I'd like to create a simple HTTP(S) proxy allowing only certain IPs XXX.XXX.XXX.* and requiring authentication. Also it won't be forwarding IP of the guest. What solution should I choose? If you have created such a thing using some tutorial, link would be welcome.
Is there a limit where you stop getting honeypot questions? I'm 3 flags shy of 700 flags on ServerFault and I'm still getting honeypot questions. I'm fairly sure I know what I'm doing at this point. — tombull89May 15 at 20:05
@tombull89 unlikely: I have about twice as much flags at Programmers (~1300) and I am getting honeypots. BTW I am also still failing some of them, as well as making mistakes in flagging - which is probably the reason why audits won't stop (no matter how good one is now, there's no guarantee this will be always so in the future) — gnatMay 15 at 21:53
When you format the result as a list, ("Get-Acl | Format-List"), in addition to the path, owner, and access list, Windows PowerShell displays the following properties and property values:
-- Group: The security group of the owner.
@syneticon-dj I see what you did there but like most questions associated with LAMP/WAMP/XAMP etc developer stacks its still crappy. All you've done is change it from OT to NARQ as it really doesn't provide any useful information - I don't understand why you wasted your time :(
@pauska this seems to be the "primary" POSIX group of the principal. I am not sure where it comes from, though - System.Security.Accesscontrol which is the object type returned by Get-ACL does not have this property.
Thanks man , Since my question considered not real question :( i post those specifications here , dovecot , sendmail , over 1,200,000 users and 25,000 active users , migrating from an exchange server which was windows 2008 R2, i am the developer of web client and mobile client, they asked me to give my opinions on how they should configure the server specially partitioning , my developing environment was CentOS 6.3 ~ 6.4 — Synxmax3 hours ago
@syneticon-dj Where is this set? I know that some users had a different primary group than Domain Users in the past, but this has all been revoked to Domain Users long time ago..
@Iain The question seems answerable as @Adrian has proposed an answer. And it likely is on-topic too, although it certainly is not suffering from being too descriptive. I think having it re-opened and the answer accepted is of more value to the site than leaving it as is.
This is a big server... It's probably overkill for the application of just running mail.
We're missing so many details here that it's difficult to give a real answer. Consider:
What mail solution are you using?
How many users do you have to support?
How much data needs to be stored?
Are you mi...
I feel that SE is lacking a decent process to cycle questions through the request/improve phases. Using comments for asking questions, making edits and again writing comments to notify the people asking is ... ugly at best.
The majority of our infrastructure is on AWS but there are some other elements: a few mac minis and a pool of Android virtual machines on dedicated hardware.
@syneticon-dj Yes, I understand that part, but it's not reflecting the current primary group of that user. Which is why I said that it seems like it's not something that changes when the user modifies the object (it stays at the same SID forever, apparently)
@pauska no, it would (and should) not - it is set once and is the equivalent of what is set in UNIX by issuing chown :group. No idea what is capable of changing it - probably Cygwin's chown would
Oh, I don't want to do DevOps as such - I just like the automation route that Linux has took. I know Microsoft are trying the same with Orchestra and the like, but, hmm, I'm not so convinced
or such glaring engineering mistakes... At work I have a major devops client. And while they've automated this crazy infrastructure, they're missing so many sysadmin basics that it's sad.
@pauska I do like Group Policy, besides I'm so entrenched in Microsoft that it would take me a long time to be able to realistically put Linux on my CV for the kind of job titles I want
It's funny, I look at all these jobs and they want people who have dealt previously with "massive" databases or "scaling" and things of that sort. Hard to get exposure to this sort of stuff when everyone requires experience already. :P
It's like asking for 5 years experience on Windows Server 2012!
after working for DAYS on an specific SELinux policy for this bloody oracle database, I've been forced by management to set the server to permissive. This DB admin dude says: "see, it works now" . Sometimes I feel like 3.bp.blogspot.com/-Cd3t-tecEOA/TnDVQiHX16I/AAAAAAAAAGo/…
@syneticon-dj Honestly I have no idea what they were trying to open it with (it was my resume). I ended up re-sending it as a .doc so they could open it.
@NathanC We have a catalog we send out. Could be called a product guide, product catalog, item catalog...something. Nope. They call it a "Guide to packaging profits..." I am still looking for where it instructs our distributors to make more profit from their packaging
@Travis Oh, I see. To be honest, I never noticed the compatibility pack has been updated to include older Office versions. As it has been distributed initially, Office 2000 was not supported IIRC.
How do I check whether I have access to/am accessible from Internet2? Google says I should be using Internet2 Detective, but that tool seems to have disappeared.
