whee.
Wraithan: works now thanks :) routing seems to have been sorted overnight :)
hah, that's a funny - the guy who joined with nick 'M0OML' - i know him irl... small world, didn't know he was an arp customer hehe :)
Damn
I'm hung over
But that's no problem because I'm going out partying again today!
goodie
howdy
man - it's raining hard here in pdx
plenty of rain here in the uk too
oh good, glad we're not hogging it all :)
was raining here in Ireland earlier, pissing cats and dogs. moved off now
Raining in Boston too, and I'm heading to Baltimore where it is also raining.
RandalSchwartz: You're in PDX as well?
We had a beautiful day yesterday, then today it is back to shit weather
I am, at least for now
yes, amazing difference
that's rose festival week for you. :)
Aye
I always found it amusing that we chose the most likely week for rain for our big tourist festival :)
It isn't the most likely, we are nearing the end of spring
spring = rainy
summer = slightly less rainy
fall = rainy
This year has rained more than it has in the past
winter = clear or snowy
and the summer rainy is an overstatement
well - in recent years, it's been different
but as a child, I can't recall seven days in a row without rain during summer
hence, I was trained early on to always take a coat with a hood even when it was 80 out
You were a child, your memory is faulty as it is.
my memory as an adult isn't much better. :)
I am born and raised here, summer has always been pretty sunny with a couple days of rain
maybe you're younger. :)
I probably am
Tends to be the case in tech circles
ahh yeah.  so my experience covers the 70's and 60's here. :)
as I said, it's been getting more sunny over time
probably due to the population increase
Even more likely that your memory is faulty ;)
er what?
more generated heat in the portland metro area
so less chance of rain
I
'm not so sure about that
I mean, I recognize that it is warmer due to cities
it has to be cold enough to condense the rain out of the sky
if it's warmer, it's more likely to not dump until it hits the cascades
since we're in this valley, the first scraping happens on the coast range, and the second on the cascades
that's why it's never humid here
Guy who works in my office is a atmospheric scientist, I'll see how viable this is
tomorrow
and I too have studied meterology, both early on, and because I'm a pilot
He has a PhD...
sure - I learned from guys with those. :)
And you are joe blow on the internet
No offense
not really.  I've never gone by joe blow
lol
and why do people always say "no offense" right around saying something offensive? :) :)
Because that isn't offensive to most people but some people would be offended
I'm also curious... why is the "internet" important in that sentence?
if I was talking to you at a bus stop, would you say the same thing?
yes
so you don't recognize that others might actually know things unless credentialed?
that's sad
I've made an entire successful career without having *any* credentials in the area
So have I
then your behavior is inconsistent... :)
And once people get to knwo me they trust my opinion on the things I have specialized in
and frankly, at 23 you can't have much of a career yet
I started when I was 15
8 years regardless
Ok, so you are also a sad individual, obviously I can't have a career because I am 'only' 23
I'm a programmer, age doesn't really play into this
I'm speaking only of length of career
not what you are doing right now
I started coding for money at 15, started coding earlier than that.
me too
but 25 years before you started. :)
What does length of career have to do with whether it is a career then?
you're leaving out the word "entire successful"
that was my point
you're skipping over it
*shrug* the key part was successful career, which I have, without credentials
I could probably find people with credentials to agree that eight years isn't much of a success *yet*
it *can* be eventually
but you are missing that point
The point isn't the credentials though, they are just a different form of displaying expertise. His PhD means he studied for years to get to where he is. Which means it is easier to believe him, then after talking to him and he seems crediable it makes me believe him over someone on the internet who claims to be a pilot who has learned about meterology
Do you trust some dude who claims some stuff, or a person in your life who has already proven themselves trustworthy on the topic?
I tend to start by trusting people
it's gotten me further to mop up the messes from that than to start by distrusting people
This is the internet, on IRC, full of trolls, I don't bother with trust unless I know someone
that you're unwilling to trust me tells me you got slapped around pretty badly early on, and for that, I'm sad for you.
both in what must've happened then, and how it limits you now
Limits?
Just because I don't trust information doesn't mean I don't retain it
I just verify the parts that are questionable to me
and it's not just about the information though
it's also the way you must necessarily treat me
life isn't about just gathering and verifying info
it's also about the connections you make
Are you SURE about that? ;P
What are connections with people? In my mind, that is data.
I wasn't sure until 1995
Wow... flashbacks to how I was when I was 23. :)
maybe it's just a phase, and you'll work through it.
