Skip to content
Hallway
0:00 / 0:00

#epoll#dtrace#zfs

Ubuntu Slaughters Kittens | BSD Now 103

Jupiter Broadcasting

53K subscribers • 2 videos on Hallway

YouTube
  • Recommended by 4 people: 4 on Hacker News.
  • Presented by Bryan Cantrill, CTO of Joyent
  • An opinion. It argues one side of a question engineers disagree on.

Advanced. Deep dive for people who know the field.

39K views11 years ago4 recommendations

Systems programmingOperating systemsConcurrency and parallelismPerformance and optimization

Bryan Cantrill discusses the design flaws of Linux epoll, the history of ZFS and DTrace, and stories from his time at Sun Microsystems.

Allan's away at BSDCam this week, but we've still got an exciting episode for you. We sat down with Bryan Cantrill, CTO of Joyent, to talk about a wide variety of topics: dtrace, ZFS, pkgsrc, containers & much more. This is easily our longest interview to date! Show Notes & Download: bit.ly/bsd-103

Why it's here

4 people recommend this

What people say. Highly recommended for Bryan Cantrill's entertaining and detailed critique of Linux epoll and his insights into operating system design.

protomython Hacker News • Recommends

On “Things you can do with ZFS [video]”

Probably not, since that would have allowed someone to add GPL-only code to it which couldn't be distributed with the whole. The video is very good and you can follow-up with link to this video
View on Hacker News

myzticon Hacker News • Recommends

On “Run containers on bare metal already [video]”

Amusing Interview with Bryan Cantrill (regarding his background and he also shares some stories) link to this video Also very informative (about what happened at Sun in the 90s, dtrace, ...) Quite a funny and smart guy ;)
View on Hacker News

Gracanaon Hacker News • Recommends

On ““Node.js is one of the worst things to happen to the software industry” (2012)”

This sounds like something straight out of the Unix Hater's Handbook. :) But epoll is not the fault of Unix. Linux gets it wrong, but others do not. There's a good (and wildly entertaining) discussion/rant about it in this BSD Now episode: link to this video
View on Hacker News

beefhashon Hacker News • Recommends

On “Talks I have given”

I also found the somewhat more casual BSDnow podcasts that featured Mr. Cantrill quite insightful1,2,3. He shows up around 15 or 30 minutes in. 1 link to this video 2 link 3 link
View on Hacker News

myzticon Hacker News • Mentions

On “No POST after rm -rf /”

…you don't type it out by accident, but shell scripts with unset variables might do it. And regarding Poettering's response2 (not trying to start a fight): It's Poettering, what do you expect? You can hate or love systemd, but part of why people hate it is his intellectual arrogance in everything he does. 1 He tells the story somewhere in here link to this video (quite entertaining) 2 link…
View on Hacker News

protomython Hacker News • Mentions

On “Linux 4.5-rc5: efivarfs fixed to avoid “rm -rf /” bricking UEFI”

For the few use cases where a system admin wants to "rm -rf /", there are hundreds of bad scripts that can screw up a system. I believe Solaris did the right thing and made it not work. link to this video
View on Hacker News

bcantrillon Hacker News • Mentions

On “Microsoft and Canonical partner to bring Ubuntu to Windows 10”

…ilestone.) Assuming that runs, my next questions are all common apps/programs that we knew to be especially thorny for one reason or another: Go, strace, tcpdump, systemd, etc. Anyway, it's great to see the SmartOS approach so broadly validated by Microsoft -- and hope that they both finish the job and open source it all! 1 link 2 link 3 link to this video 4 link…
View on Hacker News

gizmoon Hacker News • Mentions

On “Epoll is fundamentally broken”

The linked interview with Bryan Cantrill gives a good explanation of why it's broken: a straightforward implementation behaves in a way you never want, with bizarre behavior where the kernel can wake up a thread with an accept() on a fd that has already been closed. linux should just have adopted kqueue or an IOCP model instead. link to this video
View on Hacker News

Annataron Hacker News • Mentions

On “LinuxBoot: Linux as Firmware”

link to this video - watch it to the end. How long "people have been doing it" has nothing to do with the quality, it's a fallacy. They can't code if they can't even get the basics like polling or startup / shutdown correctly.
View on Hacker News

protomython Hacker News • Mentions

On “Maintaining 65k open connections in a single Ruby process”

…riptor will continue to be delivered to the parent -- despite the fact that the parent itself no longer has any notion of the file description! This epoll facility refuses to honor these semantics; closing the EPOLL_CTL_ADD'd file descriptor will always result in no further events being generated for that event description. Brian Cantrill on epoll link to this video…
View on Hacker News

eqvinoxon Hacker News • Mentions

On “The Demikernel Datapath OS Architecture”

Well, not a question. POSIX and Linux are not the same. There is no epoll in POSIX, it's a Linux addition. The BSDs have kqueue which according to some people is actually better cf. [video ]. But kqueue isn't POSIX either. Only poll() and select() are...
View on Hacker News

eductionon Hacker News • Mentions

On “Way too many ways to wait on a child process with a timeout”

He mentions Bryan Cantrill in there and I can’t resist posting his famous epoll/kqueue rant: link to this video I know this is related but maybe someone smarter than me can explain how closely it relates (or doesn’t) to this issue which seems more general (iirc Cantrill was talking about fs events not child processes generally)
View on Hacker News