I'm looking at the Network Solutions DNS zone for one of my clients... and I see: capital.aca .productiveBLAH.com baa.aca .productiveBLAH.com dev .productiveBLAH.com
Titan is a supercomputer built by Cray at Oak Ridge National Laboratory for use in a variety of science projects. Titan is an upgrade of Jaguar, a previous supercomputer at Oak Ridge, to use graphics processing units (GPUs) in addition to conventional central processing units (CPUs), and is the first such hybrid to perform over 10 petaFLOPS. The upgrade began in October 2011, commenced stability testing in October 2012 and it became partially available to researchers in early 2013. The initial cost of the upgrade was US$60 million, funded primarily by the United States Depart...
I'm running a VMware Workstation VM, to run a VPN so I can RDP to a server. From there I'm RDPing to a server which runs their VSphere Client. I'm then using the console to build a VM. It's painful
@JennyD Actually, the computer components are very outdated. The procurement and processing and years of acceptance and safety testing pushes them well behind what is "current"
I'm using names like a.alpha for the hostname of my linux box, but it seams that these name are not completely usable. The response of a hostname shell command is correct (a.alpha).
But the name printed after my user account is "user@a" instead of "[email protected]". When I use avahi, I can reach (b...
@NathanC I see bitcoins like buying shares in minor companies: it's a gamble, don't use money you're not prepared to lose; you may get lucky but don't count on it
I'm looking at the Network Solutions DNS zone for one of my clients... and I see: capital.aca .productiveBLAH.com baa.aca .productiveBLAH.com dev .productiveBLAH.com
@NathanC Yeah, I did a quick Google to see how to mine them because I have a dormant server I could do it on that has 64GB of RAM and dual, 6 core processors
@Dan I think the only way you're going to have conflict though is if you actually expect the hostname to have the dot in it. If you understand that capital.aca with $ORIGIN productiveBLAH.com is really hostname capital then it's acceptable. It seems when BIND does AXFR to a slave it will set $ORIGIN "correctly" on the slave. Either way it will resolve. So, if it's done as a shortcut on the master to avoid creating a whole seperate zone file for 1 or 2 records, I see no problem.
@ewwhite It seems so. At least from an RFC and an "it works" stand point. But, it's a matter of scale and maintainability. Also, like that SF question postures. As long as you don't expect to be able to mail [email protected].
My understanding was that periods are not permitted characters. I haven't looked at any of those RFCs since before unicode was allowed, so I may be out of date or misremembering.
@ScottPack Right! If you only have 3-4 DNS records that are hostname.dev or something, I think it's fine to keep them in the parent zone with a common $ORIGIN rather than setting up another zone file. But, if you expect that "dev" subdomain to grow, it might be better off to do it now rather than later?
I'm not sure how really "insecure" it would be anyway since the only thing the CA is doing is providing a certificate for the firewall to do LDAPS queries
@Travis Also, it's not enough that the CA server itself be secure (which basically means it's locked in a cage, not reachable by the internet at large or even internal systems other than those absolutely necessary, not allowing even SSH logins, etc). The main issue is making sure that absolutely everything done by sysadmin, CAA, etc, is always logged and always traceable
@Travis Given that, scratch my latest screed. I was thinking of a CA made for really important purposes
@AaronCopley Ah, ok, I get it. We actually do quite a lot of that here. So, for example, it.example.com is the subdomain used for internal IT resources and is a fully separate zone. Then you'll have security.it.example.com for the security office, net.it.example.com for the Networks stuff, sysops.it.example.com for the SysAdmins stuff, etc. All of those still exist within the it.example.com zone.
@JennyD Right. Really no risk if it is compromised. All it would do is prevent the firewall from performing LDAP queries to determine which user is using what service..
@Travis Any time you want to look into hardening something you need to take a hard look at the actual risk profile of the system. You don't want to end up putting a $5000 padlock on a shed storing a $100 mower, for example.
@Travis It should be as secure as the ldap server that is trusting the certificates, then.
All of this is IMAO of course; but I've been setting up an internal CA actually used to give out credentials to banking customers, so our security requirements were very high...
@ScottPack Right, but in your example the zone $ORIGIN would be example.com and the records security.it. net.it, and sysops.it. It's dirty when considering dozens of records. But, only a handful?
@AaronCopley Well, in my example example.com would be a zone, it.example.com would be another zone, and the others would just be subdomains within the it zone. How messy that gets from a management environment I can't say. That's what the DNS guys are for. :)
@AaronCopley I do know that the arrangement works completely fine in both BIND and MS-DNS.
@ScottPack Yea, I see no problem for a handful of records. But, if you are part of a rapidly growing organization, you may want to go ahead and create a separate zone to save the trouble later.