Slashdot's Setup, Part 1- Hardware

As part of our 10-Year anniversary coverage, we intend to update our insanely dated FAQ entry that describes our system setup. Today is Part 1 where we talk mostly about the hardware that powers Slashdot. Next week we'll run Part 2 where we'll talk mostly about Software. Read on to learn about our routers, our databases, our webservers and more. And as a reminder, don't forget to bid on our charity auction for the EFF and if you are in Ann Arbor, our anniversary party is tomorrow night.

CT:Most of the following was written by Uriah Welcome, famed sysadmin extraordinaire, responsible for our corporate intertubes. He Writes...

Many of you have asked about the infrastructure that supports your favorite time sink... err news site. The question even reached the top ten questions to ask CmdrTaco. So I've been asked to share our secrets on how we keep the site up and running, as well as a look towards the future of Slashdot's infrastructure. Please keep in mind that this infrastructure not only runs Slashdot, but also all the other sites owned by SourceForge, Inc.: SourceForge.net, Thinkgeek.com, Freshmeat.net, Linux.com, Newsforge.com, et al.

Well, let's begin with the most boring and basic details. We're hosted at a Savvis data center in the Bay Area. Our data center is pretty much like every other one. Raised floors, UPSs, giant diesel generators, 24x7 security, man traps, the works. Really, once you've seen one class A data center, you've seen them all. (CT: I've still never seen one. And they won't let us take pictures. Boo savvis.)

Next, our bandwidth and network. We currently have two Active-Active Gigabit uplinks; again nothing unique here, no crazy routing, just symmetric, equal cost uplinks. The uplinks terminate in our cage at a pair of Cisco 7301s that we use as our gateway/border routers. We do some basic filtering here, but nothing too outrageous; we tier our filtering to try to spread the load. From the border routers, the bits hit our core switches/routers, a pair of Foundry BigIron 8000s. They have been our workhorses throughout the years. The BigIron 8000s have been in production since we built this data center in 2002 and actually, having just looked at it... haven't been rebooted since. These guys used to be our border routers, but alas... their CPUs just weren't up to the task after all these years and growth. Many machines plug directly into these core switches, however for certain self contained racks we branch off to Foundry FastIron 9604s. They are basically switches and do nothing but save us ports on the cores.

Now onto the meat: the actual systems. We've gone through many vendors over the years. Some good, some...not so much. We've had our share of problems with everyone. Currently in production we have the following: HP, Dell, IBM, Rackable, and I kid you not, VA Linux Systems. Since this article is about Slashdot, I'll stick to their hardware. The first hop on the way to Slashdot is the load balancing firewalls, which are a pair of Rackable Systems 1Us; P4 Xeon 2.66Gz, 2G RAM, 2x80GB IDE, running CentOS and LVS. These guys distribute the traffic to the next hop, which are the web servers.

Slashdot currently has 16 web servers all of which are running Red Hat 9. Two serve static content: javascript, images, and the front page for non logged-in users. Four serve the front page to logged in users. And the remaining ten handle comment pages. All web servers are Rackable 1U servers with 2 Xeon 2.66Ghz processors, 2GB of RAM, and 2x80GB IDE hard drives. The web servers all NFS mount the NFS server, which is a Rackable 2U with 2 Xeon 2.4Ghz processors, 2GB of RAM, and 4x36GB 15K RPM SCSI drives. (CT: Just as a note, we frequently shuffle these 16 servers from one task to another to handle changes in load or performance. Next week's software story will explain in much more detail exactly what we do with those machines. Also as a note- the NFS is read-only, which was really the only safe way to use NFS around 1999 when we started doing it this way.)

Besides the 16 web servers, we have 7 databases. They currently are all running CentOS 4. They breakdown as follows: 2 Dual Opteron 270's with 16GB RAM, 4x36GB 15K RPM SCSI Drives These are doing multiple-master replication, with one acting as Slashdot's single write-only DB, and the other acting as a reader. We have the ability to swap their functions dynamically at any time, providing an acceptable level of failover.

