@Braiam I like the concept that specific technical requirements could be an opinion. IMO motherboards need a power supply plugged into it, but that's just my 2 cents, ymmv
@Braiam I completely agreed with your opinion, it's just funny to hear that sort of thing said as opinion, it's like saying IMO american wall sockets are usually 120v AC
@Braiam Ah you're from dominician republic (I enjoyed a very nice quiet time in Bayahibe a few years ago, beautiful country you've got there), english is your second language?
may need to set such as your IPv6 gateway, but my networking skills are based upon ability to flip switches and toggle toggles until something works, so I'm a bad advisor on the topic
@Braiam It would make more sense to say "I think [what you believe but are uncertain of]", opinions are relegated to things that can't be true or false like "IMO I'm an asshole" it can't be objectively true or false, it's completely subjective, or you could use opinion like you did above and it just makes your sentence sound a bit more creative
@Braiam you're lucky to live there though I gotta say: Driving through Santo Domingo was like driving in an arcade game. That was pure insanity, one of the most exhilirating, scary, hilarious things I'd ever done in a car.
@Braiam Luckily I was only passing through, from la romana area up to jarabacoa, but yeah, I could not have expected that, I've driven rentals in other Caribbean countries and never ever seen anything like that before
@JourneymanGeek Huh? I'm given to think I don't really know much about IPv6 (which is true), because I'm unfamiliar with why IPv6 would change the reachability of devices behind a NAT?
@Braiam This is good info! I never knew the IPv6 spec said anything about removing NAT
though it makes sense, hard to make a cause for making people pay for extra IPs when there's enough of them to address every molecule in the universe 25 times over (yes, I did just make that up)
I'll be able to remote into my WDTV live while I'm at work from my phone and mute/unmute pause/play/rewind whatever cartoon my kids watching to screw with him and by proxy my wife
@Sammy I think I had a USB Wireless adaptor once that did client mode.. but it was USB it didn't go into a wired connector.. So i'm not sure you want client mode. But what I had in mind, was a regular access point. or wireless router. But connect to it with a cable.
@JimmyHoffa First you have to remember which IP is the WDTV's screen, which is the power supply, which is the remote, which is the network cable, and which is the tuner.
@Braiam It's really just self-destructive behaviour because I'm the one they'll be bothering when it acts weird, like the other day when the batteries in the remote died and she asked me how to get something to play without the remote and my only response was "Hope"
I don't think the purpose of IPv6 was to eliminate NAT. rather, the purpose of IPv6 was to allow anything that plausibly needs to have a public IP address all on its own, to have one
there are many cases where things that could benefit from having a public IPv6 address today, don't, and we have to do complicated things like port forwarding
but I think if the approach is to deliberately put things on the public Internet without being NATed, rather than making that the default, it will work out better
there are a lot of systems that run worryingly insecure software and are fine with that primarily because their LAN IP address is non-routable from the public Internet, and the router doesn't have port forwarding to their box
@JourneymanGeek from a network security perspective, there's really no good reason to make a box publicly routable on the public IPv6 Internet if it's only, say, hosting a file server on the LAN
going by the principle of least privilege, it would be safer to, by design, architect the network in a way that the very nature of the IP allocation to the box prevents it from being publicly routable, except in the case where the router chooses to forward outbound requests and send back incoming packets as part of a bidi data flow
@terdon Ahhh. Can you answer one question for me about France? Us Canadians are told that we get yelled at for attempting any Canadian French over there, if we're in the North... Is that true?
how many web servers are there hosting internal applications that listen on 0.0.0.0 and ::, that, if they had a publicly routable IPv6 address, could immediately be compromised?
@allquixotic: the main reason I want this is I need one of my systems accessible both inside and outside my lan, without needing to futz with two different ip addresses
@Hennes .... actually it's a great firewall. Blocking all inbound connections is an awesome start for a firewall, if you don't need something more powerful.
