Security updates have been issued by Arch Linux (curl and zathura-pdf-mupdf), Debian (libmad and vlc), openSUSE (enigmail), Red Hat (collectd, Red Hat OpenStack Platform director, and sensu), and SUSE (firefox, ghostscript, and mysql).
Robert Haas writesabout the sharding capabilities that PostgreSQL will someday have."The capabilities already added are independently useful, but Ibelieve that some time in the next few years we're going to reach a tippingpoint. Indeed, I think in a certain sense we already have. Just a few yearsago, there was serious debate about whether PostgreSQL would ever havebuilt-in sharding. Today, the question is about exactly which features arestill needed."
"Security is hard" is a tautology, especially in the fast-moving worldof container orchestration. We have previously covered various aspects ofLinux containersecurity through, for example, the Clear Containers implementationor the broader question of Kubernetes andsecurity, but those are mostly concerned with container isolation; they do not address thequestion of trusting a container's contents. What is a container running?Who built it and when? Even assuming we have good programmers and solidisolation layers, propagating that good code around a Kubernetes clusterand making strong assertions on the integrity of that supply chain is farfrom trivial. The 2018 KubeCon+ CloudNativeCon Europe event featured some projects that couldeventually solve that problem.
Security updates have been issued by Arch Linux (runc), Debian (curl), Fedora (xdg-utils), Mageia (firefox), openSUSE (libreoffice, librsvg, and php5), Slackware (curl and php), SUSE (curl, firefox, kernel, kvm, libapr1, libvorbis, and memcached), and Ubuntu (curl, dpdk, php5, and qemu).
In a rather short session at the 2018 Python Language Summit, LarryHastings updated attendees on the status of his Gilectomy project. The aim of that effort isto remove the global interpreter lock (GIL) from CPython. Since his status report at last year's summit, littlehas happened, which is part of why the session was so short. He hasn'tgiven up on the overall idea, but it needs a new approach.
In a filesystem track session at the 2018 Linux Storage, Filesystem, andMemory-Management Summit (LSFMM), Darrick Wong talked about the onlinescrubbing and repair features he has been working on. His target has mostly beenXFS, but he has concurrently been working on scrubbing for ext4.Part of what he wanted to discuss was the possibility of standardizing someof these interfaces across different filesystem types.
At KubeCon+ CloudNativeCon Europe 2018, several talks explored the topic ofcontainer isolation and security. The last year saw the release of Kata Containers which, combined withthe CRI-O project, provided strong isolationguarantees for containers using a hypervisor. During the conference, Googlereleased its own hypervisor called gVisor, adding yet anotherpossible solution for this problem. Those new developments prompted thecommunity to work on integrating the concept of "secure containers"(or "sandboxed containers") deeperinto Kubernetes. This work is now coming to fruition; it prompts us to lookagain at how Kubernetes tries to keep the bad guys from wreaking havoc oncethey break into a container.
At the 2018 Python Language Summit, Carl Shapiro described some ofthe experiments that he and others at Instagram did to look at ways toimprove the performance of the CPython interpreter.The talk was somewhat academic in tone and built on what has been learnedin other dynamic languages over the years. By modifying the Python objectmodel fairly substantially, they were able to roughly double the performanceof the "classic" Richards benchmark.
In a combined filesystem and storage session at the 2018 Linux Storage,Filesystem, and Memory-Management Summit (LSFMM), Tim Walker asked for helpin designing the interface to some new storage hardware. He wanted somefeedback on how a multi-actuatordrive should present itself to the system. These drives have two (or, eventually, more) sets of read/write heads andother hardware that can all operate in parallel.
Eric Snow kicked off the 2018 edition ofthe Python Language Summit with a look at getting a better story formulticore Python by way of subinterpreters. Back in 2015, we looked at his efforts at that point; thingshave been progressing since. There is more to do, of course, so he ishoping to attract more developers to work on the project.This is the start of the Python Language Summit coverage for this year; articles are being collected on a dedicated summit page as they are finished.
Here's aposting from Canonical concerning the cryptocurrency-mining app thatwas discovered in its Snap Store. "Several years ago when we startedthe work on snap packages, we understood that we could not instantlyimplement an alternative that was completely safe from all perspectives. Inaddition to being safe, it had to be useful. So the challenge we gaveourselves was to significantly improve the situation immediately, and thenpave the road for incremental improvements that could be rolled outgradually."
