HNHacker News
TopNewBestAskShowJobs

mwpmaybe

907 karma · joined July 27, 2015

[ my public key: https://keybase.io/mwp; my proof: https://keybase.io/mwp/sigs/IgRJbBXIMpIG5ILaErmbmjkyttiEoe4ov_JqABTpuoQ ]
submissionscomments
mwpmaybe··on Do we really need swap on modern systems?
My personal rules of thumb for Linux systems. YMMV.

* If you need a low-latency server or workstation and all of your processes are killable (i.e. they can be easily/automatically restarted without data loss): disable swap.

* If you need a low-latency server or workstation and some of your processes are not killable (e.g. databases): enable swap and set vm.swappiness to 0.

* SSD-backed desktops and other servers and workstations: enable swap and set vm.swappiness to 1 (for NAND flash longevity).

* Disk-backed desktops and other servers and workstations: accept the system/distro defaults, typically swap enabled with vm.swappiness set to 60. You can and likely should lower vm.swappiness to 10 or so if you have a ton of RAM relative to your workload.

* If your server or workstation has a mix of killable and non-killable processes, use oom_score_adj to protect the non-killable processes.

* Monitor systems for swap (page-out) activity.

mwpmaybe··on Do we really need swap on modern systems?
If you have an SSD, try setting vm.swappiness to 1 (not 0).
mwpmaybe··on Docker in Production: An Update
Were you using Device Mapper purposefully, or just because you couldn't get a more typical storage driver to work? The general setup instructions recommend either 1. installing linux-image-extra and using AUFS (for kernel versions < 4) or 2. modifying the dockerd invocation to specify the OverlayFS storage driver with `-s overlay` (for kernel versions >= 4).

Admittedly, the latter strategy is a bit frustrating and counterintuitive on 16.04. When you install Docker it will start automatically and hang because it can't find the AUFS driver and Device Mapper isn't set up. The solution is to either 1. modify policy-rc.d to prevent services from automatically starting or 2. set up Device Mapper (install dmsetup and run `dmsetup mknodes`) before installing Docker and changing the storage driver. Unfortunately, these workarounds are not particularly well-documented.

I would be interested to see how Docker 1.12 or 1.13 with a modern kernel and OverlayFS would be able to handle your described workload.

mwpmaybe··on Docker in Production: An Update
I have (or had, I guess) an 11 year career in Fortune ~100 finance and healthcare. Agreed on all points. However, they're already experimenting with Debian, Ubuntu LTS (just not 16.04), kernels, custom AMIs, and different Docker versions. It's not a stretch to imagine they have the ability to deploy a back-ported kernel, or build and/or use an AMI that includes one. Also, HFT practices (or lack thereof) are far from the staid enterprise practices you may be thinking of.
mwpmaybe··on Docker in Production: An Update
And a quick run through the previous article[0]:

> Docker Issue: Breaking changes and regressions

Yes, it's new, and it changes regularly. This is why you do what you do with any other piece of software in an enterprise environment: Pick a stable release (e.g. 1.7), deploy it, then upgrade your development environment to the next stable release (e.g. 1.11) and work through the breaking changes. Regressions are frustrating, I'll agree, but it's free software: report the regressions and help the community fix them.

> Docker Issue: Can’t clean old images

A built-in feature to do so was added in 1.13 (a few months after this article was published).

> The only way to clean space is to run this hack, preferably in cron every day: docker images -q -a | xargs --no-run-if-empty docker rmi

That's not a hack, that's how you do things in Unix.

> As a long-standing goal, the AUFS filesystem was finally dropped in kernel version 4.

> There is no unofficial patch to support it, there is no optional module, there is no backport whatsoever, nothing. AUFS is entirely gone.

While the first point is technically true (actually, based on some light googling, I'm not sure it was ever merged in the first place), many distributions provide it as an optional kernel module. For example, Ubuntu provides it in linux-image-extra. Yes, Virginia, you can build kernel modules from source.

> How does docker work without AUFS then? Well, it doesn’t.

It does: Btrfs, Device Mapper, ZFS...

