I think I may have found a fix for this, albeit the most fucking retarded one I've ever seen.
Installing the manufacturer's color profile.
I have one of these laptop webcams:
https://files.facepunch.com/forum/upload/262814/70af0f38-5478-4641-afa6-562146dcdcfc/webcam.jpg
I want to connect it to a USB cable but can't figure out what exactly those marking stand for.
USB+ and USB- are data lines I assume? Which one of those would be power and ground though?
USB+ and USB- (also called D+ and D- sometimes) are your two USB lines yes. DGND would be your ground point, but as for power you're correct in that you wouldn't want to hook it up to the +5V line on your USB cable. Since logic levels for USB are 3.3V TTL, you'll be fine hooking everything else up but you'll still need a regulator to provide 3.3V from the 5V on the USB cable.
At work I learned how to Linux LVM storage on my remote VM and I only corrupted my filesystem once!
NOPE! HASN'T FIXED THIS FUCKING SHIT!
Usually GPU based glitches show up on screenshots too.
My guess: Monitor is bad, have seen this happen with bad caps. Mine got awful pink lines that were most common on white backgrounds.
Signs your programmers don't have enough to do: someone's checked a 1v1 prophunt map of the office into the git repo.
It was me. I did it. In my defense, they pushed the go-live backagainand there are literally zero open issues in the tracker, there's nothing to do on the DB/backend side. I've been cleaning and optimizing for months now, there's no blood left to squeeze from those stones. And working on an actual time-waster project makes me feel slightly less guilty than working on a personal project.
I have been given a domain controller admin at work to get Kerberos SSO on Apache working on a production AD server.
Who wants to bet $10 that I'll somehow break the DC trying to get pure Kerberos auth working, as I'm not supervised and not even my team leader that usually deploys ADs knows wtf to do. So far I got to the point where krb5 communicates and logs in to the AD, but when it comes to Apache it keeps failing when logging in, which makes me suspect that I somehow screwed up the keytab file generation.
This is literally black magic for me.
So excited to use my new Smartphone Device™
https://i.imgur.com/KZFQa9M.png
thats an iphone 4
What would be a good way to replace strings in a 16 bit dll file? I can read them fine with resedit and binary ninja and I guess hex editing could work but it would be nice to be able to completely change them
you could write a python script to open up the dll as bytes, search for the string, and replace them. note that the replacement string must be the same length
Ugh I just had to take apart this MSI GS63VR and re-lubricate the fan bearings. Very nerve-racking at the end when it would not power on. It just needed to be woken up by the AC adapter and now it's working fine, and not making a god-awful squealing noise whenever the GPU fans turn on to boot.
Hoping this is the place to ask silly random questions that google can probably answer so here we go;
Dual monitors, do they make a systems perform worse in general? I'm running a GTX 1050ti on an i72300K. My main monitor is on my GTX but my secondary monitor is on my CPU, and even though my system specs meet up the requirements for most games I play I end up with a lot of jitters and dropped frames. Running both monitors on the GPU don't seem to change much.
Microcenter is awesome, can't speak for Fry's though.
Just running a desktop on a second monitor shouldn't affect framerates in games. Obviously if you stretch a game across two 1080p monitors, you'll get half the framerate because you're rendering twice the pixels, but just having one plugged in shouldn't affect anything in the ballpark of a 1050Ti.
Try either windowed or real fullscreen mode on the game. One of them causes more issues than the other but I can never remember which was the worse one.
Rented an instance from google to run Parsec and do some gaming since I'm not at home rn and only have an old notebook, and damn if it doesn't work surprisingly well. I expected like, 70ms latency but it feels more like 2 to 3 frame at worse.
The preemptive instances are really cheap too, I'm very impressed.
Upload is where it counts, and you're gonna pay for it:
https://s.gvid.me/s/2019/03/06/KUm464.png
That's right folks!
For the low low price of $120/TB of transfer!
Backups: Free
Your office burnt down and you need to recover all data by yesterday: Priceless, apparently
After googling, downloading a bunch of stuff, even using decompilers on the actual .dll, I think I managed to find something that will actually let me do what I want. The problem is that it's a 16 bit application as well, so...
https://files.facepunch.com/forum/upload/109680/f4fe10ce-ca68-4efc-a045-079f798e7d05/VirtualBox_Win95_06_03_2019_18_14_15.png
Good ol' 95
My friend deserves the credit for drawing it
https://twitter.com/malfubabbu/status/1019022495775625217
ANYWAY, turns out it was both DNS and IE, since the Linux server was not registering the FQDN and IE can't really work with IPs added into intranet. Had to set the A record manually on the AD.