Security updates have been issued by Arch Linux (firefox, llpp, and webkit2gtk), Debian (kwallet-pam), Fedora (kernel and pam-kwallet), Gentoo (mpv), Oracle (389-ds-base, firefox, libvirt, and qemu-kvm), and Ubuntu (php5 and php5, php7.0, php7.1, php7.2).
Security updates have been issued by Debian (tiff and tiff3), Fedora (glusterfs, kernel, libgxps, LibRaw, postgresql, seamonkey, webkit2gtk3, wget, and xen), Mageia (afflib, flash-player-plugin, imagemagick, qpdf, and transmission), openSUSE (Chromium, opencv, and xen), SUSE (kernel), and Ubuntu (firefox).
Technologies like containers, clusters, and Kubernetes offer the prospectof rapidly scaling the available computing resources to match variable demandsplaced on the system. Actually implementing that scaling can be achallenge, though.During KubeCon+ CloudNativeCon Europe 2018, Frederic Branczyk from CoreOS (nowpart of Red Hat) held a packed sessionto introduce a standard and officially recommended way to scale workloadsautomatically in Kubernetesclusters.
The efail.de site describes a set ofvulnerabilities in the implementation of PGP and MIME that can cause thedisclosure of encrypted communications, including old messages. "In anutshell, EFAIL abuses active content of HTML emails, for exampleexternally loaded images or styles, to exfiltrate plaintext throughrequested URLs."The EFF recommendsuninstalling email-encryption tools that automaticallydecrypt email entirely. "Until the flawsdescribed in the paper are more widely understood and fixed, users shouldarrange for the use of alternative end-to-end secure channels, such asSignal, and temporarily stop sending and especially reading PGP-encryptedemail."
The 4.17-rc5 kernel prepatch has beenreleased. "So I think we're in pretty good shape. Please go keeptesting, though, to make sure we're not missing anything."
Gian-Carlo Pascutto postsabout the sandboxing improvements in the Firefox 60 release."The most important change is that content processes — which renderWeb pages and execute JavaScript — are no longer allowed to directlyconnect to the Internet, or connect to most local services accessed withUnix-domain sockets (for example, PulseAudio)."
Much has been written on LWN about dynamically instrumenting kernelcode. These features are also available to user-space code with aspecial kind of probe known as a User Statically-Defined Tracing(USDT) probe. These probes provide a low-overhead way of instrumenting user-space code and provide a convenient way to debug applicationsrunning in production. In this final article of the BPF and BCC serieswe'll look at where USDT probes come from and how you can use them tounderstand the behavior of your own applications.
Security updates have been issued by Arch Linux (libmupdf, mupdf, mupdf-gl, and mupdf-tools), Debian (firebird2.5, firefox-esr, and wget), Fedora (ckeditor, drupal7, firefox, kubernetes, papi, perl-Dancer2, and quassel), openSUSE (cairo, firefox, ImageMagick, libapr1, nodejs6, php7, and tiff), Red Hat (qemu-kvm-rhev), Slackware (mariadb), SUSE (xen), and Ubuntu (openjdk-8).
The Rust team has announcedthe release of version 1.26.0 of the Rust programming language. "The past few releases have had a steady stream of relatively minor additions. We’ve been working on a lot of stuff, however, and it’s all starting to land in stable. 1.26 is possibly the most feature-packed release since Rust 1.0."
I was sure that somewhere there must bephysically-lightweight sensors with simple power, simple networking, anda lightweight protocol that allowed them to squirt their data down thenetwork with a minimum of overhead. So my interest was piqued when Jan-Piet Mens spoke at FLOSSUK's Spring Conference on "Small Things for Monitoring". Once he started passingworking demonstration systems around the room without interrupting thedemonstration, it was clear that MQTT was what I'd been looking for.
Security updates have been issued by Arch Linux (freetype2, libraw, and powerdns), CentOS (389-ds-base and kernel), Debian (php5, prosody, and wavpack), Fedora (ckeditor, fftw, flac, knot-resolver, patch, perl, and perl-Dancer2), Mageia (cups, flac, graphicsmagick, libcdio, libid3tag, and nextcloud), openSUSE (apache2), Oracle (389-ds-base and kernel), Red Hat (389-ds-base and flash-plugin), Scientific Linux (389-ds-base), Slackware (firefox and wget), SUSE (xen), and Ubuntu (wget).
