@JourneymanGeek WOW, how that NAA survived almost a year????

@JourneymanGeek "chui Nee Asilodcuo ren chea" Translation bout the only thing it good for is getting chicks into the tub.

lol I started using Chatroulette and wanted to start looking for groups but this web application just located somebody from Ukraine and started showing his webcam session as he was doing a bad habit (ahem...). Of course I disabled mine along with microphone. Oh shit you feel shy suddenly looking on somebody live and talking.

@Boris_yo You mean you shy away from dudes holding thier private parts in chat roulette/.

@Psycogeek No. What made you think so?

lol they ask me to verify account by SMS. The heck with them! I ain't providing my mobile number for spamming.

Ping to ajax.googleapis.com (74.125.196.95) is extremely laggy. How can I troubleshoot?

@ThatHelpVampireGuy: pathping?

I'd also probably try a proxy/ssh tunnel to rule out your ISP

@JourneymanGeek No package by that name on apt.

oh, its a windowsism

@JourneymanGeek Ahn... Huh? Is that those VPN things geeks talk about?

linux equivilent is mtr

ssh tunneling? kind of

My lousy cheap router eats all my traceroutes and burps asterisks only.

Will that prevent me from mtr'ing?

that might

Yeah, no answer.

But at least I could load SE.

Whole day it wouldn't even go past "connecting ajax.googleapis.com"

ow

On different news, I'm still trying to build my DIY NAS. I finally managed to get the damn FreeNAS on a flash drive, it boots on my machine, but apparently it needs a wired connection and I got only wifi.

ow

speaking of which

I need to dig my old atom box out of storage to see if it still runs

I managed to get Ubuntu to temporarily run on my NAS box, but it boots so slooooooooooooooooooooooooooooooooooooooooooooooooooooooooow

I tried on another PC today and it is still slow. Maybe it's because the flashdrive is formatted on NTFS?

Would it slow it down?

i guess it might

Getting a linux-based OS to boot from the flashdrive is an art almost as inexact and occult as getting browsers, video tags, containers, codecs, resolutions and bitrates just right.

@JourneymanGeek If you tunnel, you are still connecting through ISP.

On some computers, it will boot only if I preformat. On some, if it's FAT32. On others, if it is anything but FAT32. Sometimes it has to be NTFS. Sometimes I need to use unetbootin, sometimes LILO on Windows, sometimes Ubuntu's own tool...

not really

all your traffic is through the other system

For any DNS wizards out there, or anyone who knows the Plesk panel, I'm in a dilemma with ports and nameservers. I have a question posted, if you want to check it out. superuser.com/questions/669015/my-nameserver-isnt-registered

@jflory7 wait, so the nameservers provided by OVH are controlling your actual A records, CNAMEs, etc?

you bought the domains through namecheap -- do they not provide a control panel so that your domains are answered by their servers?

Yes, they let me manage all of my DNS settings from my control panel.

@jflory7 and what IP are you trying to put into namecheap as the DNS server to forward requests to for your domain?

the OVH IP, or your server?

And Namecheap does provide nameservers and DNS management, but I'm hoping to set up email accounts on my dedicated server with my domain.

@jflory7 running your own DNS server is not a prerequisite for hosting email

yup

Plesk is probably running on your dedicated server, so I'm a bit confused by using Plesk on your dedicated server to tell OVH's nameservers to take responsibility over your domain....?

you just need to set up mx records

I'm not too clear on what stuff is running on what within OVH -- you've got your dedicated server or VPS or whatever, which is "your box" -- let's call it IP 1.1.1.1 -- then there's another (two?) DNS servers run by OVH, which you do not have exclusive control over, but you are able to tell those servers to host your domain, IPs, say, 2.2.2.2 and 3.3.3.3 ?

For Namecheap, I'm just trying to change the nameservers for the whole domain. I have two nameservers that are provided to me by default to use. I already have the nameservers in place, so I was hoping I could just figure out how to get them working again. // And yes, Plesk is running on my dedicated server. I have a primary and secondary nameserver – the primary one is the one that points directly to my server and is the one that is also giving me problems. The secondary one is the one […]

provided by OVH.

er.... but.... I guess I don't see how having Plesk installed on your dedicated server lets you control the DNS entries on another server owned by OVH, unless they have some custom software going

if you're trying to point Namecheap at OVH's DNS servers -- which are not the box you are renting from them -- and configuring your DNS management within Plesk on your dedicated server, that's... not going to work

@allquixotic Oh, wait, I don't think I was clear. I'm renting a dedicated server from OVH. The nameservers are for the same box.