Because the Hyper-V VM was acting really weird, I was forced to nuke it and redo it from scratch in a new VM, only to stumble onto a weird kerberos error. This might not really be a place to get help with things like this but it's a worth a shot, because I have no idea anymore.
In Apache I get this:
[Wed Mar 06 09:47:45.018516 2019] [auth_kerb:error] [pid 4323] [client 172.16.20.110:63405] gss_acquire_cred() failed: Unspecified GSS failure. Minor code may provide more information (, No key table entry found matching HTTP/mediawiki.company.local@)
After the "@" the domain is missing and I can't figure out why.
/etc/kbr5.conf:
[libdefaults]
default_realm = COMPANY.LOCAL
kdc_timesync = 1
ccache_type = 4
forwardable = true
proxiable = true
[realms]
COMPANY.LOCAL = {
kdc = company.local
admin.server = company.local
}
[domain_realm]
.company.local = COMPANY.LOCAL
company.local = COMPANY.LOCAL
Generated the keytab like this:
ktpass -princ HTTP/mediawiki.company.local@COMPANY.LOCAL -mapuser kerb.srvc@COMPANY.LOCAL -pass nopassword4u -crypto RC4-HMAC-NT -ptype KRB5_NT_PRINCIPAL -out C:\kerberos.keytab
/etc/hosts is like this:
127.0.0.1 mediawiki.company.local mediawiki localhost.localdomain localhost
::1 localhost6.localdomain6 localhost6
# The following lines are desirable for IPv6 capable hosts
::1 localhost ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts
I'm able to get a ticket from the AD on the Linux server with kinit
Ticket cache: FILE:/tmp/krb5cc_1000
Default principal: kerberos.srvc@COMPANY.LOCAL
Valid starting Expires Service principal
03/05/19 08:47:53 03/05/19 18:47:53 krbtgt/COMPANY.LOCAL@COMPANY.LOCAL
renew until 03/06/19 08:47:49
Even when the fucker 500s because of kerberos failing, it STILL shows an NTLM login prompt with the setting the address as intranet one in IE. Can't show the apache config right now because even after several months I still don't have VPN access after requesting it and I'm not in the office rn.
As I said, this is literal black magic to me and I'm not friends with Windows servers at all. My deadline is Wednesday next week because that's when the coworker that gave me this assignment is coming back from a holiday and that's when we are supposed to demo the SSO implementation for internal mediawiki documentation for a different IT team. I think I'm doomed.
That's for the cloud instances? Motherfuckers.
Yup, that's the "general" Egress pricing that applies to basically all services.
Those prices are basically the same as Amazon, haven't changed significantly in a decade, and are a huge part of why I moved to colocation my own hardware (bandwidth is literally 1/100 the price).
Aside from that one big issue, the thing I hate the most about this whole affair is how hard it is to keep track of running costs, their console just doesn't help much.
It's on purpose.
I still remember legacy Glacier pricing:
How I ended up paying $150 for a single 60GB download from Amazo..
Is absolutely correct:
Had I initiated the retrieval of a 3TB backup this way, the bill would have been $6,138.00 plus tax and AWS data transfer fees.
Oh hey, speaking of backups and how I really don't have enough: what's a good setup for a home NAS box? I only need a few TB, probably a two- or four-disk system, would be nice if it could stream video/audio but not 100% necessary. Are there good prebuilts for this, any brands to look for/avoid, or would I be better off building my own? The last NAS I built was so long ago it was using IDE hard drives, I clearly have no idea what the modern way to do it is.
I can really only relay my personal use cases and solutions (and I'll still preach Snapraid for most infrequently-written-to home NAS usage ) since I used rolling my own with plain old linux on it as a learning experience
I think most others around here run a NAS-centric-OS (like unraid or openmediavault) which are a little more plug-n-play friendly so they'll be more knowledgeable if you want to go that route
Whether you're going for DIY or pre-built is always depending on if you're either willing to spend money or time. Theoretically all you need is a raspberry pi and some hard drives.
Sorry, you need to Log In to post a reply to this thread.
