Slashdot Mirror
Canonical Patches Two Kernel Vulnerabilities In Ubuntu 14.04 (softpedia.com)
jones_supa writes: Canonical has announced that a new kernel update is now live in the default software repositories for the Ubuntu 14.04 operating system. According to the security notice, two Linux kernel vulnerabilities have been fixed. The first security flaw was discovered in the SCTP (Stream Control Transmission Protocol) implementation, which conducted a wrong sequence of protocol-initialization steps. The second kernel vulnerability (discovered by Dmitry Vyukov) was in the Linux kernel's keyring handler, which tried to garbage collect incompletely instantiated keys. Both vulnerabilities allow a local attacker to crash the system by causing a denial of service. To fix the issues mentioned above, Canonical urges all users of Ubuntu 14.04 to update their kernel packages on all platforms.
It depends if all systems are remotely attackable, even if only http/ssh is opened, for instance. Basically if you are using SCTP, instead of TCP/UDP you may be at risk. But not so many applications are using that protocol.
Slashdot, fix the reply notifications... You won't get away with it...
Is that safe?
Safer than not, in this case :-)
Slashdot, fix the reply notifications... You won't get away with it...
I use the wily kernels with 14.04 and I haven't ran into any issues. Using patched kernels is probably safer than what I do.
It was supposed to be the successor to TCP with 1 -> N connection abilities IIRC, but to be blunt it seems to have died on its arse.
Which versions have the vulnerabilities and where are they fixed? Did Ubuntu use an old, out-of-date kernel?
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
I wouldn't be surprised to find more clients using it because it is a TCP/UDP mishmash that is good for multicasting, IPTV, and telco stuff. Of course, it will require new firewalls, since most will look at the packets, go, "is it TCP/UDP/IGMP/ICMP... if not, just drop them."
As for local attacks, I'm glad they are taken care of. Although not as show-stopping as a remote root bug, with containerization becoming mainstream, a bug that panics a kernel and drops a compute node can cause some headaches, even if the services are redundant.
Is that you, Bill?
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
I think that those who still run 14.04 are running servers. And I hardly think a lot will update and reboot.
That's an very strange assumtion. Of course server vulnerabilities are patched, and the machines rebooted if they need be. What did you expect? "Oh noes, my uptimes! I can't rebootz!"
My client is currently in the process of rolling out a new line of products based on Ubuntu 14.04 (the choice of distribution was not mine). Of course we'll be using patched kernels for new machines we build. Simply upgrading to whatever happens to be the latest version of Ubuntu this week is not an option. This has been a year in testing. The next major update is likely two to four years down the line. The previous one (which is still being shipped) is based on Ubuntu 8.04 (Hardy Heron).
May we live long and die out
I'm running 14.04 on my desktop PC. But since these vulnerabilities are both to local attackers and the worst they do is force a reboot, I'm not rushing to reboot.
This space intentionally left blank
I've never been one to whine about stories being posted here, but this one has me particularly puzzled. Is there something novel about this particular set of patches? I ask because I've seen many, many kernel updates released by Canonical to my 14.04 boxes involving potential local exploits, since 14.04 was released. Anyone know why this one warrants a story, or is it just a slow news day?
I think that those who still run 14.04 are running servers.
14.04 is the most recent LTS release, so I would imagine that many desktop users are still running that version. Hell, 12.04 is still under support.
How many users do you have on your servers?
well, my desktop PC (also running 14.04) takes about 15 seconds off its SSDs, 6 of which spent by the BIOS. But it's not definitely a server.
And which hardware are you using? Mine takes 5+ minutes just for hardware initialization.
Thanks to Zarquon I am not running virtual fluff as the reboot would downtime all virtual machines.
I run an OCS on my ones and a reboot is to be planned with a couple of months in advance with a test on a clone system.
A reboot is not a bad thing, especially if it's needed for security patches.
But within next April we'll get the new LTS. Instead of planning a reboot we are working on a transition.
Bottom line: that patch came too late in my opinion.
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.