I did.
but I had to get burned a few times before learning
it's not just data, it's asynchronous yet synchronized data with tangible values associated to the relationship independent of either side
and it's also context
in some cases, the context of the data is even more important than the data
data only becomes knowledge with context and understanding :P
until you understand why having a round of golf with a potential client, even if you never discuss the business, is important... you won't get it. :)
depends what sorta businesss you're in i guess
all the ones that matter. :)
meh
substitute "round of golf" with an appropriate activity, of course
RandalSchwartz: re: your ipv6 discussion last night, no need for .0 or .255 to be different in ipv4 either.
there has to be a brodcast addr in v4
but maybe the .0 can be unspecial
there doesn't *have* to be a broadcast in ipv4
there does, but it doesn't have to be .255
no, there doesn't
I mean .255 in the sense of "all 1's"
depends on the netmask, etc
that's where I was going with that, yeah.
ahh - ok
well - I didn't have a fast way to say that
a /31 for example is perfectly valid and has no broadcast address
haha. Fair enough ;)
most people are familiar with /24's
that said, /31s are mostly annoying hehe :)
so I was just pretending we were talkin gabout those
bob^^: with a /31 you're the broadcast address ;)
hehe yeah i guess :)
some implementations would treat the odd address of a /31 as a broadcast though, right?
not everything supports a /31
ahh ok
on the most part i wish they didn't exist
stupid address, everyone should use a /30 instead for point to points ;)
if we just soaked up ipv4 faster, we could force moving to ipv6 sooner
too much stuff can't support ipv6 yet
... http://penrose.uk6x.com/
otherwise i'd agree :)
bob^^: how much stuff *can't* support it, and how much of that needs to be public-facing?
i've got a lot of kit in work that has no way of doing ipv6 (yet, it's just slack on the manufacturers part tbh) and has to be public
there's no reason old gear can't get stuck in an ipv4 ghetto the same way we don't worry about ipx networks that don't speak ip :P
this is kit that runs linux, so absolutely no way it shouldn't do ipv6 - it's configured through a web interface that has no options for v6 though :(
plus
in europe
RIPE are being absolutely annoying about assinging ipv6 addresses just as they are fro ipv4
we applied for a small 'test' range, and were denied
are they?
oh yeah
I thought ipv6 was *everywhere* there.
nope
far from it
in fact, things seem further ahead in the US to me
doubt that.
honestly :)
every DC I've been in over there has had native ipv6
rare over here :P
which DCs?
heh - http://ipv6.he.net/statistics/
420 days!  smoke'em if you got'em!
just had a new /16 :)
took a lot of work to get it though :)
I've been using miredo a lot
bob^^: I can imagine, /16 is huge.
so I'm actually on v6 even behind v4
already all gone :(
and I connect to my ARP box via v6
... how did you manage to assign all of a /16?
ISP :)
RandalSchwartz: sounds inefficient ;)
we've got two /16's, a /17 and a /18
all full pretty much
"everybody gets an IP!  everybody gets an IP!" </oprah>
bob^^: even if you're assigning them all a /48 isn't that still 4 billion of them?
ah - this is v4
so no :)
oh
I was going to say :P
RIPE are *really* annoying with v4
RandalSchwartz, look under your chaaaiiirs
but i was surprised how annoying they were with v6 tbh, given how many there are
"ooh!  10.42.69.69!"
YOU WIN!
now pay taxes on it
RandalSchwartz: I dunno, I'll put more faith in that countdown when they run out of ranges to reassign.
tbh
they should be forcing the original /8 assignments back into free use :)
those /8s are held by very big companies, I don't think ARIN has a lot of leverage :P
yeah :(
even the US gov't couldn't afford litigation with IBM ;)
i know of one educational instituion here in the UK that has a *seriously* large assignment from way back when the internet started
and they don't use it
at all
but they refuse to give it up :(
of course not, it's going to be worth something some day.
that would be irresponsible.
indeed :)
one thing I would suspect is that a lot of these original blocks are actually assigned internally.
yes, you are right
i know IBM do that with 9/8
I think I heard GE does as well.
you'd think they could free up at least some of the US gov't assignments too
i'm guessing HP probably do too
DISA has ... well, at least the 4 /8s I see.
heh - hitting red.stonehenge.com from the he.net in one wilshire is 2 hops away
... http://lg.he.net/
there's also, imho, a *lot* of waste around multicast
pfft, traceroute lies.
bob^^: spoken like someone who doesn't use multicast ;)
oh no, we use mcast
but mostly internal :)
same.
is 5/8 allocated yet?
it says IANA-reserved
it's one of those stupid ones that people have used for private stuff
hamachi uses 5/8 for example
oh.  evil
even if it was assigned, chances are it'd end up polluted :(
just like 1/8
and I meant that also about hamachi :)
yeah :)
let's just take back all the ipv4 we gave china
they're only abusing it anyway :)
true!
does ford really need 19/8 ?
we'll give them a /30, everything else they can NAT ;)
doubt it, but i bet it's used internally on everything :(
23/8 is reserved too?
hehe
http://www.iana.org/assignments/ipv4-address-space/ipv4-address-space.xml
;)
that's too easy
240/8->255/8 is still free too
the old class D?
bob^^: meh, 69/8 was successfully assigned.
... eventually.
fucking bogons lists.
hehehe
those things piss me off. They're like RBLs but worse because they never get turned off when they're unmaintained :P
yeah, unmaintained bogons = very bad news
yeah.
well maintained = good news
... I guess, but still.
but on the whole you're right, it's never kept maintained
we drop the usual suspects at network borders
rfc1918 mostly
I had a 69.something block (I forget which now) and would get random network problems because some idiot didn't update.
i read a lot about 1/8 being assigned as well and causing all sorts of trouble
because i believe cisco used to train people to use 1/8 for point to point internal stuff
haha.
indeed :/
I can't imagine the fun 1.2.3.4 gets.
iirc, a route reflector ended up having about 400mbit/s shoved up it :)
meh, 400mbit...