Everything is restricted to just one box that I'm renting from OVH.

@jflory7 the DNS servers provided to you by OVH are almost definitely only for outgoing queries originating from your server, intended for the server to use to resolve external domain names

I think when you said "nameservers" in the sentence above, you really meant "domains"? unless you mean there are two IP addresses e.g. 2.2.2.2 and 3.3.3.3 which are not on your box that are also DNS servers...

it would make sense to me if you have one box and you're trying to run a DNS server on that same box and that DNS server is intended to answer queries for two domain names

that configuration makes sense, and is valid

@allquixotic I think I get what you're saying. I know the secondary nameserver (i.e. sdns1.ovh.ca) wasn't really relevant and in the help article I had read, and it was the primary nameserver (i.e. nsXXXXXX.ip-xx-xx-xx.net) that was relevant. // Yes, that is what I'm trying to do – sorry for not being too clear.

@jflory7 OK, so what you need to do is obtain the public IP address of your dedicated server, the IP address that you SSH into, and tell Namecheap to use that IP address for both of the domains... then you can use Plesk to configure your DNS settings

but most likely you will need to install a DNS server and/or open a port in the firewall first, judging from your question

and don't worry about a secondary DNS server -- hopefully they don't require you to set one -- there should only be one server that you're pointing Namecheap to -- your box -- nothing else

So just create an A record for the domain to point to my server? Alright. It's just the thing that's weird is I changed the nameservers for one domain, but when I did it for the second, it gave me that error saying it wasn't registered.

Like, for one domain, I just changed its nameservers in Namecheap TO the nameservers in my server. When I tried doing that for my second domain, no bueno

@jflory7 honestly, unless you think there is something wrong/unreliable with the namecheap DNS servers, I really don't see the point in making this so difficult -- you can just create your MX records on namecheap's DNS management web site and that's it

@jflory7 I'm surprised that worked at all -- as I said above, the nameservers that are in the /etc/resolv.conf file are not for you to use for controlling your domains -- at least I don't think so, unless OVH specifically said that, and then they'd have to have some separate management panel where you can configure them

your Plesk install on your server isn't going to tell OVH's central DNS servers what names to resolve and how -- it just isn't

your Plesk can only tell your own dedicated server that you're specifically renting how to answer DNS queries

besides, unless your server is pretty darn beefy, you're going to chew up a significant bit of resources (bandwidth and RAM mostly) running your own DNS server, especially if there's some kind of DDoS ongoing against the domain name system -- most people prefer to have their names hosted on large round robin nameserver providers with a geographically distributed network

It's more so just for the convenience of managing my domains from one place and just using the bells and whistles that Plesk provides for having a domain registered in there. I'm not doing anything really major with them. I created this images to see if they help better illuminate what I'm asking, or if it's what you already are saying.

@jflory7 okay, that's weird... so the something.ip-192-95-30.net is your server?

also, AP European History :)

graybeard back in my day, we didn't have websites for our high school classes! we learned entirely from a giant book, which we had to carry 15 miles in the snow with no shoes -- up a hill both ways!

@jflory7 I'm confused, because it says that crystalcraftmc.com is registered with "InCapDNS.net", which is owned by GoDaddy

that seems to conflict with the claim that it's owned by Namecheap

let me walk you through it: the IP that comes back for crystalcraftmc.com is 199.83.134.131, which if you perform a reverse DNS lookup on it, you get something.incapdns.net, which when you look them up in whois, gives you godaddy.

Yes, that's correct. And the reason for that is because I just transferred from GoDaddy to NameCheap two days ago. // Hahah, well, if it consoles any, we still have our giant textbook!

that's not even an OVH server

@jflory7 well it may be that you may have to wait some more time before the transfer is complete, because you may not be authorized to control the nameserver records for crystalcraftmc.com yet

from namecheap's side, I mean

@allquixotic I already have them changed to another set of nameservers, so I know that I am able to edit and change them. It's just that this one nameserver in my panel shows up as "unregistered" and Namecheap won't take it. The OVH rep. said that my port 53 was closed, but I have no idea how to reopen it.

@jflory7 well you can troubleshoot your port closure in a number of ways -- check iptables -L and netstat -pan and ps -ef

try to see what your INPUT chain policy is in iptables; see if any DNS servers are listening on port 53 in iptables; and run ps -ef | grep ns to see if something like dnsmasq is running

if it's not running for some reason, you may want to start it

@allquixotic For INPUT chain in iptables, the policy is accept. I'm not seeing anything for Port 53 anywhere, and same for dnsmasq.

