When a deleted primary device file only takes 20 mins out of your maintenance window, but a whole year off your lifespan

In praise of UNIX and knowing when to deliberately drive a server off a cliff


Who, Me? The weekend has been deleted. Pause a moment before you start your own workplace odyssey and enjoy another's trip to Oopsville courtesy of Who, Me?

Today's story comes from "Jim", and concerns the time he and a colleague were performing an all-night hardware, OS, database and application upgrade of a daily newspaper's publishing system, running on a Sun/Sybase combo.

The fate of Sun Microsystems is sadly well documented, while Sybase continues to be a thing (although SAP has long since ditched the name).

Back then both were in rude health, and Jim and his pal were gainfully employed as engineers for a Sun/Sybase VAR and ISV.

The upgrade was going to plan. "Users and our team of trainers were expecting to arrive in the morning and log into a fully upgraded system," he explained. "In the middle of a critical phase of the upgrade, my buddy (he's still my buddy) suddenly got quiet – always a bad sign."

It transpired Jim's partner had been playing with Solaris's whizzy GUI file manager "and accidentally deleted the Sybase master device file while it was running."

It is difficult to describe how big a disaster this was. The loss of the primary device file would leave Sybase decidedly poorly. It is, however, an easy mistake for the unwary to make.

This hack fondly remembers a time toward the end of the last century when one particularly overconfident DBA decided to remove the chaff from a Microsoft SQL Server database directory by running del *.* "because the files it needs will be locked, right?"

More than 20 years on, still etched into my memory is his expression as Windows NT cheerfully shredded production database after production database, as ordered. And no, there were no recent backups.

For Jim, things weren't so dire. "In Solaris (and other UNIX boxes), deleting a file merely unlinks it from its directory.

"The file space isn't reclaimed as long as the file is held open by some process."

So the database would continue to work, even though a relatively major organ had been excised. However, no new processes would find the file, so a dump of the system databases wasn't an option. Nor was a graceful shutdown since the file would be closed and its bytes cast to the wind.

Jim had wisely made backups, but a recovery from them would burn through the maintenance window "and probably kill the project for another week."

And that's without considering the employment prospects once the silliness had been found out.

What to do?

Out of ideas, Jim decided to crash (rather than halt) the system by typing the BREAK sequence at the console. The server would not get the chance to close the file cleanly...

"We said a small prayer, crossed our fingers, booted the server, and waited for the file system check (fsck) to repair the damage we had done," he recalled.

"I've never typed the letter 'y' more carefully than when asked if we wanted to re-link orphaned inodes."

With an elevated heart rate, Jim logged in and checked the file system's lost+found directory.

Sure enough, there were a handful of files with integer names ("all fsck knows is the inode number, so that becomes the file name", he explained.) After a bit of investigation, he put the most likely file back in place, held his breath, and fired up the database.

"Using up all my good fortune, the database took off and we finished the upgrade.

"I wouldn't be so dramatic as to say I have PTSD from this, but retelling the story still raises the hair on the back of my neck.

"It only took 20 minutes from our maintenance window, but at least a year from my lifespan."

Ever had your bacon saved by the designers of the Unix file system? Or seen a simple task suddenly take on job-threatening proportions thanks to a co-worker's curiosity? Share your tales of near misses and near hits with an email to Who, Me? ®


So it appears some of you really don't want us to use the word 'hacker' when we really mean 'criminal'

The votes have been cast and counted... and it's a landslide

Register debate Last week, we argued over whether or not the media, including El Reg, should stop using the word hacker as a pejorative.

This debate came about after infosec pro Alyssa Miller and a few others from the Hacking Is Not A Crime movement politely asked Register vultures on Twitter to quit using the h-word as a lazy shorthand for criminal.

We said we'd think about it. And we thought about it, and we thought about it some more. And in the end, since we're writing for you, we decided to put it to the audience: we published an article for and an article against the proposal, and let everyone vote for whichever side they agreed with.

On Wednesday, Alyssa argued in favor of the media no longer using hacker as a pejorative.

Continue reading

Vodafone chief gushes over OpenRAN, says commercial deployments to start this year

But still some way to go before standards-based tech can match mainstream products

Last year Vodafone bet big on OpenRAN, announcing it would shift a huge portion of its tower estate to the standards-based tech. Now Andrew Dona, the telco's director of network and development, has shed some light on how this will work.