@Hennes by "firewall", you mean "explicit deny rules on inbound ports", which most systems don't provide by default. publicly routable IPv6 for devices that are not configured to default-deny-inbound will get absolutely destroyed if someone discovers their box is routable and port sniffs it
@CanadianLuke do that :) But really, if you have any French, speak it. The English of the average Frenchman is atrocious. Like the Spanish and Italians.
@DarthAndroid well, the semantics matters and it's technically not considered a firewall. but yeah it is good security particularly if one has no hardware firewall
it would've been nice if all IPv6 routing hardware (and software) since the beginning of time shipped by default with a firewall rule out of the box denying inbound connections by default
but that isn't the case, and there is still new IPv6-capable hardware being shipped today that allows everything if your ISP lets you have a public IPv6 address
I also get hung up on the terminology of "firewall" -- any layer 3 router that is capable of being configured by the network administrator to selectively block connections based on some ruleset is technically a firewall, but you can configure (or fail to configure) a firewall in such a way as to allow all traffic, which is not secure
when people say "you need a firewall", what they mean is that you need a firewall configured in such and such a way
but seriously, to maintain backwards compatibility with the user expectation of making all inbound ports non-routable out of the box (which is the effective behavior that NATv4 provided, with or without a firewall), networking hardware vendors are gonna need to lead the charge
perhaps when huge ISPs like Comcast and Verizon turn on IPv6 for all residential customers, they should have a 90-day period where all inbound connections are blocked at the ISP level, unless you call them or login to a website and opt-in to opening it up
and then advertise the hell out of "you need to secure your router"
I mean, if they just turn on IPv6 without any real user communication, there are going to be hundreds of thousands of SMB/CIFS fileservers running on peoples' old boxes and laptops that are going to start being routable to the public internet
@DarthAndroid I dunno, I'm pretty sure that most modern modem+router combo devices for DSL, FIOS, etc. are fully ready for IPV6 out of the box; the only thing they've been waiting for is for the ISP to start providing a DHCP6 server for them to get IPs from
god forbid you buy one that was manufactured in the last year or two -- the chances of it not supporting IPv6 at all are essentially zero, although the chances of it being fully disabled by default (some drop list in the web admin page "Enable IPv6" set to "Disabled/Off") are higher
I'm observing a fairly odd behavior in one of our webapps: we have three HTML form fields: two radio buttons and a text box... the text box will be visible if one radio is selected, or hidden if the other is selected... on IE 8 (don't ask), if you type something in the text box and then click the radio to hide it, then submit the form, in the HTTP POST request, the value of the text box in the form is cleared
but in Firefox, the text box's value that was set before it was hidden is retained
I'm going to go with Firefox as probably having the right behavior here, so wtf is up with IE
But someone on the forums said it was because of some bug report. I read the bug report and it was like "i had problems, then I disabled this setting and now the problems are gone"
@allquixotic it's an enabled vs disabled flag on the control if I recall. I fixed that bug a couple times last year, some people use the wrong Dom flags to hide controls and browsers rightly don't post field data from fields in those states. There's other appropriate ways of hiding or disabling fields that will retain the field data in post
Don't remember the precise wrong and right states. Alternatively the code cleared the field value when it hid it
@CanadianLuke Regarding your comment on my question, do I need to change something? The share-specific directives are different from the global ones and I put the force ones in the share section because the global section is generated by FreeNAS and I didn't want to mess with it too much
My Samba knowledge is limited to Google... Sorry. We run Samba as our domain controller too. I'll PB our smb.conf file, and it seems to work for us. I'll let you do with it as you will
@allquixotic Nah, but google knows well about that issue. Just look up form post didn't include form field or something. Generally you said CSS flags to change visibility/etc in a safe way as opposed to things like flipping enabled/disabled or read-only which are element level attributes. The CSS attributes should simply affect the display and not the browsers handling of the element.
It's a common issue, there's lots of causes and fixes varying from "Well it shouldn't post that data, but you need to check on the server if the field was posted before just naively updating the field with a null" to "It really needs to be in state X so you need to stuff the value into a hidden field which you use the value from on the server side"
@allquixotic Yeah like I said, it's a common bug for many obvious reasons. It's fixable though, just throw it to the dev with whatever info you can imagine and it's not difficult to fix if he's remotely OK.
alternatively if you want out of QA fix it yourself, send him the patch with the bug ticket and ask him to review it and check it in if he like's the fix.