2 Dual Opteron 270's with 8GB RAM, 4x36GB 15K RPM SCSI Drives These are Slashdot's reader DBs. Each derives data from a specific master database (listed above). The idea is that we can add more reader databases as we need to scale. These boxes are barely a year old now — and still are plenty fast for our needs.

Lastly, we have 3 Quad P3 Xeon 700Mhz with 4GB RAM, 8x36GB 10K RPM SCSI Drives which are sort of our miscellaneous 'other' boxes. They are used to host our accesslog writer, an accesslog reader, and Slashdot's search database. We need this much for accesslogs because moderation and stats require a lot of CPU time for computation.

And that is basically it, in a nutshell. There isn't anything too terribly crazy about the infrastructure. We like to keep things as simple as possible. This design is also very similar to what all the other SourceForge, Inc. sites use, and has proved to scale quite well.

CT: Thanks to Uriah and Chris Brown for the report. Now if only we remember to update the FAQ entry...

Comments

Windows?

[mseidl]

I'm like sooooooooo surprised you guys aren't running nt4 boxes. IIS was this sh!t back in the day

Re:Windows?

[QBasicer]

Do you know how many people would be heartbroken if they found out Slashdot was run off windows?

Re:Windows?

[truthsearch]

Probably the same number of people that were elated to find out that Microsoft's site is cached using Linux. I found it deliciously amusing.

Re:Windows?

[ObsessiveMathsFreak]

Not as many people who were traumatized to hear that Slashdot is run on a Dell.

Re:Windows?

[BadMackTuck]

Microsoft isn't running linux servers, they're using Akamai who does.

http://news.netcraft.com/archives/2003/08/17/wwwmicrosoftcom_runs_linux_up_to_a_point_.html

Savvis

[garethwi]

Nice to see you're hosted by a Microsoft Gold Partner. That's a benchmark of quality.

Re:Savvis

[Anonymous Coward]

They've changed hands several times and names even more times since we moved in.

Re:Savvis

[bunco]

Twice, actually. Slashdot is hosted in an Exodus legacy data center. Exodus was bought by Cable & Wireless who then sold their US network assets to Savvis.

Depending on who you talk to, you'll get different responses about Savvis. This is mainly due to the heritage of various customers. i.e. Savvis/Bridge/Intel vs Exodus reputation.

Savvis is actually the conglomeration of _many_ companies.

Exodus == (Exodus, AIS, Arca, Cohesive, Network-1, Global Center)
C&W US == (MCI (IP backbone), Exodus, Digital Island)
Savvis == (C&W US, Intel Hosting, Bridge Networks)

Redhat 9

[Anonymous Coward]

Tell me that's a hilarious joke...

Re:Redhat 9

[MisterFuRR]

If it works, and theres no need to change -- why introduce unknown incompatibility...its a production network -- not your home box.

Re:Redhat 9

[Shakrai]

and is therefore completely unsupported.

I've been in the mission critical provider business (ISP, then an insurance agency, now in health care) for almost eight years now and I'd take community support (websites, forums, and *gasp* USENET) over corporate support any day of the week.

Yes, it's nice to be able to sell your project to the PHB/MBA by saying "We'll always have support if we have problems!", but anybody that has spent a few hours on the phone with their vendor and who realizes that they know the product better then the Level Four support guys do, comes to realize that support isn't the sliver bullet that it's presented as.

Have my vendors helped me out in the past? Absolutely. But will I upgrade away from a rock-solid system that has served me well for years merely to gain support with a newer version? No way in hell......

Re:Redhat 9

[Shakrai]

Rule #1 of a mission-critical application is "if it ain't broke, don't fix it."

Slashdot is mission-critical? That's what I'll tell my boss the next time he asks why I'm spending so much time on it ;)

Kidding aside though, I'm rather fond of the old joke that I found in 'fortune -o'. "Working computer hardware is a lot like an erect penis. It stays up as long as you don't fuck with it."

