I know, but still the point is to make the address valid. So the person looking at your question doesn't get confused by using by seeing you try to assign a broadcast address to an interface.
Anyway, can you ping your public address from the outside?
But not the other? That seems odd. Are you able to do a tcpdump on the Tomato device? If you ping the non-functioning address do you see the packets arrive?
@Zoredache: Edit conflict, I'd guess. Two of the iptables lines were running together so I changed the format of the PRE block. I didn't see any diagram during my edit.
Is there a trick to running procmon with elevated privileges from a limited account on Win7? A coworker is trying to troubleshoot an application that isn't working on Windows 7 in a non-admin account. It sure would be useful to see that stupid thing the application is trying to do.
Some questions sound like a question posed by someone new to system work, then are swarmed by people that discovered some toy to share. Anyone else feel that way? serverfault.com/questions/230592/ssh-server-on-windows
nine times out of ten that question should be answered with, "Don't do that."
It's like asking how to equip your cat with weapons to defend your home as a home security system because you have a cat already and don't want to invest in some fancy "real" security system.
Hi,
I know how to canonicalise the case of URLs on incoming request to IIS7.5, in fact, there's a built in rule template to start from. But how about outbound (without changing the code)?
This is where I got to so far:
<outboundRules>
<rule name="Outbound lowercase" preCo...
@BartSilverstrim specifically, the last 2 free /8s were handed out to APNIC earlier this week, which left 5 /8s in the free pool. Under a 2004(?) agreement, each RIR is entitled to one of the last /8s.
@BartSilverstrim Yeah, the fact that they're having to speak in their non-native language and translate technical terms on the fly is pretty frustrating :)
@BartSilverstrim I read a psychological study about those kinds of filler noises at one time, it was pretty interesting. Some kind of unconscious remnant of our brains attempting to perform data lookups or some such.
@packs yup. It's supposedly one of the hardest things for speakers to avoid doing. I try pretty hard when I give talks and I still find myself doing it.
What I remember was that "fillers" were used to keep people from interrupting us, and to keep people from thinking you're "turning the conversation over" as people think a break is a point to interrupt.
My broadcasting courses tried to teach us to adopt a sort of persona when speaking, the good ole fashioned Radio Voice (tm). The rest fell out with practice.
@BartSilverstrim The study that I remember is fairly recent, mid 2000s. I can't remember enough useful information about it to figure out any meaningful search terms, unfortunately.
The habit I thought came from trying to keep people from thinking you were done and that you hate dead air...silence is golden but annoying to us, I guess.
Unfortunately one of my more aspergian traits is a very strong aversion to hearing "uuuhm um ummmm" from speakers.
I end up deleting podcasts from certain keynote speaker events and such when people give talks who clearly should not be on stage.
If you're not prepped to give talks without constant fillers I can't stand listening to you :-)
picture a bunch of networking nerds in the smoky back room of a bar doing IP horse-trading. "I've got a /22 available under 11.12/16. What'll ya give me for it?"
The real reason we're switching to ipv6 is to find out how many autistic-spectrum people are in IT, as they're the ones that won't be using address shorthand.
Interesting idea, but it would have to be quite a super site to get people over...and that would appeal to end users, not the providers that actually have to do the work, no?
I wasn't arguing against it, @packs. They drove adoption of home video technologies.
I just thought home users were too ignorant of things like IPv4/6 to actually demand it, let alone call Comcast and request it because they want to get to Sex.V6.com :-)
I can guarantee that the transition to IPv6 won’t be transparent to everybody. I fully expect ISPs to some day send notices to users that they need to bring in their modems to be exchanged. Rogers has already forced users to upgrade their modems when it upgraded from DOCSIS 1 to 2; they will likely be doing so again when they finally start spending the money we pay to upgrade to DOCSIS 3.
Yes, but it will mean a hardware upgrade, which is exactly what will happen when ISPs are forced to migrate to IPv6, regardless of the users’ tech level. What will be interesting is when the ISP tells some people that they must upgrade your OS.
Today's News TechNet Radio: What the Cloud Means to Microsoft CIO, Tony Scott
Tony Scott, Microsoft CIO is leading the Microsoft IT organization to invest in the cloud. Listen in and find out how it will bring new possibilities and benefits to businesses. Chris Caldwell http://bit.ly/h7Z5iP
also that's not a bad thing -- @Synetechinc has a point that for a lot of users Win9x is sufficient (email and a browser). For that same class of users NAT or public IP doesn't matter -- their mail and browser work fine.