Let me put it in a paste

@allquixotic dev.bukkit.org/paste/sxfb0h7apui5rvt9

@jflory7 you do have named running which is a DNS server, but I don't know why it isn't listening on 53

linux.die.net/man/8/named

check the dns server configuration steps for either Plesk or your underlying OS

something's hosed

o wait.

tcp        0      0 192.95.30.137:53            0.0.0.0:*                   LISTEN      11452/named
tcp        0      0 127.0.0.1:53                0.0.0.0:*                   LISTEN      11452/named

so it is listening on port 53

Hmmm - so what do you think that means?

@jflory7 either the guy is lying and it's not closed (you can check with nmap), or it's closed because named isn't configured properly and is closing the connection with an error

@allquixotic or he's not checking udp ;)

wait what? why is tcp:53??? isn't udp:53??

err, ignore the stupid clickbait pic

the link is to a funnier thing

o0

@allquixotic Seal!

anybody here?

we have like 20 PCs in work and one router

is it possible to connect to router via some web interface from any single PC?

the internet keeps falling down so the technician from a company who is responsible for our network sent us an email that he will connect to us via TeamViewer

and will make changes on router

so that the internet works 24/7 in our network

we don't know to which PC should the technician connect to via TeamViewer

all PCs are Windows 7

@EinsteinsGrandson Set up an isolated virtual machine for him

yeah, thats probably a good idea

@EinsteinsGrandson I think this is the default when dealing with routers, no?

@JemTay Hello, new user! This is Root Access, a chatroom for Super User, and I am this channel's helpful chat bot. You will need to earn 20 reputation to chat; meanwhile, if you need help, please see the Help page for Super User.

@Mathev Hello, new user! This is Root Access, a chatroom for Super User, and I am this channel's helpful chat bot. You will need to earn 20 reputation to chat; meanwhile, if you need help, please see the Help page for Super User.

Why are we so unusually quiet today?

Is it some internet holiday today or something?

worse. Its monday

this monday feels like saturday to me :D

At least you guys aren't in an almost-drunk state of dizzyness like me.

I had a great saturday; in fact I only woke up sunday at 2PM =P

But now I can't focus at work

I wanna go home!!!

as long as it dosen't feel like friday with chinese food its ok

And yes, I just earned myself a special place in hell for that.

and to make up for that

@JourneymanGeek The video effects on the first video's start are priceless.

No, really, no one should have to pay a cent for that.

@ThatHelpVampireGuy: I am really wondering if its some ironic commentary on society and fame.

this stuff is just so.. perfectly bad

like a snowflake frozen in a moment, just before it is vaporised by two trains crashing, while a shark jumps over it.

o.o

Did you ever get your hands on a Model M, @JourneymanGeek?

@JourneymanGeek There's a mock video by her and some talk show host saying exactly that. Can't find it.

@Bob: alas, not yet

@JourneymanGeek ...haven't heard that second one. What the fuck.

@Bob: its horrible enough to want to hide under the bed.

@ThatHelpVampireGuy Looks like they ran it through one of those basic image filters. Just the raw output.

Ok, that's the keyboard ordered. Should be here before Christmas, with any luck.

(amusingly the BW is the oldest part of my desktop. It was a VERY strange build when you get the keyboard, then the monitor, then the case, then everything else)

@Bob: what did you get?

@JourneymanGeek The Kogan one. Not really a branded one, but it's a decent price for its features.

ahh

And I'll be able to compare the noise of the browns with my current keyboard :P

I expect my current one to be louder.

lol

browns are reasonably quiet

and blues arn't as loud as a truely crappy keyboard ;p

@JourneymanGeek BEHOLD:

Eh, my keyboard is ridiculously loud. Especially for a membrane.