For those who are curious about the rather complex way in which X serverpointer acceleration works, Peter Hutterer has put together a four-partseries on the topic:part 1,part 2,part 3,andpart 4."The input for the acceleration profile is a speed in mickeys, a threshold (in mickeys) and a max accel factor (unitless). Mickeys are a bit tricky. This means the acceleration is device-specific, the deltas for a mouse at 1000 dpi are 20% larger than the deltas for a mouse at 800 dpi (assuming same physical distance and speed)".
The CoreOS blog is carrying anarticle describing the path forward now that CoreOS is owned by RedHat. "Since Red Hat’s acquisition of CoreOS was announced, wereceived questions on the fate of Container Linux. CoreOS’s first project,and initially its namesake, pioneered the lightweight, 'over-the-air'automatically updated container native operating system that fast rose inpopularity running the world’s containers. With the acquisition, ContainerLinux will be reborn as Red Hat CoreOS, a new entry into the Red Hatecosystem. Red Hat CoreOS will be based on Fedora and Red Hat EnterpriseLinux sources and is expected to ultimately supersede Atomic Host as RedHat’s immutable, container-centric operating system." Someinformation can also be found in thisRed Hat press release.
The amount of available data is growing larger these days, to the pointthat some data sets are far larger than any one company or organization can create and maintain. So companies andothers want to share data in ways that are similar to how they share code. Some of thosecompanies are members of the Linux Foundation (LF), which is part of why thatorganization got involved in the process of creating licenses for thisdata. LF VP of Strategic Programs Mike Dolan came to the 2018 Legal andLicensing Workshop (LLW) to describe how the Community Data LicenseAgreement (CDLA) came about.
Mozilla has released Firefox 60. From the releasenotes: "Firefox 60 offers something for everyone and a littlesomething extra for everyone who deploys Firefox in an enterprise environment. This release includes changes that give you more content and more ways to customize your New Tab/Firefox Home. It also introduces support for the Web Authentication API, which means you can log in to websites in Firefox with USB tokens like YubiKey.Firefox 60 also brings a new policy engine and Group Policy support forenterprise deployments. For more info about why and how to use Firefox inthe enterprise, see this blog post."
Security updates have been issued by Debian (kernel), Gentoo (rsync), openSUSE (Chromium), Oracle (kernel), Red Hat (kernel and kernel-rt), Scientific Linux (kernel), SUSE (kernel and php7), and Ubuntu (dpdk, libraw, linux, linux-lts-trusty, linux-snapdragon, and webkit2gtk).
Version 1.14 of theBattle for Wesnoth role-playing strategy game — the first release in over threeyears — is available. "Along with the long-awaited debut on Steam,this new release series brings forth a vast number of additions and changesin all areas: a new single-player campaign, a visual and functional refreshof the multiplayer lobby and add-ons manager, a refurbished display engine,new unit graphics and animations, and much more."
<p>In a plenary session on the second day of the Linux Storage, Filesystem,and Memory-Management Summit (LSFMM), Dave Chinner described his ideas fora virtual block address-space layer. It would allow "space accounting to beshared and managed at various layers in the storage stack". One of thetargets for this work is for filesystems on thin-provisioned devices, wherethe filesystem is larger than the storage devices holding it (and administrators areexpected to add storage as needed); in current systems, running out ofspace causes huge problems for filesystems and users because the filesystemcannot communicate that error in a usable fashion.
The removal of an old joke from the GNU C Library manual might not seemlike the sort of topic that would inspire a heated debate. At times,though, a small action can serve as an inadvertent proxy for a moresignificant question, one which is relevant to both the developers and theusers of the project. In this case, that question would be: how is the project governed and whomakes decisions about which patches are applied?
At the 2018 Linux Storage, Filesystem, and Memory-Management Summit(LSFMM), Allison Henderson led a session to discuss an XFS feature she has beenworking on: parent pointers. These wouldbe pointers stored in extended attributes (xattrs) that would allow various tools toreconstruct the path for a file from its inode.In XFS repair scenarios, that path will help with reconstruction as well asprovide users with better information about where the problems lie.