Ah, I'm a brit, and I haven't seen much of it over here. Was wondering if it would become more widespread so as to avoid disconnecting the hordes of 15-year old dust factory PCs.
Besides, not everybody will (or even can) upgrade to the latest and greatest tech. Not only are there many people in North America with lesser computing needs and less money, but you have to remember that there is a whole world out there. Sure, there is 7 Starter Edition, but many third-world countries use whatever is affordable, even if it’s not up-to-date.
@SmallClanger the problem as @MuraliSuriar pointed out is that carrier NAT will break a lot of stuff that defines the "modern" internet -- Great-Grandma is happy with her browser and email, but Grandma these days is using Skype to see the grandkids across the country, and Mom & Dad are playing CoD on the XBox
@Synetechinc and this is part of the reason v4 will be around for a while, but the nature of technology is that eventually investment in legacy equipment won't make sense anymore. The world has to move on if everyone and their toaster wants an internet connection...
@Synetechinc we have internet-enabled refrigerators now... not much of a stretch from those to enabling the oven, then the toaster.... It would be eminently useful to me if my roomba had an IP addres so I could connect to it and retrieve it when it gets stuck somewhere.
@Synetechinc I own a Roomba because I'm lazy. Why should I have to crawl under my couch with a broom handle when I could theoretically connect to it with a web browser (or custom client even) and steer it out manually?)
Just what we need, IP enabled Roombas. Some zombied-bot hackers will crack them and we'll have an army of them hit the highway to attack a business as a physical DOS attack.
@MuraliSuriar During the Obama inauguration if you hit CNN you got either Akamai or LimeLight (not sure how they figured out which one) and if you got LimeLight you received a popup that basically said, "If you want the best video performance, install this plugin. Trust us, it's cool."
it's good about not getting into impossible situations, it just gives up too easily
dude this roomba must have been lobotomized at the factory :)
It also manages to eject its dust bin on its own sometimes. I'll come home and the roomba is on the charging dock with its dust bin like 5 feet away in the middle of the floor :)
Wait until the Roomba gets it own IP address on the Internet and you wake up to have it watching you from the foot of the bed in the middle of the night.
I already have a fish to watch me in the middle of the night. It's actually kinda sweet in a creepy "If I had legs I'd climb out and try to eat you" sort of way
@BartSilverstrim hey, if I get my latte and my news when I walk in the door what do I care if some guy gets his kneecaps shot out for snitching on the mob? :-)
What's the general thought on this.. a comment to your answer improves the answer in such a manner that it really ought to have been included in the answer in the first place. Is it acceptable to edit the answer to include the commenter's information, or is that considered rep-theft?
Also, with a valid AuthzLDAP setup, it's possible to restrict based on attributes
Require ldap-filter &(eduPersonPrimaryAffiliation=*Staff)
or
Require ldap-filter &(memberof=CN=group,OU=unit,DC=container)
The the person really wanted REP they would have provided their own answer. In my opinion it is fine to update a answer to be more correct based on comments that are left. To be polite, you can mention the person's contribution in your edited version.
I think getting good answers is the most important issue. As for "rep theft", you don't get rep for comments and people can upvote as many answers as they like in the same question so you're not cutting anyone's air off. Just be polite and thank the contributor for helping improve the answer
@Ben generally speaking if someone makes a comment that really improves my answers (or makes me rethink a section) I'll edit and add that in after a bar, mentioning who suggested the change. If it's a deletion I'll <s></s> the text
You can post a lot of answers, and since upvotes are +1-, and downvotes are only -2, and many people ignore answers that aren't wrong, the chance of loosing rep from providing a quick-n-dirty answer is very small.
Still, I don't get some of the miss-spellings. Does he not have a spell-checker in his browser?
@RobertMoir My biggest problem with SF is when someone gives an answer that only works in one particular variant of an OS -- I try to give broadly applicable answers whenever possible
Please fix any errors of mine you see. I am far from perfect. I started answering lots of questions on forums many years ago, in an effort to improve my writing ability.