I do like it, though :P

Maybe just familiarity, but at least the keys don't fell gummy. Quite a few brand-new membrane keyboards do :(

@JourneymanGeek Or a M :P

lol

@anteatersa Welcome to Root Access chat for Super Users! I am this channel's helpful chat bot. Please don't ask if you can ask or if anyone's around; just ask your question, and if anyone's free and interested they'll help. For bot commands, type !!listcommands

I WILL get one eventually

Ooh, I just looked at the order details - apparently it comes with a keycap puller :D

@JourneymanGeek True M or replica (Unicomp?)?

unicomp

same tooling actually

IBM spun off lexmark, who sold it to unicomp

o.O

what's a "USB 2.1 hub"?

amazon.com/Das-Keyboard-Professional-Mechanical-DASK3MKPROCLI/…

@ThatHelpVampireGuy is a bottle drinking another bottle, cannibalism ?

o0

@JourneymanGeek And bullying too! He throws it on a garbage can!

lol

o.o

jbhifionline.com.au/computers-laptops/tablets/…

apparently it's more or less the same as the N7, spec and performance wise

@JourneymanGeek Any favoured mice? :P

Of all the random accessories, that's probably the thing I need to replace most. Mouse wheel click's dying :(

> Two-port USB 2.0 hub allows for syncing and charging iPhone, iPod and USB compatible devices

maybe?

@Bob "A hub that is slower than USB 3 for 500 Alex"

I wonder if there are any browser or website settings that can help reduce or eliminate the incidence of leaking information due to the successful execution of a reflected XSS attack

the situation is: a web app accepts an HTML form field (text box), stores that in a database (SQL injection is not possible, thankfully), but the field's value gets reflected verbatim back into a different webpage in the same site, and can be interpreted as code, meaning that putting the right value into the input field means that somewhere else in the app there will be arbitrary javascript and html executing in the page context

from there it's pretty trivial to create an <img> or <iframe> tag (using document.write in JS) with a URL to a remote site, cross-domain, and including document.cookie in the query string to the remote site -- and suddenly the browser has given away your cookie

can the original site code or the browser do anything to prevent this that doesn't completely break the web? for one thing, cross domain <img>s are widely used and accepted practice, so you can't just disable those

I mean, aside from actually fixing the XSS vulnerability....

@allquixotic Sure, you can define cookies as not accessible via JS.

@Bob that sounds like something the browser would have to do -- how about the vulnerable site?

@allquixotic No, that's something the site does.

oh?

An HttpOnly cookie (as set by the site) cannot be accessed via client-side JS.

oh, right, I forgot about that flag :o

Obviously the browser controls this, but the site would have to set it when setting the cookie.

=> en.wikipedia.org/wiki/HTTP_cookie#HttpOnly_cookie

> The HttpOnly attribute is supported by most modern browsers.[17][18] On a supported browser, an HttpOnly session cookie will be used only when transmitting HTTP (or HTTPS) requests, thus restricting access from other, non-HTTP APIs (such as JavaScript). This restriction mitigates but does not eliminate the threat of session cookie theft via cross-site scripting (XSS).[19] This feature applies only to session-management cookies, and not other browser cookies.

I'm pretty sure our site doesn't set HttpOnly cookies -- we looked into that a while ago -- not setting it basically acts as a precursor to other attacks, which is the state of where we are right now

it's a real mess, because right now there's a sort of exploitable synergy between a bunch of seemingly pointless vulnerabilities on our site

Generally, the idea is to have it always set except when you need JS access to the cookie. Well, that would be the secure way to go about it.

the more vulns you have the more using them together can become a nightmare scenario

or just fix the XSS vuln... there are other potential problems with that.

well it is an HTTPS site so removing HttpOnly from the cookie would be hard, you'd have to MITM TLS

Wait what?

I mean for an attacker to remove HttpOnly from the cookie

HttpOnly should not care if it's HTTP or HTTPS.

Ah.

Yea, you'd have to hijack the HTTP session. Client-side scripting, by design, can't touch it.

hopefully we can get at least a subset of the vulns fixed before production -- a subset such that the remaining vulns aren't exploitable in practice

codinghorror.com/blog/2008/08/…

@allquixotic Any reason you decided not to?

@Bob oh. I have absolutely zero control over what the devs do

@allquixotic Well, do you know why they decided against it? :P

if they decide not to do something or put it off, it's probably because there are so many layers of bureaucracy to cleave through to get it done that it will take a year or two

Ah.

I don't think they decided against it -- they probably decided that it would be politically unpopular for their career to attempt to tell the people who can enable it that they should enable it

On a small project without any politics and bureaucracy to worry about, it's a five-minute job to set it and maybe an hour to make sure no JS actually needs the cookie.

on a small project it's much easier to do security, because you don't have 5 different, mutually hostile offices that are in some way responsible for security or security theatre, and none of them likes being told that their stuff is insecure or that they aren't doing their job right

@allquixotic I'd be surprised if it wasn't on one of those checklists for securing a server/website.

