| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Comments:
<BBS> how are they going to have peers?
<Aurelius> BBS: GET THEIR OWN KTHX
<BBS> ok, how about addressing some of the questions I brought up?
<Aurelius> ok...
<BBS> it's irresponsible to take something that for all intents and purposes is a fad, as seen by other mmo's, and build an entire network that could be useless in 2 years
<BBS> no company who values their bottom dollar is going to do that
<Aurelius> you can always "debuild" the network
<BBS> so outsourcing is the way to go
<BBS> pure and simple
<BBS> and what, sell the shti on ebay?
<Aurelius> but if you are smart
<Aurelius> you come up with a new product to take the place of the dying product
<BBS> wow, thank god you don't run a company, I'd feel bad for your employees
<BBS> you try to do that
<BBS> doesn't always work
<BBS> EQ has EQ2, they're still floundering compared to their original numbers (partly due to wow)
<BBS> SWG, was big when it first hit, now it's **** too
<Aurelius> BBS: don't be so condescending. you're starting to sound like peerce
<BBS> no, I'm being realistic
<BBS> there's a big picture and you're not seeing it
<Aurelius> no, i am
<Aurelius> it's not that hard to build a network with multiple peers in a single location...
<BBS> but it's not a single location
<Aurelius> they aren't hosting in a single location?
<BBS> they have servers east coast, west coast, europe, asia....
<BBS> no
<Aurelius> ok
<Aurelius> me being not a gamer, i don't know where they have their servers
<BBS> if it was that simple, sure, buy a cage at equinix
<BBS> done
<Aurelius> yeah
<Aurelius> you were just going off about how stupid they are
<BBS> dude, they have like 6 million subscribers world wide
<Aurelius> and how a single router hosed them
<BBS> I NEVER said a single router
<BBS> they have ONE provider, one host, ATT
<BBS> ATT buys NO transit
<BBS> from anyone
<Aurelius> err, not you... cow did
<BBS> if something goes down, they're ****ed
<BBS> they have peering agreements, sure
<BBS> and they lose some income when they have outages
<Aurelius> sorry for hte mistaken attribution :)
<BBS> but it's different than a tier 2 that diversifies
<Gimp-Cow> router down again
<Gimp-Cow> itiots
<Gimp-Cow> idiots
<BBS> and PAYS for transit
<Aurelius> yeah
<Aurelius> that much definitely is true
<BBS> I'd go with a tier 2 any day of the week, unless you're doing your own diversification
<Aurelius> yeah, unless you wanna become a tier 2, go with tier 2 :)
<BBS> I don't think there's anything wrong with being singly homed, just not with ATT, MCI, L(3), etc
<BBS> and then of course, if you're going to be singly homed, there damn well better be hsrp **** going on
<BBS> heh
<Aurelius> heh, ****, i haven't heard hsrp in years.
<BBS> but that's another story, and that's internal to the hosting company
<BBS> well, I don't care if it's hsrp or some other form of dynamic routing
<Aurelius> where 2m could be impacted
<Aurelius> err
<Aurelius> i just haven't heard that term in years :)
<BBS> point is that you damn well better have multiple physical paths to a customer that large
<Aurelius> made me have a flashback hehehe
<BBS> the problems blizzard has with wow are the same every time. they really only ever have two problems
<BBS> 1) their updates **** up and their servers take longer to get back up
<BBS> 2) the network breaks
<BBS> their network breaks WAAAAAAY too often
<Aurelius> not much else to break ;)
<BBS> at least once a month
<BBS> I can forgive the former problem. we're talking about going live on 150+ servers nationwide all in one day, on live servers that are actually huge clusters and have databases the size I can't even imagine
<Gimp-Cow> they have too much load for whatever infrastructure they have
<BBS> the major upgrades no doubt involve DB schema changes, etc
<Aurelius> perhaps yeah
<Gimp-Cow> yeah
<BBS> **** happens, it's hard to test that sort of thing properly on devel servers, you can't account for idiosyncrasies
<Gimp-Cow> but non patch days are just reindexes
<Aurelius> i can accept that happening periodically
<BBS> but the network, a proper network is designed with failover in mind
<BBS> Gimp-Cow - I've never seen major delays on non-patch days
<BBS> it's the patch days that **** **** up
<Aurelius> sometimes bugs can't be found unless you have all the users hitting that ****
<BBS> and again, I'm fine with that
<Aurelius> like yer mom
<BBS> I'd be more pissed if they never added content or fixed **** :)
<Aurelius> i had something i wrote once that under all sorts of random and non random tests _never_ broke
<Aurelius> but the moment i had real users hitting it, it randomly would segfault
<BBS> anyway, I still smell a bit like Aurelius' mom. her sweat rubbed off on me while I was making her scream "who's your daddy", so I'm going to go shower
<Aurelius> clobber the stack and we never did end up finding it
<Aurelius> BBS: hope you enjoyed it
<BBS> not really :(
<Aurelius> then why would you go back?