Speaking to Telecom TV, Dona said Vodafone had already deployed two OpenRAN sites to its production network, situated in the southwest of England. These deployments are part of its testing process, which Dona said would conclude in May.

The wide-scale macro rollout, which will replace roughly 2,600 4G masts with OpenRAN alternatives, is expected to commence later this year, winding up in 2027 in time to meet the UK government's edict to excise high-risk vendors from the telecommunications networks.

In 2019, Vodafone's then-CTO, Scott Petty, said 32 per cent of its 4G base stations used Huawei-made kit. The following year, he said Vodafone's Huawei-based 5G NSA (non-standalone) RAN equipment was "inextricably linked" to its legacy networks, which include 5G. Removing Huawei's equipment from the RAN and legacy core networks is expected to cost approximately €200 million (roughly £170 million) over a five-year period.

Continue reading

Swedish startup Logical Clocks takes a crack at scaling MySQL backend for live recommendations

Takes a 'different approach' to YouTube's Vitess to munch complex transactions in microseconds

Swedish startup Logical Clocks is launching a new key-value database as a managed service, based on the MySQL derivative MySQL NDB Cluster.

The vendor told us its RonDB can be used to provide live data to machine learning models for real-time decision-making – as commonly used in online recommendations and fraud detection.

Although it has a history going back to the late 1990s, the new open-source distribution is currently in closed beta, with interested users encouraged to apply to participate. General availability is expected in the second quarter.

Logical Clocks said the database can respond in 100-200 microseconds on individual requests, in less than a millisecond on batched read requests and perform complex transactions in a highly loaded cluster within 10 milliseconds. It can perform hundreds of millions of read or write operations per second, the company added, and apparently offers 99.9999 per cent availability – no more than 30 seconds of downtime per year.

Continue reading

Microsoft quantum lab retracts published paper: Readings that cast doubt on crucial discovery went AWOL

Quasiparticle eggheads were 'caught up in the enthusiasm of the moment'

A paper published in Nature two years ago and spearheaded by a Microsoft scientist has been retracted after it emerged that the data presented simply didn't add up.

The work was produced at a quantum computer lab set up by Microsoft and QuTech, a research center co-founded by the Delft University of Technology (TU Delft) in the Netherlands. The study, led by Microsoftie and TU Delft Professor Leo Kouwenhoven, reported the discovery of a theoretical quasiparticle the academics believed would prove useful for future quantum computers.

"A 2018 academic paper published in Nature and led by one of our scientific directors, primarily in his capacity as a Professor at TU Delft, was retracted,” Zulfi Alam, a Microsoft Quantum unit veep, told The Register on Monday.

“As part of proposing the retraction, the authors of the paper took feedback from the scientific community, re-analyzed the data, wrote a new paper based on the analysis, and embraced the paper’s examination by independent experts in the field. This is an excellent example of the scientific process at work.

Continue reading

Deploy AI workloads with confidence using OpenVINO

Write once, deploy anywhere

Sponsored Artificial Intelligence techniques have been finding their way into business applications for some time now. From chatbots forming the first line of engagement in customer services, to image recognition systems that can identify defects in products before they reach the end of the production line in a factory.

But many organisations are still stuck at where to start in building machine-learning and deep-learning models and taking them all the way from development through to deployment. Another complication is how to deploy a model onto a different system than the one that was used to train it. Especially for situations such as edge deployments, where less compute power is available than in a datacentre.

One solution to these problems is to employ OpenVINO™ (Open Visual Inference & Neural Network Optimization), a toolkit developed by Intel to speed the development of applications involving high-performance computer vision and deep-learning inferencing, among other use cases. OpenVINO takes a trained model, and optimises it to operate on a variety of Intel hardware, including CPUs, GPUs, Intel® Movidius™ Vision Processing Unit (VPU), FPGAs, or the Intel® Gaussian & Neural Accelerator (Intel® GNA).

This means that it acts like an abstraction layer between the application code and the hardware. It can also fine tune the model for the platform the customer wants to use, claims Zoë Cayetano, Product Manager for Artificial Intelligence & Deep Learning at Intel. “That's really useful when you're taking an AI application into production. There's a variety of different niche challenges in inferencing that we've tackled with OpenVINO, that are different from when models and applications are in the training phase,” she says.