@Bob oh, it certainly is

we already tested for it previously :P

I remember looking through one of the DoD declassified ones for application servers a while ago. Good lord that was extensive... though it seems to deal more with human procedure and access than the programs being run.

but because of all the political interlocks, knowing about a security vulnerability is only about 5% of the battle for us -- for most smaller and more agile teams, knowing about a vulnerability is 90% of the problem, and fixing it is only 10%

@allquixotic Ah, yes. Let's argue about it and throw blame around rather than fixing it -_-

I'm so glad I don't have to worry about security for what I'm currently doing.

@Bob more like, we're afraid of those guys over there and what they could do to us to make our lives miserable, so we probably shouldn't bring this up with them

even though we're all within the same overarching organization, it's amazing how much hostility there is... you know about those stories where a security vulnerability researcher (or just a curious random person) reports a vuln to a big company, and as "thanks" they get slapped with a lawsuit, injunction, etc?

the culture seems to be: don't point out other peoples' mistakes, even if they sorely need to be fixed and pose a significant risk

silence, at least, doesn't make anyone else uncomfortable

@allquixotic Which seems all well and good until someone maliciously attacks and suddenly you've lost everyone's credit card info.

the implication being, if you don't point out their mistakes, they won't point out yours, and everyone gets to look good in front of their management

(Or even worse in your situation, since you're working with some rather sensitive data - hellooo identity theft!)

but that's so dysfunctional -_-

@Bob yeah, and the sad thing is, many times more heads will roll after there's some kind of breach than would have rolled (if any...) if the issue were addressed candidly and fixed when it was discovered

the culture needs to be, "yay, we found a vulnerability! fix it; be glad it wasn't exploited; praise the people who found it, and don't hate on the people who left the vulnerability in their code/server/configuration, because we're all human and we make mistakes, and we have these people checking after them for a reason"

lol:

(that was from MSDN: msdn.microsoft.com/en-us/library/ms533046.aspx)

@Bob problem being, in my case, the attacker can effectively obtain the "username" and "password" (I use those terms very loosely) that would allow them to login as the user, just by passing a remote server document.links and scraping values from certain pages

so the cookie is not strictly required, as you can essentially get all the info you need through other means than sending document.cookie

I'm pretty sure the only way through this is to get the XSS vuln fixed

they claim it would be very difficult to fix the vuln even assuming they were able to jump the political hurdles... I'm not quite sure what the technical difficulty is, but it may have something to do with making sure the site actually renders properly after it's fixed

@allquixotic o.O

That's... more than slightly broken...

@Bob don't get me wrong, they'd still have to use CSRF to actually navigate to those pages as the user; they can't just GET a URL from any old browser without having the user's cookie

but assuming they could get XSS again on the pages that provide the requisite info, they could really get everything they need

I wonder if it is possible to continue navigating to other pages within the site while "passing" your vulnerable XSS context from one page to another?

without using CSRF

with the premise being: you have successfully executed javascript on a given page of the app, and you can't use CSRF, but you want to execute arbitrary javascript on at least one other page (which requires a full page load and either a POST or a GET to a known URL) in order to obtain additional information

oh... I know how you'd do it!

within your compromised context, make iframes with the src set to the place where the info you need is contained

you can then look inside the iframe from outside, or vice versa, because it's within the same domain, so the SOP does absolutely nothing to secure the user

so yeah, HttpOnly buys us nothing

once the (stored/reflected) XSS is performed, whether it's through CSRF or some other means, all bets are off

the XSS has to be fixed

(the CSRF too, that's really high priority)

:\

heck you could even use XMLHttpRequest with a POST method if GET isn't good enough for you

it's the same domain so the browser would happily execute the request

and probably no other way to fix it, either

right

the iframe one was a maybe, but any HTML attributes wouldn't be set by the script anyway, this has to be controlled by the browser, whether locally or via a HTTP flag

(the iframe maybe: stackoverflow.com/questions/13247798/…)

I think the discoveries I've made in the above wall of text will help me help my coworkers build a case that we need to get this fixed, even if reputations or political brick walls are in place

the people who have to fix it really won't like it, mainly because it sounds like it's going to be actually difficult (on a purely technical level) for them to fix it... lots of code changes

@bmike Welcome to Root Access chat for Super Users! I am this channel's helpful chat bot. Please don't ask if you can ask or if anyone's around; just ask your question, and if anyone's free and interested they'll help. For bot commands, type !!listcommands

Pro-Tip: Always install windows 8 on your hard-drive from an old machine. It ensures you get the classic boot style so windows 8 has no ties into your hardware and doesn't complain about grub or anything, also it doesn't take 3 extra partitions

glad I ugpraded to Windows 8 before upgrading my Mobo

@ChatBotJohnCavil ha - boy am I glad you're here to welcome me. feel free to let me know if you'd like any tips on human-computer interaction or how to deal with sarcasm.