@Voretaq7 - answering for just one version of something... that's not so bad if someone asks about "Windows 2008 SP2 with the following patches installed" but yeah, not so good when people assume everyone is doing it their way
Something that reads my mind and highlights errors where I use a homonym improperly, or mix up its, or something. I tend to do that frequently. Loose/Lose is something I never get right.
Well now, that is certainly an issue. Just a change of how the answer is framed: "Well I'm running OSX and the command line options for top that I have are...." as opposed to "here are the command line switches for top"
well the English language is often a bit loose on how stuff is pronounced and taught, so its easy to lose your way.
@RobertMoir I don't mind if the question says "I'm on [this OS], how do I do [task]", it just bugs me when I see blatant linux-isms in reply to a generic "unix" question
Mostly because when I get new junior admins I have to un-train their blatant linux-isms (BSD shop)
definitely, that's what I'm saying... assume that you're running what I'm running
grin that sounds familiar. I first came across Linux in an IBM mainframe environment that had AIX and SCO Unix (yeah it was a while ago) so we had first hand examples of how things were different
@BartSilverstrim, perhaps not an intention checker. It would be nice if there was a way to have your spell checker highlight words that are commonly miss-used. A slight highlight would be a reminder to check that you are using the words correctly.
@BartSilverstrim sometimes yes (like if they're asking for ps flags), but a lot of the time it's some common task where there's a generic unix-y way and an OS-specific way
and like @Zoredache pointed out there are pretty wide differences between distros, but the "generic unix-y way" works on upwards of 99% of them (and on the BSDs, Solaris, AIX, etc.)
You betcha. Which is why I tend to phrase answers like, "This is how you'd do it from the command line, but your distro's persistent config tools will be different."
@sysadmin1138 I do that a lot on most questions -- or more often if it's something where flags are different I'll just say "You want to use tools X, Y and Z -- here's an example. See the man pages for X, Y and Z for more details"
my expectation is if you're good enough to be on ServerFault (doing this "professionally") you should be able to read a manpage and generalize from my specific example. :-)
When it comes to tools, I admit I frequently assume the GNU tools over BSD and other variants. The GNU tools can be installed on your BSD box with a bit of effort.
and I'm tolerant of stuff like "use wget" because even though it's not in my BSD base system I install it routinely
Where one tool (usually GNU) has a superset of functionality I try to assume the lowest common denominator and give a solution that will work on both platforms.
I'll admit I don't always succeed (I've encountered machines without egrep before, but often I'll give egrep-style regular expressions because... well they're less painful :-)
So -A, and -a translate to ForwardAgent (on|off). I was kinda hoping someone had a nice table. that had every command line option and the equivalent configuration option
@voretaq7, I am reading through the man pages you may be right since I am not seeing anything. Though I don't understand why there wouldn't be a config option.
@Ben I'm not sure about that... ProxyCommand is supposed to set up a pipe basically (between your SSH client and some remote ssh server port) isn't it? Or am I misunderstanding its purpose?
@voretaq7, chaining doesn't help much if you want to use scp to copy something from a remote system you have to go through an intermediate server to access.
Am I caching right? is there any way i can improve this? the site gets about 3k unique hits a day
nginx.conf:
gzip_types text/* application/* image/*;
open_file_cache max=1000 inactive=20s;
open_file_cache_valid 30s;
open_file_cache_min_uses 2;
open_file_cache_errors off;
if_modif...
im trying to run a hello world python script automatically on apache server, when you open the default index.html page, but i can't figure out how to enable it to run automatically
@brainydexter I suggest that you write it up on a question, and then if we answer, it's an answer that could help others too, as the main site gets better indexing than the chatlogs
Hi,
I know how to canonicalise the case of URLs on incoming request to IIS7.5, in fact, there's a built in rule template to start from. But how about outbound (without changing the code)?
This is where I got to so far:
<outboundRules>
<rule name="Outbound lowercase" preCo...
Also, with a valid AuthzLDAP setup, it's possible to restrict based on attributes
Require ldap-filter &(eduPersonPrimaryAffiliation=*Staff)
or
Require ldap-filter &(memberof=CN=group,OU=unit,DC=container)
Am I caching right? is there any way i can improve this? the site gets about 3k unique hits a day
nginx.conf:
gzip_types text/* application/* image/*;
open_file_cache max=1000 inactive=20s;
open_file_cache_valid 30s;
open_file_cache_min_uses 2;
open_file_cache_errors off;
if_modif...