Re:Redhat 9

[WhatAmIDoingHere]

How do you BRICK a computer short of taking an axe to it? Boot from install media and reinstall. If the hard drive is shot, the hard drive is shot. But a dead HDD doesn't mean it's bricked. I can see maybe fucking up a BIOS upgrade but even with that there are ways to undo the damage.

You people keep using the word "brick" to refer to "broken software that can easily be reinstalled."

the powers that be

[ebolaZaireRules]

The hardware that powers slashdot?

I wanna know about the power that powers slashdot... are you really as green as the default colour scheme?

Re:the powers that be

[the_tsi]

Only the bad guys have red LCDs. That's how you tell them apart from the good guys, who always have blue or green. ... or purple, I guess, if you're a bad ass motherfucker.

Redhat 9?

[eli pabst]

It'll be interesting to read the software section. It was surprising to see that they use an EOL'd version of Redhat (RH 9) that is no longer supported by Redhat. Granted, they're just webservers, but you'd think that would still require a lot of manually updating to keep things patched.

Write-only database?

[jolyonr]

That sounds useful! I use /dev/null as a write-only database. Very efficient.

Jolyon

Re:Write-only database?

[Ron Harwood]

I'm trying to assume that's humour... but that said...

If you have a farm of replicated mysql servers (which are read only - as replication is one way here) you need a db to write to.... not reading from it reduces the load on that server.

So, assuming that your read-mostly - it's actually a nice way to balance the load across multiple systems.

Re:Write-only database?

[cbiltcliffe]

Magic.

Possibly obtuse question

[athloi]

What determines why you run Red Hat 9 on some systems, and CentOS on others? Was BSD even considered? (You wouldn't run on Macs, would you?)

Re:Possibly obtuse question

[Precision]

Deployment date. The Redhat 9 machines were deployed 3 years ago and just haven't needed to be reinstalled yet. BSD, not so much.. we have a team of great linux admins, introducing another variable isn't likely to happen.

Re:Possibly obtuse question

[saterdaies]

Usually these decisions are made based on familiarity, availability, and the like. If you're staff and you are all really familiar with RedHat, why would you force them to run BSD or Debian? Each system has pros and cons, but to be honest, the largest pro or con is usually familiarity. It's really easy to get familiar enough with any *nix to get Apache running. The issue is whether you have the knowledge to deal with it when your live webserver suddenly stops responding to requests.

Stability and familiarity are more important than the latest cool distro. Is there a reason that they should have picked BSD over RedHat? Of course there are some. There are others to pick RedHat over a BSD. In the end, you have to go with what you're comfortable and familiar with in order to ensure that you can deal with sudden, unexpected problems.

Re:Possibly obtuse question

[bogie]

Just because you don't have an ISO to download doesn't mean it isn't Free. How do you think CentOS gets made?

Reference Materials

[Ided]

This may be slightly off but I was wondering if anyone could recommend some good reading materials for setting up clustered sites or how to spread out work loads like they're doing with their systems.

Re:Reference Materials

[the_tsi]

http://www.linuxvirtualserver.org/ or anything about F5 BigIPs. Most of understanding load balancing is about understanding (a) how to fool layers above you in the OSI stack (switching on layer 4 through 7 -- particularly 7 -- can take a while to wrap your head around) and (b) the algorithms to pick which physical server gets the next connection (round robin, least connections, predictive, whatever).

Finally some adult stories

[Jack Malmostoso]

Oh yes, geek pornography finally appears on /. :)
Thanks for the report, looking forward to the software part!

Re:Interesting

[CatPieMan]

Non-logged in user see the same page, so its basically a static page that gets updated every couple of minutes.

Logged in users can have a bunch of customization options on the front-end, which would take more resources.

I find it just as interesting that the logged-in readers use up that much more CPU.

Re:Interesting

[ZachPruckowski]

Who said subscribers have two dedicated servers to read the main page? The article/summary says that two servers serve for ACs reading the main page, and 4 for logged-in users. I saw no subscriber/non-subscriber distinction.

bandwidth usage and cost?

