Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Comment: Re: A Microsoft Killswitch (Score 1) 214

by Thundersnatch (#46002593) Attached to: Microsoft Remotely Deleted Tor From Windows Machines To Stop Botnet

What, exactly do you think Yum/Apt and other FOSS package systems do do? They give the same root permissions to a random package maintaine; an individual who likely would more easily be swayed by the money of organized crime or the NSA than a fairly rich and likely highly audited MSFT employee. As far as I know, there are no audits at all done of the actual binaries distributed by Linux package managers.

Comment: Re:No (Score 2) 213

by Thundersnatch (#42911575) Attached to: WebKit As Broken As Older IE Versions?

Webkit is open source, with an active community that cares about standards, has an explicit policy of trying to behave like other browsers where possible...

All evidence to the contrary. The number of "broken in latest Chrome" bug reports we've had coming out of QA recently is quite alarming. Things like certain tags not appearing in the layout at all, or massive layout gaps that don't appear in any other browser.

Personally, I think Chromium is moving too fast, and now Mozilla is following. Many of the bugs we've encountered were regressions, broken in Chome say 15, fixed in 17, and then broken again in 24.

Comment: Re:Postgresql (Score 1) 116

by Thundersnatch (#42656219) Attached to: Fedora 19 Nixing MySQL in Favor of MariaDB

The closest thing to case-insensitive collation is the citext data type. It works basically transparently as a case-insensitive replacement for varchar and text.

The major feature keeping $dayjob from using PostgreSQL over MSSQL in new development is the lack of an accent-insensitive collation. Making an index using a custom function marked IMMUTABLE that calls (lower(unaccent(text)), and then calling the same function in nearly every query, is simply too hackish to stomach.

Comment: Re:the point, exactly? (Score 1) 191

Consider also that without such pioneers as Chuck Yeager we would not have transsonic or supersonic airliners.

Umm... we don't. That 1970s french/brit thingy crashed, and that was that. It just costs to damn much to fly an airliner past Mach-1. It's like commuting 30 miles to work every day in a Bugatti Veyron at 2 mpg.

Comment: Re:Close to re-entry speed (Score 3, Informative) 191

And while they're busy doing that they often manage to put on one hell of a show:
* this effort
* the autonomous vehicle DARPA Challenge
* other random bits that we read about
* certainly other random bits we have no idea about, but I bet they're cool!
-nB

also..
* the fucking Internet

you kids these days need to learn your history

Comment: Re:Wait, hang on (Score 1) 336

by Thundersnatch (#39748307) Attached to: India Test Fires Long-Range, Nuke-Capable Missile

Yes, Israel has been at war with its neighbors, but only because it has been repeatedly attacked by those neighbors. Not that Israel plays especially nice these days - in fact they act like total dicks. I probably would act a bit dickish too if 90 million of my neighbors had vowed to wipe my people from the face of the Earth, and repeatedly attacked me with tanks and artillery to prove that they weren't just talking.

Comment: Re:Be creative but have rules (Score 2) 429

by Thundersnatch (#39287017) Attached to: Server Names For a New Generation

You just gave away your inexperience there.

You really want to touch configuration files on hundreds/thousands of machines just because a server IP has changed? How do you migrate a server to new hardware without downtime?

There's this thing called a "DNS Resolver Cache", in every OS. And another thing called a "TTL" on a DNS record. They'll save you hours of scripting work the next time you need to do a service migration.

Comment: Re:Sabu is unemployed - what a surprise (Score 1) 511

by Thundersnatch (#39261709) Attached to: LulzSec Leader Sabu Unmasked, Arrested and Caught Collaborating

You do realize that by denying people access to employment after their jail term has ended, you're leaving them only one option: Criminal activity, correct?

He can dig ditches, mow lawns, shuck corn, whatever. That's gainful employment. What he cannot do is expect to ever be put in a position of trust by his employer. That's the way it works for convicted felons - it ruins your life, even after you are out of prison. It's been that way since Greece ruled the Mediterranean, and will likely always be that way.

Comment: Re:What we need (Score 1) 209

by Thundersnatch (#39117591) Attached to: Apache 2.4 Takes Direct Aim At Nginx

Nginx 1.0+ supports backend keepalives with a patch and module, but they are still not in official release. But this code comes from the principal nginx author, so it will make it into release soon.

That said, your back-ends are usually very close network-wise to nginx proxies, and connections can be established and torn down in less than 1 ms. Since the back-ends are usually thread-based, this is a good idea anyway (which is why everybody has to turn off HTTP keepalives in Apache when they start to scale). Disabling HTTP keepalives SUCKS for the client's experience, especially if they are on wireless/mobile connections or on another continent.

I manage a medium-sized SaaS application with about 0.7M users, and we front dozens of honking physical JBoss/Tomcat boxes with a single-core linux VM running nginx with 1 GB of RAM (with a hot standby of course). Nginx is only proxying to back-ends, not serving static files (except for a small 512MB set of really hot files using proxy_cache which stays in the filesystem cache). Nginx itself uses only about 100 MB with 8 worker processes. This isn't surprising: even the biggest $50K F5 load balancers have very wimpy specifications for CPU and RAM, but like nginx they use an event-driven model to keep RAM usage and context-switching to a minimum.

One problem running nginx on Linux is that asynchronous IO on Linux is horribly broken by design, and only works for databases that use direct uncached IO. So we are looking at moving nginx to FreeBSD so we can take advantage of asynchronous disk IO as well as the default asynchronous network IO.

The one-thread/process-per-connection model of Apache really just doesn't cut it for web-scale workloads. We were able to re-purpose our dedicated Apache front-end boxes as application servers instead because of the RAM savings. So nginx saves us about $2k per month in colo costs.

Nothing happens.

Working...