I think what they can do that should be recognized as a browser-independent resolution is to use the fact that a form field set to disabledMUST NOT be considered successful
so you could have an onclick on the radio button that hides it to basically set the field to hidden style, then disable it
and on show, set it to enabled, then un-hide it
by setting the disable after the hide, and the enable before the show, the user will never have a chance to see the form field get rendered on screen as disabled
but the behavioral effect will be for any standards-compliant browser to not post the field
@JimmyHoffa actually the desired behavior is for the hidden field in question to never post to the server.
don't worry; our server code is already anticipating how to gracefully handle the case of the user shoving it down the server's throat anyway using an intercepting proxy or something
but client-side it just makes sense to not submit it at all if we aren't going to process it
over a large number of users and a large number of fields with significant text in them, it could make a minor performance difference... perhaps... especially since this is HTTPS
@allquixotic That tends to create server-side complexity because you don't know if they set the value to blank or it wasn't posted so you have to check if it was posted at all before handling it. And if it's multiple fields you now have to do it per-field. There's various settings some frameworks have that will allow you to make the page always post the old data if it doesn't have a new value etc.
@JimmyHoffa well, what we do is read the values given for the radio buttons, and depending on which radio button was selected, determines which conditional form fields we care about
if a value is given for a form field that we know, server-side, was hidden, then we just ignore that value
and if the user specified somehow in the url-encoded POST body that both radio buttons were selected, using an intercepting proxy or so, then we just throw them an error
obviously a standards compliant web browser won't let two radio buttons with the same name be selected simultaneously, but you can still modify the http request and break the rules of DOM
@JimmyHoffa crud; I just tested something and IE 8 is submitting the hidden field now, in a trivial thing I wrote that doesn't use our app's framework or anything, so I'm thinking the odd behavior difference between IE 8 and Firefox is related to clientside JS being loaded/not loaded by one of the browsers :(
@allquixotic Anybody breaking rules of DOM is never worth even thinking about though, you just catch them and redirect them to youtube and forget that some arsehole exists trying to screw with things
@JimmyHoffa they're worth thinking about from a security perspective because the server needs to anticipate all possible inputs when thinking of the client, essentially, being in control of an arbitrary bitstream being sent to your server
so you can't trust the client to do things in a sane or even pseudo-sane way based on any kind of standard
the important thing is to be able to detect these "normally impossible states" like having two mutually exclusive radio buttons checked, and to handle them gracefully
not even attempting to detect them can lead to security problems down the road
@OliverSalzburg Because I'm looking for places to flag comments... when I get bored I troll around chat until I find a blue person that says they are bored. Then I fill up their queue with obsolete comment flags.
@Undo you'll probably have pretty good luck with flagging comments on questions with either substantial upvotes, or several answers, or both... though I'm sure you already knew that
I'd point you to one of my recent answers that had a ton of comments, but they were already removed
@DarthAndroid we still created it, but as part of the ToS of agreeing to participate in the site, we promised SE that we'd agree to the CC copyright license for our content
since it's still implicitly copyrighted by the respective author, that means that, if you wrote something on SE, you could make a copy of it and re-license that copy under any other license (no matter how restrictive), and publish it in any way (no matter how restrictive), but you wouldn't be authorized to retroactively tell someone who read your CC-licensed content that they are now violating your copyright
@Braiam can't you just click on a message to see it in the transcript, then from the transcript, reply?
@Braiam reload the page, you might.... the Force JS might be acting up
Does anybody know how the polish bottle au-de-burning-nose in their mustard? I think this would make a spectacular energy drink/sausage marinade combo. I fucking love it.