[TwoWheelTomy]

wonder how much bandwidth slashdot is using and how much it costs.

Re:bandwidth usage and cost?

[Lumpy]

That's crazy. just lease a dark fiber. WE do that for a point to point that is 12 miles and pay $1500.00 a month. bring my own gear and I'm running 1000Mb happily.

The savings pays for the gear in less than 2 years plus we have 10X the band width as well as full control over the connection.

Re:bandwidth usage and cost?

[Precision]

The average monthly bandwidth usage for /. is around 40-50mbit/sec, which is relatively small. As for cost, you can contact your local ISP for a guesstimate, we get fairly deep discounts since we push quite a bit more with all the sites consolidated.

Re:bandwidth usage and cost?

[anticypher]

For a 100Mbit/sec commit on a GigE connection for a full internet feed, I've been getting quotes from California of around US$10-$16 per Mbit, depending on the data centre and provider. For a 1 Gbps commit, the price drops to around US$6/Mbit. Those are prices from sales droids without any attempt at negotiating a better price.

For a site like /., a 95th percentile bandwidth of 50 Mbps for the month would cost between US$500 and $800, less if the total commit for all of OSDN's traffic was much higher. Add in hosting costs estimated around US$2,500 per month for a cage with room for 6 racks and matching electricity and cooling, and you can calculate how much operating costs /. runs.

After that, you have to count up all the amazing 7 figure salaries of Rob and the gang who keep things running :-)

the AC

Re:bandwidth usage and cost?

[anticypher]

I was shopping for transit in the U.S. this summer, and those were the reasonable prices from companies that I also work with here in Europe. I don't know of any tier-1 who will bother with 1Mbps, most tier-2's won't either. My "standardized" quote is for 100Mbps commit on a GigEthernet port, that can handle sustained traffic of 800Mbps. This lets me compare without giving away details of my clients before contracts and NDAs can be signed.

However, I had a strange split in quotes I received. Some were in the range I expected, from about US$10 for a 100 Mbps commit (minimum bill about US$1000) up to around US$20-$25/Mbps. Then there was a huge jump up to the $500/Mbps range you speak of. Companies that were obviously not one of the tier-1 or 2 players, just resellers of tier-2 bandwith, but who didn't seem capable of competing.

Quite a few places seemed to think they could obfuscate the quote by refusing to deal in Mbps/month, and instead would offer traffic totals of 100 Gigabytes for inbound+outbound together. There were others who offered peak+offpeak or other ways to hide the usual Mbps/month quote.

One place was offering GigE ports, but I discovered later their internet transit was just a pair of 100M copper links. They sold their traffic as a package but when you calculate out 50 Gigabytes in one month into a traffic figure, you come up with something like 1-2 Mbps, for the low price of US$500. This may be where you are getting your quotes from.

As a very general rule of thumb, the tier-1s don't want to deal with a monthly bill of less than US$10,000, the tier-2s don't want anything less than US$1,000, and the tiny resellers will try to sell you everything they can (rackspace, metered electricity, port costs, traffic) to try to keep the bill upwards of $300-$500/month.

Just for comparison, even with the US dollar in free fall this summer, US prices were well over twice what we pay in Europe for internet transit.

the AC

Why CentOS?

[bogaboga]

I am not saying that CentOS is any inferior at all but wonder why they chose it over all the possible serious systems in the Linux world. Is there anything CentOS does better than say OpenSUSE or Ubuntu/Debian and the rest?

Re:Why CentOS?

[pak9rabid]

CentOS is RHEL, minus the support. CentOS is 100% binary compatible w/RHEL as well, meaning the RPMSs you'd get from RHEL would work just fine in CentOS and vice-versa.

Re:Why CentOS?

[Precision]

We use a combination of CentOS and RHEL. The reason we chose CentOS over say debian is because it is basically identical to RHEL, we end up with a "single" platform that we have to deploy, test, and build packages for regardless of support. Depending on the system we will deploy either RHEL or CentOS accordingly based on support requirements.