I want adamantium exoskeleton...

@JimmyHoffa you mean you'd rather boot via BIOS than UEFI?

that's a mistake IMHO, especially given that UEFI GOP + Secure Boot + IRST boots from POST to login in about 8 seconds

@allquixotic I'd rather boot via MBR, it's easier to control; The BIOS is UEFI and let's me make easy choices about which MBR

@allquixotic I hardly ever boot. If I turn the computer off it's generally for reason, otherwise I sleep it which is extremely fast boot as well

@JimmyHoffa what's so easy to control about it that isn't easy to control about UEFI? your UEFI firmware itself should display a UEFI boot menu of all the OSes you have installed on your hard drive; in fact, you can install Windows Loader and GRUB on the same drive and choose which to boot by pressing a key during POST, eliminating the need for chain loading and all that old cruft... you can even install GRUB and Windows Loader into the same EFI System Partition

they seriously thought this UEFI thing through pretty well, and except for a few annoying UEFI bugs in various motherboards by sleazy manufacturers, the overall execution of it is fantastic -- there's really no reason to stay on the old guard anymore, it's just cruft and legacy and limitations that you really don't need

@allquixotic Part of the reason I'm avoiding it is how quickly it's supposed to boot too, that will supposedly make it extremely difficult to get my f8 in for safe mode I've heard

@bmike you're just meeting Cavil now, eh? (he's my bot, been here a long time too)

@allquixotic Say my UEFI partition gets slopped, can windows boot off old-approach then? Do I have to reinstall windows UEFI then?

@JimmyHoffa there is no more "F8 for safe mode" on Windows 8 -- it doesn't exist

@allquixotic wrong :) it's just hard to get to

I required it though, the nvidia driver built into windows booted to a blank screen and wouldn't give me any video at all

had to figure out how to turn on safe mode so I could get in and swap the driver out

@allquixotic It's just off by default, you can turn it on with bcdedit from the repair command prompt from a windows 8 install disk

@JimmyHoffa you can reinstall the Windows Loader into the EFI System Partition either using the Windows recovery command line (from a USB stick or DVD) or the "Reset my PC" feature

@allquixotic "Reset my PC" uninstalls all your software though it says

@JimmyHoffa well you've already used bcdedit once, so I don't see why you couldn't use it again, just to reinstall Windows Loader this time in case you accidentally bork it up

my current way has boot + system all in the same contiguous disk space, so the whole thing would have to go to hell, and it doesn't depend on the MBR in that way

I could jigger MBR any way I want and it won't break my windows bootability, MBR from a different drive could boot to that windows if I wanted

but when windows installs itself with UEFI, it sets up the installed OS to not be able to boot from any other way

UEFI could boot my current windows OS or MBR could or a grub rescue prompt

GRUB rescue prompt can't boot a UEFI installed windows OS

I've never been able to get a GRUB rescue prompt to boot any Windows OS, unless it gets to stage 2... the rescue prompt (can't find stage1.5 or stage2 binaries) can't find the disk or load the NTFS driver or whatever it needs to root (hd0,0) and chainloader +1

thing is, you don't need GRUB to boot from the Windows partition if it's installed via UEFI

you just tell your BIOS, "Hey, I'd like to boot Windows" during boot-up... probably F11 or F12 key

@allquixotic It's actually a nice touch.

Is there a FAQ on how to make one? It might be fun to add one to the chat I inhabit mostly from time to time.

!!tell bmike help

@bmike github.com/Zirak/SO-ChatBot/wiki/Interacting-with-the-bot

my fork is at github.com/allquixotic/SO-ChatBot and github.com/allquixotic/so-chatbot-driver

@bmike The simplest way is to run the bookmarklet.js inside your own browser (recent FF/GC).

@allquixotic has a headless build if you prefer to run it on a headless server

@Bob that'll get you up and running so you can see the bot, of course, but then you're tied to keeping that computer/browser running 24/7 if you want the bot to be stable

@allquixotic touché

@bmike what @Bob is saying is correct -- basically you can try out the bot in your browser, but personally, if you want it to be "dedicated" in the channel all the time, if I were you, I'd run so-chatbot-driver, which makes it effectively(*) headless and able to be run on a dedicated server