Continue reading

China outlines plan to boost economy with AI, a cloud OS it controls – and bringing in skilled foreigners

Other fun bits: An 'asteroid patrol', brain:computer fusion, DNA storage, enhanced privacy laws

China has put quantum communications networks and a brain:machine interface on its to-do list in plans unveiled at its annual "Two Sessions" parliamentary sittings.

The centerpiece of the Two Sessions, which sees 5,000 of the nation's political elite gather for meetings of the National People's Congress (NPC) and top political advisory body the National Committee of the Chinese People's Political Consultative Conference (CPPCC), is discussion of a new five-year plan for the nation's development.

The 14th Five-Year Plan, a document outlining objectives from 2021 until 2025, is not allowed to be released before finalization. However a 142-page long draft in Mandarin was made legally public and select parts have been translated by Chinese journalist, Zichen Wang of state-controlled Xinhua News.

According to his translations, the five-year plan has two sections that pertain to technology.

Continue reading

Mobile World Congress seemingly serious about in-person Barcelona event in June, shares safety plan

Is Spain really ready for 50,000 people at one venue? Sounds like a super spreader event ready to happen

Mobile World Congress appears determined to run its annual Barcelona super-conference as an in-person event this year, mid-pandemic, posting a safety plan online on Monday.

The tech-fest is due to take place at the end of June, having been pushed back from its usual late February slot, giving it less than four months until doors open: a risky timeline given that the vaccination rate for Spain and the Catalan region currently stands at just under nine per cent.

But the organizers reckon that the global COVID-19 pandemic can be defeated within the walls of its conference venue with a few simple steps: social distancing, personal hygiene, event hygiene, and training staff.

Continue reading

GitHub bug briefly gave valid authenticated session cookies to wrong users

Don’t panic: Fewer than 0.001% of sessions compromised through flaw that couldn’t be maliciously triggered

If you visit GitHub today you’ll be asked to authenticate anew because the code collaboration locker has squished a bug that sometimes “misrouted a user’s session to the browser of another authenticated user, giving them the valid and authenticated session cookie for another user.”

GitHub disclosed the problem today, explain that it could only happen under “extremely rare circumstances” and “occurred in fewer than 0.001% of authenticated sessions on GitHub.com.”

The service knows which users’ sessions were exposed by the flaw and says it has contacted them with guidance and additional information.

Continue reading

Azure flings out free virtual trusted platform module for cloudy VMs

Take that, rootkits and other low-level nasties - if they take a crack at fresh VMs, on certain instance types under a handful of OSes

Microsoft has revealed that its Azure IaaS platform now offers free a virtual trusted platform module.

Dubbed “Azure Trusted Launch for virtual machines” and launched as a preview on March 8th, Microsoft’s CTO for Azure Mark Russinovich said the new offering “allows administrators to deploy virtual machines with verified and signed bootloaders, OS kernels, and a boot policy that leverages the Trusted Launch Virtual Trusted Platform Module (vTPM) to measure and attest to whether the boot was compromised.”

All of which is pretty familiar stuff on-prem, as TPM has been around for over a decade and is just-about standard issue on modern servers. Google brought virtual TPM to its cloud in mid-2018 and made it the default server configuration in April 2020.

Continue reading

Cisco issues blizzard of end-of-life notices for Nexus 3K and 7K switches

Service options decline starting next year... so there may be a Nexus 9K switch in your future

Cisco has in recent days issued a blizzard of end-of-life and end-of-sale announcement for switches in its Nexus 3000 and Nexus 7000 ranges.

By The Register’s count, the networking giant has announced that the 18 devices, listed below, across the ranges will soon be sent to the knacker's yard.

The initial batch of notices advised users that the listed devices would not be sold after late August 2021, with shipments to end in November of the same year and support services dwindling as of August 2022. November 2025 was set as the last date on which a service contract could be renewed.

However, Chipzilla has since updated a handful of the notices and extended some of the deadlines mentioned above by as much as 18 months. You can find the 3K notices here and the 7K notices here. The last day of hardware support will be sometime in 2026 or 2027, depending on the model.

Continue reading

Biting the hand that feeds IT © 1998–2021