Security updates have been issued by Debian (libdatetime-timezone-perl, libmad, lucene-solr, tzdata, and wordpress), Fedora (drupal7, scummvm, scummvm-tools, and zsh), Mageia (boost, ghostscript, gsoap, java-1.8.0-openjdk, links, and php), openSUSE (pam_kwallet), and Slackware (python).
The 4.17-rc4 kernel prepatch is out."Two thirds of the 4.17-rc4 patch is drivers, which sounds about right.Media, networking, rdma, input, nvme, usb. A little bit of everything, inother words." The codename has been changed, for the first timesince 4.10, to "Merciless Moray".
The mount()system call suffers from a number of different shortcomings that has ledsome to consider a different API. At last year's Linux Storage,Filesystem, and Memory-Management Summit (LSFMM), that someone wasMiklos Szeredi, who led a session to discuss hisideas for a new filesystem mounting API. Since then, David Howells has beenworking with Szeredi and VFS maintainer Al Viro on this API; at the 2018LSFMM, he presented that work.
The DMA zone (ZONE_DMA) is a memory-management holdover from thedistant past. Once upon a time, many devices (those on the ISA bus inparticular) could only use 24 bits for DMA addresses, and were thuslimited to the bottom 16MB of memory. Such devices are hard to find oncontemporary computers. Luis Rodriguez scheduled the lastmemory-management-track session of the 2018 Linux Storage, Filesystem, andMemory-Management Summit to discuss whether the time has come to removeZONE_DMA altogether.
A system's page tables are organized into a tree that is as many as fivelevels deep. In many ways those levels are all similar, but the kerneltreats them all as being different, with the result that page-tablemanipulations include a fair amount of repetitive code. During thememory-management track of the 2018 Linux Storage, Filesystem, andMemory-Management Summit, Kirill Shutemov proposed reworking how pagetables are maintained. The idea was popular, but the implementation islikely to be tricky.
Security updates have been issued by Debian (jackson-databind, quassel, and redmine), Fedora (community-mysql and php), Red Hat (chromium-browser), Scientific Linux (java-1.7.0-openjdk), and Slackware (seamonkey).
At a plenary session heldrelatively early during the 2018 Linux Storage, Filesystem, andMemory-Management Summit, the developers discussed a number of problemswith the kernel's get_user_pages() interface. During the waninghours of LSFMM, a tired (but dedicated) set of developers convened again inthe memory-management track tocontinue the discussion and try to push it toward a real solution.
<p>Chris Mason and Josef Bacik led a brief discussion on the block-I/Ocontroller for control groups (cgroups) in the filesystem track at the 2018 LinuxStorage, Filesystem, and Memory-Management Summit. Mostly they were justaiming to get feedback on the approach they have taken. They are trying toaddress the needs of their employer, Facebook, with regard to the latencyof I/O operations.
Memory hotplugging is one of the least-loved areas of the memory-managementsubsystem; there are many use cases for it, but nobody has taken ownershipof it. A similar situation exists for hardware pagepoisoning, a somewhat neglected mechanism for dealing with memory errors.At the 2018 Linux Storage, Filesystem, and Memory-Management summit, MichalHocko and Mike Kravetz dedicated a pair of brief memory-management tracksessions to problems that have been encountered in these subsystems, one ofwhich seems more likely to get the attention it needs than the other.
The memory-management subsystem is a central point that handles all of thesystem's memory, so it is naturally subject to scalability problems assystems grow larger. Two sessions during the memory-management track ofthe 2018 Linux Storage, Filesystem, and Memory-Management Summit looked atspecific contention points: the zone locks and the mmap_semsemaphore.
Security updates have been issued by CentOS (firefox, java-1.7.0-openjdk, java-1.8.0-openjdk, librelp, patch, and python-paramiko), Debian (kernel and quassel), Gentoo (chromium, hesiod, and python), openSUSE (corosync, dovecot22, libraw, patch, and squid), Oracle (java-1.7.0-openjdk), Red Hat (go-toolset-7 and go-toolset-7-golang, java-1.7.0-openjdk, and rh-php70-php), and SUSE (corosync and patch).
LWN.net