;)
http://labs.ripe.net/content/pollution-18
wasn't that much :)
it was just on a very crappy link and ended up a little oversubscribed :)
ended up dropping announcments for 1.1.1.0/24 and 1.2.3.0/24 hehe
at the end of the day, I'm still amazed anything works on the net
it's like... magic
:)
come work for an isp ;)
well - I mean I know how it works technically
been there, done that ;)
it's still just... full of wonder and amazement
no kidding.
considering that here we are, having this conversation
and aren't paying compuserve 10 cents a minute
or dialed in to an 8-line BBS
and that I own this machine...
no wait... this *virtual* machine
somewhere in the bits inside a real machine in LA
so that I can have a net presence
hehe
which is a gross presence minus the overhead. :)
anyway, I wax poetically now :)
well, we officially killed our vps :P
in single user mode now trying to fix pdns
ouch
ZFS may have been a bad idea on a vps
not enough memory
amdprophet: I haven't had problems with ZFS.. yet.
it isn't ZFS's fault directly
we're running apache with a couple of php apps and a couple of rails apps
apache memory usage skyrockets and ZFS has no memory left to use, which causes it to freeze up
ahh
I've got quite a few vhosts on my web server, but I've cached it well enough (and I use lighttpd) that my memory stays pretty healthy.
yea, lighttpd should be a lot better, we're looking at switching to nginx as a proxy, lighttpd for php apps, and unicorn for rails apps
nginx is very good
I don't like lighttpd because of the memory leaks, I think they may have been fixed now
yea that's what i heard about lighttpd, monit should take care of that though
I hear mostly good things about nginx
I gotta get them on FLOSS Weekly
do you run FLOSS Weekly?
au: lot of people use it in huge production environments, must be solved to some degree.
I do
awesome
i'm going to have to subscribe to it
twit.tv/floss
120 back issues
you'll be there for a while :)
video too?
the recent ones, yes
oh nice, one on haiku!
they're not all in the RSS feed
but the names are all regular
curl -O 'http://www.podtrac.com/pts/redirect.mp3/twit.cachefly.net/floss[0-0122].mp3'
Oh hmm. the older ones aren't there
:(
I'm asking... they used to be there.
they recently renumbered to be 4-digit numbers
probably just a mistake in the files
ahh - the older ones were named FLOSS-nn.mp3
FLOSS-nnn.mp3
changed around episode 95
between 92 and 93
so curl -O 'http://www.podtrac.com/pts/redirect.mp3/twit.cachefly.net/floss[0092-0122].mp3'
so curl -O 'http://www.podtrac.com/pts/redirect.mp3/twit.cachefly.net/floss[0093-0122].mp3' then
and
curl -O 'http://www.podtrac.com/pts/redirect.mp3/twit.cachefly.net/FLOSS-[001-092].mp3'
that'll download all of them
thanks
my isp is acting up again, i'm getting around 1 mbps down on the speedtests
jdoe: only large sites I know of that use it are the pirate bay and torrentfreak haha
i should be getting 30-ish
wordpress use nginx I believe
amdprophet: I get 8 kbit/s down, you shouldn't complain ;)
dialup?
nope, I live in Australia :P
oh rofl...
I have 300kbit/s normally, but once I go over 3 GB, slowed to 8 kbit/s
au: youtube, yahoo
I'm used to it, haha
youtube run google web server I though
au: i'm friends with a girl near brisbane, she has the same issue... telstra?
yep
amdprophet: well bigpond, which telstra own
basically, telstra own all the lines here
own all the exchages
exchanges*
they have a monopoly
that sucks
hopefully you'll get the NBN soon and that guy planning on filtering it gets forced out
then you'll have some nice & fast unfiltered fibre lines ;)
1
yep
stupid labor govt
labour *
:P
wait
australians spell it the american way?
people are lazy.
hah
yes it's spelt labour, but they call themselves labor I think
yep http://www.alp.org.au/
jdoe: according to netcraft, youtube run apache and yahoo run 'yts'
dunno about netcraft, I assume lighttpd's site isn't lying.
it doesn't mention yahoo, I forget where I read that.
maybe outdated
I remember youtube ran lighttpd
but then their was 502 gateway error I saw one day
now it just seems to be apache
arpnetworks - Apache/2.2.11 (Ubuntu) Phusion_Passenger/2.2.4 PHP/5.2.6-3ubuntu4.5 with Suhosin-Patch mod_ssl/2.2.11 OpenSSL/0.9.8g
boo reporting versions...
haha ubuntu
he should have his servertokens set to Prod
RandalSchwartz: so I can brag about knowing you now that I know you're famous and all :)
twit's cool, though havent' listened to it for a bit
amdprophet, I think you mean infamous :D
lol