> So, the docker guys wrote a new filesystem, called overlay. [..] Note that it’s not backported to existing distributions. Docker never cared about [backward] compatibility.

Docker supports multiple storage drivers, one benefit of which is to be able to support older systems: AUFS on older distributions, OverlayFS on newer. The container abstraction allows you to not care about the underlying storage subsystem.

> Right now. We don’t know of ANY combination that is stable

In November 2016? Are you kidding?

0. https://thehftguy.com/2016/11/01/docker-in-production-an-his...

mwpmaybe··on Docker in Production: An Update
I am neither a Docker expert nor evangelist, and I have my own gripes and frustrations with it, but this article is full of misinformation and FUD. To wit:

> CoreOS is an operating [system] that can only run Docker and is exclusively intended to run Docker.

No, it isn't. (Maybe if you substitute "Docker" with "containers.")

> First, the main benefit of Docker is to unify dev and production. Having a separate OS in production only for containers totally ruins this point.

No, it doesn't. Completely the opposite, in fact: Docker makes it possible to not care about OS differences between dev, test, and prod.

> Docker on Debian is major no-go

> Docker is 100% guaranteed suicide on Debian 8 and it’s been since the inception of Docker a few years ago

> Debian froze the kernel to a version that doesn’t support anything Docker needs and the few components that are present are rigged with bugs.

rubs temples It's Linux. It's Debian. You can run any kernel you want. There are plenty of repositories out there with binary kernels, including Debian's own back-ported 4.9 kernel, or you can build one from source. Because, you know, open source. I've run many instances of Jessie on Linode with their 4.8 and 4.9 kernels and it's no problem. Hundreds of companies are running Docker in production on Debian and Debian-derived systems.

> I am not aware of any serious companies than run on Ubuntu.

Just because you are not aware does not mean they do not exist. How about Netflix, Snapchat, Dropbox, Uber, and Tesla for starters?

> I cannot comment on the LTS 16 as I do not use it.

It's been out since last April. If you're serious about this survey, this is a no-brainer.

> I received quite a few comments and unfriendly emails of people saying to “just” use the latest Ubuntu beta

What? No. Just use the latest LTS.

> I am moderately confident that there is no-one on the planet using Docker seriously AND successfully AND without major hassle.

LOL

mwpmaybe··on Latest DevOp Blogs, Boards and Books to Check Out Today
I found this[0] one and this[1] one. Not sure which one the GP is referring to.

0. https://devopsengineers.com

1. https://devopschat.co

mwpmaybe··on Ubuntu 16.04.2 LTS released
Are you sure that was specific to Ubuntu and not general to Debian?
mwpmaybe··on Ask HN: I’ve been applying to a million jobs
Check out the "Who's hiring?" [0] thread and consider posting in the "Who wants to be hired?" [1] and "Freelancing" [2] threads (there's one every month).

I remember feeling like you do when I was young and thought I knew everything. As others have mentioned in this thread, starting your own company or building a freelance practice might be the best strategy. Quite frankly, if you're so great, you should work for yourself. If you're not, you probably (and unfortunately) need to "pay your dues" and put in two or three years as a junior developer (this is key) at an established company. I'm looking at your web site and can't easily figure out whether or not you've already done that, so that's not a good sign. You'll build skills, a network, salary and job title history, and most importantly, credibility.

Also, have you considered going to university? A lot of companies in the US won't hire you if you don't have an undergraduate degree (or 10+ years of corporate experience). I see "Foundation in Applied Arts" on your web site but it's not clear what if anything that translates to.

FWIW, I did the CS degree and dues-paying thing and most days wish I'd started a company instead.

0. https://news.ycombinator.com/item?id=13541679

1. https://news.ycombinator.com/item?id=13541681

2. https://news.ycombinator.com/item?id=13541680

mwpmaybe··on High-Memory Instances and $5 Linodes
Yeah, that's... not practical.

What if they upgrade the 1024 plan to 1536? (That has actually happened, by the way.) Now you're extracting the number and comparing it to a range?

> Give me the plan where the RAM is >= 1024 and < 2048.

What if they upgrade the 1536 plan to 2048?

> Nevermind, just give me the plan with two cores.