Re:Why CentOS?

[idontgno]

Yes indeed; RedHat with the serial numbers filed off.

My GNU/Linux server experience is pretty much all in the RH family tree, except for a brief fling with SUSE for SPARC in the early '00s. For desktop, though, I have to admit that Kubuntu is pretty sweet. I just haven't had to (or had the chance to) learn the guts like I've had to (or had the opportunity to) learn the ugly white underbelly of server OSs.

But Hell, I come from a SunOS and AIX background, so almost nothing in the GNU/Linux universe can really scare me much.

Multiple master DBs

[atomic777]

"These are doing multiple-master replication, with one acting as Slashdot's single write-only DB, and the other acting as a reader."

Isn't that a contradiction? If you have only one write DB, why do you need multiple masters, aren't the other 6 just slaves at that point? Or are there separate master/slave pairs (I'm assuming these are MySQL databases)

Re:Multiple master DBs

[Bellum Aeternus]

master-master allows really fast fail over because you don't need to down the system to re-cofig a slave as a master. I've actually worked with companies that have master-master-master clusters.

One way it's better

[Synn]

It's familiar to people who are used to working with Red Hat.

Considered a CDN?

[xmpcray]

I was wondering if you ever considered using a CDN service like Akamai to serve content? Most of the big sites (Apple/MS etc) use it.

Re:Considered a CDN?

[Precision]

Actually many of our sites do use a CDN, however the /. devs long ago decided against it for some reason or another. Heck it's still even all setup for them.

Re:Interesting

[jamie]

Yeah, I wasn't sure what he meant either. We have 2 webheads serving static pages (like the non-logged-in homepage), and 4 serving specifically the dynamically-generated homepage for all logged-in users. Plus 1 that serves all SSL traffic, which subscribers can use.

People often say "subscriber" when they mean "logged-in Slashdot user," not specifically a paying subscriber.

I want my sense of childlike wonder back!

[foo fighter]

I always imagined slashdot ran on hundreds (perhaps thousands) of modded Dreamcast consoles powered by lucky, randomly selected registered users running in hamster wheels who were lured by blocks of Wisconsin cheese dangling just out of reach.

Thanks for destroying my sense of childlike wonder, you insensitive clods!

Re:I want my sense of childlike wonder back!

[An ominous Cow art]

I always imagined slashdot ran on hundreds (perhaps thousands) of modded Dreamcast consoles Are you seriously saying that you imagined a Beowulf Cluster of them?

backup?

[nido]

Well, let's begin with the most boring and basic details. We're hosted at a Savvis data center in the Bay Area. Do you ever worry that a big earthquake will hit and your datacenter goes offline? Do you at least keep an offsite backup?

Re:backup?

[statikuz]

From the website: "SAVVIS has done extensive engineering to ensure that any Datacenter located in a region prone to seismic activity is braced for such events. Design elements include, seismic isolation equipment to cushion facilities against movement as well as seismic bracing earthquake bracing on all equipment racks. All SAVVIS Datacenters have racks anchored to the concrete slab below the raised floor."

Re:backup?

[Eponymous Bastard]

Well, let's begin with the most boring and basic details. We're hosted at a Savvis data center in the Bay Area. Do you ever worry that a big earthquake will hit and your datacenter goes offline? Do you at least keep an offsite backup? First rule of offsite backups: Never talk about your offsite backups.
Second rule of offsite backups: Never talk about where you keep your offsite backups.

You thought I was going somewhere else with that didn't you?

In all seriousness, that sounds like it would be in the software article instead.

Re:backup?

[Precision]

Of course we do offsite backups, but also we're currently preparing building a new primary data center in Chicago away from Earthquake land.

Re:backup?

[shokk]

At which point, Slashdot is the least of people's worries. This is a news entertainment site, not a critical care facility.

Thanks

[debrain]

To the editors:

Thanks for this. It's really very interesting.

-B

read the entire series

[clem]

I can't wait for "Slashdot's Setup, Part 8 - Root Passwords".

