Tuesday.

Most end users of managed service providers "have no idea" whose software keep their networks humming, said CEO Fred Voccola of the breached software company, Kaseya.

He estimated the victim number in the low thousands, mostly small businesses like "dental practices, architecture firms, plastic surgery centers, libraries, things like that."

Voccola said only between 50-60 of the company's 37,000 customers were compromised. But 70% were managed service providers who use the company's hacked VSA software to manage multiple customers. It automates the installation of software and malware-detection updates and manages backups and other vital tasks.

Kaseya said it sent a detection tool to nearly 900 customers on Saturday night.

The REvil offer to offer blanket decryption for all victims of the Kaseya attack in exchange for $70 million suggested its inability to cope with the sheer quantity of infected networks, said Allan Liska, an analyst with the cybersecurity firm Recorded Future.

But Kevin Reed of Acronis said the offer of a universal decryptor could be a PR stunt because no human involvement would be needed to pay a $45,000 base ransom demand apparently sent to the vast majority of targets. Analysts reported seeing demands of $5 million and $500,000 for bigger targets, which would require negotiation.

Analyst Brett Callow of Emsisoft said he suspects REvil is hoping insurers might crunch the numbers and determine the $70 million will be cheaper for them than extended downtime.

Sophisticated ransomware gangs on REvil's level usually examine a victim's financial records — and insurance policies if they can find them — from files they steal before activating the ransomware. The criminals then threaten to dump the stolen data online unless paid, although that does not appear to have happened in this case. But this attack was apparently bare-bones. REvil seems only to have scrambled victims' data.

Dutch researchers said they alerted Miami-based Kaseya to the breach and said the criminals used a "zero day," the industry term for a previous unknown security hole in software. Voccola would not confirm that or offer details of the breach — except to say that it was not phishing.

"The level of sophistication here was extraordinary," he said.

It was not the first ransomware attack to leverage managed services providers. In 2019, criminals hobbled the networks of 22 Texas municipalities through one. That same year, 400 U.S. dental practices were crippled in a separate attack.

Active since April 2019, REvil provides ransomware-as-a-service, meaning it develops the network-paralyzing software and leases it to so-called affiliates who infect targets and earn the lion's share of ransoms. U.S. officials say the most potent ransomware gangs are based in Russia and allied states and operate with Kremlin tolerance and sometimes collude with Russian security services.

___

AP reporters Jim Heintz in Moscow, Jan Olsen in Stockholm, Kirsten Grieshaber in Berlin, Jari Tanner in Helsinki and Sylvie Corbet in Paris contributed to this report.

CNN —

Related video above: Hubble commemorates its 30th year with new images

Universe, we've got a problem: The payload computer aboard the Hubble Space Telescope that has shaped our understanding of the cosmos for over 30 years has stopped working.

After the June 13 malfunction, Hubble's payload computer stopped receiving the "keep-alive" signal that's a "standard handshake between the payload and main spacecraft computers to indicate all is well," according to a NASA news release.

Following the main computer's automatic shift of all science instruments to safe mode, control center personnel at NASA's Goddard Space Flight Center in Greenbelt, Maryland, restarted the payload computer but it soon halted once again on June 14. The telescope and other science instruments remain in good health, NASA reported.

The payload computer -- a NASA Standard Spacecraft Computer-1, or NSSC-1, system built in the 1980s -- is part of the Science Instrument Command and Data Handling unit, a module on the Hubble Space Telescope that communicates commands to Hubble's science instruments and formats data for transmission to the ground. The current unit is a replacement that was installed by astronauts on shuttle mission STS-125 in May 2009 after the original unit failed in 2008, which delayed the final servicing mission to Hubble while NASA constructed the replacement.

The payload computer's purpose is to control and coordinate the science instruments on the telescope, and monitor their health and safety, according to NASA. The computer's programs also analyze and manipulate the data it collects. The computer is critical, but there is a second computer the operations team can switch to if problems arise.

Based on early data, the Hubble operations team initially thought a degrading memory module halted the computer. After the failed restart and attempt to switch to a backup memory module, the team tried to gain more information while again trying to bring the memory modules online, which also fell flat.

After several tests of the computer's memory modules, the investigators found that a different piece of computer hardware might have been the culprit behind the memory errors: The operations team has started exploring whether the problem lies within the Standard Interface, or STINT, hardware -- which facilitates messaging between the computer's Central Processing Module and other parts -- or the CPM itself.

The operations team members are also designing tests they'll soon run to pinpoint potential problems and a possible solution -- which, for future reference, would be instrumental for figuring out which hardware is still working properly when something else isn't.

If the problem isn't fixed, the Hubble operations team will be prepared to switch to the STINT and CPM hardware in the backup payload computer, NASA reported.

If the operations team members use plan B, they'll then need several days to see how the backup computer performs and to resume the normal operations for which the main payload computer is typically responsible. The backup hasn't been used since its installation in 2009, NASA reported, but it was "thoroughly tested on the ground prior to installation on the spacecraft."

Launched aboard Space Shuttle Discovery in 1990, the Hubble Space Telescope has shared its observations of stars, galaxies and other astronomical objects "that have captured imaginations worldwide and deepened our knowledge of the cosmos," NASA reported.

That includes its role in narrowing down the universe's age from the once estimated 10 billion to 20 billion years old to about 13.8 billion years, a number now used to understand the chronological development of stars, galaxies and more.