Okay, do you want the $20/mo plan or the $120/mo plan?

> Fine, sort the plans by cores and memory and give me the smallest plan.

Oops, they introduced a new plan with one core and less (i.e. not enough) storage.

I'm not saying it's impossible. Just impractical.

mwpmaybe··on High-Memory Instances and $5 Linodes
That's interesting. Thank you.

I realize that I am doing that thing where you try to rationalize your preconceived notions, but here are some high-level observations:

* Looks like there are normally setup fees associated with provisioning new servers, although they are suspended at the moment.

* Support seems next to non-existant, which I suppose is not surprising considering they're a low-cost provider.

* SSD-based servers appear to be frequently out-of-stock.

* If you need a KVM attached to your server, it is $30 for 24 hours or $200+ for a week.

* 250Mbps bandwidth (presumably in and out) cf. Linode which is 40G in and 1-10G out.

All that being said, my interest is piqued. I could see this being a good fit if 1. you have a more-or-less dedicated sysadmin, 2. need a lot of storage and/or memory and have solid sizing requirements ahead of time, and 3. cloud (VPS or otherwise) isn't an option due to cost or other facts. It could be great for running your own VM or container cluster. Thanks again for sharing!

mwpmaybe··on High-Memory Instances and $5 Linodes
Just being able to programmatically provision a server at a moment's notice—even in normal day-to-day (i.e. non-catastrophic) situations—is a critical tool in my toolbelt.

You can take a snapshot of a VM, you can move it to a different host, you can spin up a new instance to test a major upgrade, you can clone a cluster for a few hours of testing, you can scale up and down...

mwpmaybe··on High-Memory Instances and $5 Linodes
And one server will only ever be one server. For many deployments it makes more sense to have several smaller nodes distributed between different geographical areas than one big node in one datacenter. And/or multiple nodes in a a single datacenter so you can perform rolling upgrades.
mwpmaybe··on High-Memory Instances and $5 Linodes
What is your definition of a "root server" and where do you rent them? Do you mean a dedicated server? I've checked a few providers and none of them come close the costs/benefits you're listing. OVH's EG-16 (4c/8t, 16G, 2x4T) is $79/mo. 1-and-1's L4i (4c/4t, 12G, 2x1T) is $80/mo.
mwpmaybe··on High-Memory Instances and $5 Linodes
Well, they've changed in the past... it's fine as long as the plans corresponding to the PlanIDs never get smaller or more expensive. This particular change violates that assumption and so is definitely not fine.
mwpmaybe··on High-Memory Instances and $5 Linodes
PSA: If you're using the Linode API, the 2048 plan is now PlanID 2 and the new 1024 plan is now PlanID 1. The other PlanIDs were incremented by one as well. I wish they hadn't done that, but there it is.
mwpmaybe··on High-Memory Instances and $5 Linodes
I think it makes good business sense. It comes less than a year after they bumped the $10/mo instance to 2 GB RAM. They must have realized they were losing market share just by dint of there being no $5/mo option to compete with DO. Even though this is the bottom of the market, as you say, some of those "one-and-done" projects must hang around and need to scale eventually, and it's clearly easier to scale up on an existing platform than migrate to another. I know this takes away one of the two or three big reasons I ever have for using DO instead of Linode!
mwpmaybe··on High-Memory Instances and $5 Linodes
Those are legacy options and to the best of my understanding are not available on new accounts or after switching legacy accounts to hourly billing.
mwpmaybe··on Is PostgreSQL good enough?
http://rachbelaid.com/postgres-full-text-search-is-good-enou...
mwpmaybe··on Ask HN: If you could move anywhere for a tech job, where would you go?
It's made affordable for most people because it's heavily subsidized by the primary's employer or the government. Without a subsidy it can cost anywhere from a few hundred to a few thousand per month.
mwpmaybe··on Ask HN: If you could move anywhere for a tech job, where would you go?
You've been saying company, and the GP set the context of company, but okay, fine. Big organization: company, not-for-profit, union, government, etc.
mwpmaybe··on Ask HN: If you could move anywhere for a tech job, where would you go?
You're still assuming a big company. Imagine you work for a startup and have to purchase insurance on an exchange.
mwpmaybe··on Ask HN: If you could move anywhere for a tech job, where would you go?
An unsubsidized silver plan (i.e. on an exchange) for a family of four can run into the thousands.
mwpmaybe··on Ask HN: If you could move anywhere for a tech job, where would you go?
If it's a big company, and if it's just you...
mwpmaybe··on Zelda: Link's Awakening: Kennel Glitch
Oh good grief, that happened to me by accident once and I had no idea what was going on. I think I had to restart the game! Thank you for clearing up this mystery for eight-year-old me.
mwpmaybe··on In Support of Free Speech
You are walking around one day when you overhear someone accosting someone else. You run over to witness the accoster shouting vicious racial, religious, and/or sexual epitaphs at the accosted. You interrupt them and ask what's going on. It's clear the two don't know each other, and the accosted, who seems in shock, says he/she was just going about his/her day when this person noticed something about his/her appearance and started screaming. The accoster does not dispute this, and indeed, as soon as you've stopped asking questions, resumes accosting.