<BBS> I won't
<Aurelius> or are you that desperate
<Aurelius> lamer.
<BBS> but she was like "first one's free"
<Aurelius> that's not what i hear
<BBS> **** dude
<BBS> your'e right
<BBS> your mom took my wallet
<Aurelius> told you
<XTR> omg.
<Aurelius> ****er.
<Aurelius> hold XTR
<XTR> Hold me.
<Aurelius> fags.
<BBS> luckily, I just took it from your room while she was getting me a beer
<BBS> so no harm done!
<BBS> anyway, yeah, shower time
<XTR> My domain controler is so hosed.
<XTR> Fuuuuuuuuuck man.
<Ted-Lap> http://img164.imageshack.us/img164/8666/njbnjkn2dj.jpg <-- cl***y chick
<pale_blue> that's kinda hot if you factor out the river of piss. :-/
<Aurelius> would be hotter with scat
<pale_blue> i can tolerate some piss. but once the **** starts flyin', no ****ing way.
<XTR> :(
<Joe-0> hola hos
<Joe-0> xtr i think im getting some azenis
<XTR> omg.
<Joe-0> rt615
<pale_blue> s/az/p/
<Joe-0> HAHAHAHA
<Joe-0> <--- owned
<XTR> :(
<XTR> I poured myself way too much vodka.
<pale_blue> you'd better drink it all, you don't wanna waste it. there are sober kids in africa!
<Mag69> anybody knwow what would cause the lights on a RAID volumes HD's to flash every few seconds when connected to a PERC3 card?
<peerzzzzz> Mag69; its polling the drives
<peerzzzzz> or doing background surface scans, many raids do that
<Mag69> any way to turn that off?
<peerce> why?
<peerce> put tape over the ligths.
<Mag69> the same drives used to be on a Adaptec card and it didnt do that
<peerce> :D
<Mag69> peerce: hehehe
<peerce> thats nice. adaptec ****s
<Mag69> its it harmfull to the drives?
<peerce> no
<Mag69> peerce: yeah thats why we switched to perc's
<Mag69> just kind annoying because we cant tell actually disk activity
<peerce> uh, isnt it in a rack in a data center where noone even physicaly sees it for months at a time?
<peerce> it SHOULD be :D
<Mag69> well its a external raid tower unit. we have been trying to find out the performance bottle neck with our sql server
<Badtzmaru> in a perfect world, but people are stupid
<pale_blue> can you save ****? good. can you get the **** back when you ask for it? good. it's working.
<Mag69> the drives are constantly being hit. we switched to a perc3 card to see if it would help but now we cant tell actuall disk activity versus the polling
<peerce> Mag69; first place to start would be to organize the SQL data as 3 seperate tablespaces, each on its own dedicated RAID10 array
<peerce> yes, that takes a minimum of 12 spindles
<Mag69> peerce: yeah, unfortunatly they havent invested in this box.
<peerce> this is MS SQL? put the redo logs on one tablespace, the indicies on another, and the actual data tables on the 3rd
<pale_blue> and in a perfect world everybody has a bottomless IT budget.
Return to
#windows
or
Go to some related
logs:
superdan db
#heroin
ddgirls passes
shlamele
#politics
valid 700-PID
#sql
#beginner
Your message was not delivered within 1 days and 0 hours.
#beginner
|
|