Re:read the entire series

[moosesocks]

I have this funny vision of what would happen if /. got hacked, and how it would have been done:

The admins would wake up the next day to discover that the site was running perfectly normally, but was performing slightly faster than normal.

After closer inspection, they'd find that their datacenter had been emptied, and replaced by a single Apple ][ that had been hacked to run the latest version of Ubuntu, and that slashcode had been rewritten so that it would perform all of the same functions as the previous slashcode, but ran at twice the speed... on the Apple ][.

A post-it would be found stuck to the screen, stating that all of slashdot's old and now unnecessary hardware had been sold, with the proceeds being donated to the EFF. The message would likely include or be in the form of a Soviet Russia joke. Additionally, a miniaturized plastic Gnu would be left behind as a calling-card.

The news of this would be regarded as insignificant by the editors, until over a year later, it gets posted four times in the span of two days.

Re:Interesting

[avronius]

best thrustworthy news site I do not think that word means what you think it means...

Comment woes

[theantipop]

Any chance that with all that iron you can loosen up the crazy restrictions on comment posting? It can get rather ridiculous sometimes.

Artistic solution...

[Anonymous Coward]

"Really, once you've seen one class A data center, you've seen them all. (CT: I've still never seen one. And they won't let us take pictures. Boo savvis.)"

Send in a courtroom artist :-)

Yep, you're dead on the money about Level4 support

[JohnnyComeLately]

I've worked 2nd Tier Technical Support within Sprint PCS, and it even got to the point I was helping their Level 3 with future system designs. Their level 1 was often a joke. One guy wanted me to reinstall packages on a Sun Solaris machine...I said, "This is not Windows...reinstalling will result in EXACTLY the same error" which of course, it did.

Anyway, it did get to a point where I instantly got escalated to their 2 or 3 tier because if I couldn't fix it, or I couldn't find the answer withing a Unix forum on-line, they would have a hard time offering a solution. This was supporting about 300 Sun Netra systems running Solaris 9.

Re:You probably shouldn't tell us too much informa

[Precision]

Lemme know how that works out for you, considering they're doing layer 2 only and don't have an IP address.

Re:Active-Active Gigabit uplinks

[cecil_turtle]

If you're on an OC-24 or above loop your providers may be able to sell you bandwidth in gigabit increments. If you're on an OC-3 or OC-12 loop you can normally buy in 100 megabit increments. Otherwise "legacy" OC-3 is something like 155mbps and OC-12 is 622mbps if you're using the whole line (not breaking it off into T-1's, DS-3's, etc).

Once you have multiple uplinks from different providers you would typically use the BGP protocol to announce your IP space on both providers, then when people try to get to your site they will come down the provider which is the fewest hops for them. If one of the lines goes down, the other one is still active and then everybody will go down the running line until the broken one gets fixed.

Re:Interesting

[Provocateur]

Yes but did you see where all this hardware is wired to a Palm Pilot??

The Palm, so the legend goes, provides the necessary horsepower for the few that actually do read TFA's.

Yes, MySQL.

[Ayanami Rei]

NT

root passwd != externally vulnerable

[Steve Hamlin]

I can't wait for "Slashdot's Setup, Part 8 - Root Passwords"

And what would you do with them? Knowing the root password shouldn't get you into a properly configured and patched system.

I even remember one cracking contest where the owner of the machine gave out the root password to the target machine. (quick google: nope)

You could attack the bandwidth, or try to get physical access. But if Cmdr. Taco can't get in....

Re:RAID? Mail-Server?

[Precision]

Yes, all our servers are at least RAID1, as for email, this article was Slashdot specific machines only. There are quite a few shared systems, including the outgoing mail relay.

Re:You probably shouldn't tell us too much informa

[Precision]

Ding, we /used/ to use them as layer 3 routers, but they couldn't keep up after the years and alas, they've been relegated to dumb layer 2 switches now. The poor cpu's can't keep up with anything else. We do have OOB serial management on them like you mentioned however.