If you respond to this situation by any action (other than smiling and walking away or joining in the accosting), you are demonstrating intolerance for the accoster. Even simply trying to reason with him/her about his/her beliefs demonstrates some level of intolerance. But you feel sympathy for the accosted, you don't share the views of the accoster, and you want to help in some way. This is known as the paradox of intolerance and it's a challenging philosophical issue.

The argument against Shopify that is being made here is that the behavior and actions of Breitbart and similar groups and publications bring about an atmosphere that leads to intolerance towards minorities. Or, if you prefer something more concrete, a store in Indiana refusing to sell a wedding cake to a gay couple demonstrates intolerance towards homosexuals. What can we do to help or make them stop being intolerant? How can we respond to this intolerance without demonstrating intolerance ourselves? Or is intolerance of intolerance a moral obligation?

mwpmaybe··on In Support of Free Speech
These are neo-nazis buying products from a neo-nazi publication that happens to use Shopify for their online store. They're not going to opt-in to an ACLU donation. What the GGP is proposing is that Shopify take their cut of the transaction and donate it to the ACLU.
mwpmaybe··on Why Databases Are Not for Docker Containers
I am still on the journey to wholesale container acceptance, but I have been finding more and more use-cases that are delightfully solved by them. My favorite so far is a WordPress hosting platform with some shared infrastructure (web server, caching reverse proxy, and database) but each PHP-FPM instance jailed in its own container. This lets me:

* easily chroot PHP (this is surprisingly difficult otherwise)

* restrict MariaDB access by IP address

* constrain the resource consumption of each application as necessary (i.e. to prevent an out-of-control PHP script from swamping the box)

* independently determine each application's PHP version

And because each managed application is (basically) a Docker image and a Caddyfile, it's easily extensible to non-PHP things. I can feel the lightbulb flickering but I'm not yet at full k8s awareness. The shared infrastructure isn't containerized, but it could easily be, and it's all running on one VM, but it could be distributed across multiple.

Containers don't solve the common problems, they just give you more tools to work with. With databases, for example, you still need to figure out whether each application gets its own database instance? schema? user?, a replication strategy, a failover strategy, a backup strategy, etc. You can use either a bind-mounted host directory or a shared-storage volume for the backing store, just like always, or a newfangled data volume container.

I am more comfortable sharing a database instance between multiple schemas and users because I can do IP-specific grants, but if I wanted to do one per application, I could do that too!

mwpmaybe··on Speed Up a WordPress Website in 8 Steps
No, not for s-maxage seconds.
mwpmaybe··on Trump Fires Acting Attorney General
You are clearly 10x smarter than I am, but I wish you would use your powers for good.

Yes, it was clever of then to use the list, because it gives them the cover to do whatever they want to do, knowing full well that people like you and Andrew McCarthy of the National Review would defend their actions using this argument. You must be able see the difference between what the Obama administration did in 2015 and the Trump administration is attempting to do today. The fact that the two orders involve the same seven countries in both cases is irrelevant.

← PreviousPage 6 of 12Next →