I've always wondered about the merits of trying to game on a dedicated server: buy a dedi chassis with Xeons, SAS, hardware RAID, FBDIMMs, etc.; put Windows Server 2012 R2 on it; install either a FirePro or Tesla which happens to have HDMI or DisplayPort out; and install games
half the games would probably be like, "wat? I don't know wtf a Xeon is, or a FirePro, so I'm going to pretend you're running a PC from 2005... or maybe I'll crash"
@DarthAndroid I'm thinking of buying a 512 GB SSD (or maybe 1 TB if prices come down) as my primary OS boot disk, because, discounting game installs, I use way less than 500 GB of disk space on my main desktop
@DarthAndroid Have you solved the "takes absolutely forever to load games" part of the problem? That's why I went back to local after trying network share..
@JimmyHoffa when I do that sort of thing, I find myself either lowering what I consider to be "the requirements", or dealing with / living with sub-optimal performance or features, or something like that
when I shop for components, at first I tend to ignore price entirely; figure out what I'd ideally want; then figure out what's the maximum price I'm willing to pay, and try to find it a little bit cheaper than that (on sale or something) and then buy
I still think I'm getting good value, but I'm not paying bottom dollar, either
I kinda lucked into buying several components at a time when their price was low, and almost immediately after I bought them, their price skyrocketed
specifically, my 32GB of RAM and 2 x 4TB HDDs
you can't buy new or even refurbished components with similar specs to mine right now at a similar price
the prices went way, way up
haha Brandon willy has 16 rep... I bet he's super confused
@allquixotic See I do the same but then instead of deciding on maximum price I can live with, I then step it back to "Ok this is ideal, but what is the minimum performance I can live with", take the price of that, and then move forward with "Now what is the maximum I can get at that price point"
@JimmyHoffa I can't do that -- I'm not willing to live with lesser performance :P I do that at work every day on my Core i3-2100 dual core Sandy Bridge piece of silt, but that's because I get paid to do it -- if I'm paying money for the hardware and I expect to get entertainment value and satisfaction out of using it, I won't settle for crap
honestly, for money, I'd work on a Pentium III if that's really what they wanted me to do... I'd also inform them how badly that would hurt my productivity, and if they were OK with that, I'd proceed to write about 5 lines of code a day and take my paycheck and go home to my 3770K
but my desktop is really "that one thing" that I have to have the best of... everyone has a "thing"
for some people it's a car, for others a boat, for others a well-maintained, sparkling clean house, etc
I have a newish car that is functional, reliable, and not fancy at all -- it's the car equivalent of a Dell business workstation with a passively cooled Nvidia GPU and 4 GB of RAM and a 120 GB HDD
@JimmyHoffa that.... erm... well, I'll reserve judgment; maybe if you split it exactly down the middle, you'll be able to have two PCI-E x4 cards instead of one PCI-E x8 card
or maybe PCI-E lanes work like highways, and to split a road you have to build an exit lane and then make one of those clover figures and an overpass for interchange
email skype-archive AT nsa DOT gov; they will be able to retrieve a copy, and, once it has been vetted to determine that it definitively does not contain any unlawful content, will be released to your custody (but still they'll keep a copy anyway)
@JimmyHoffa one big problem with many extreme cooling systems is that, if they get too cold, water vapor starts to condense, and that's a very bad thing for electronics
@allquixotic fair point, having the cooler pad your laptop sits on start condensing water from the air causing the bottom of your laptop to be sitting in a small puddle = not good..
@JimmyHoffa I think, before long, we're gonna need to have computer systems sealed in a vacuum rather than exposed to the atmosphere, which would enable much more extreme methods of cooling / heat exchange, because there won't be any water vapor in the air around the components to condense
like, if you stuck your laptop/desktop in your freezer, it would short out because of the water vapor in there, but if all the circuitry were physically isolated from any sources of vapor, you could put it in a freezer and it'd work fine
it's either that, or the component manufs will have to keep finding new ways to increase the maximum thermal limit of components, so they don't need to be cooled as much
@allquixotic We'll just have them all bronzed, and you'll need washers around the connection points where you bolt components together tightly enough to ensure a water seal up to a depth of 100'.....
