Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us $3,400 per month for bandwidth bills alone, and since we don't believe in shoving popup ads to our registered users, we try to make the money back through forum registrations.
«750 »
  • Post
  • Reply
peepsalot
Apr 24, 2007

        PEEP THIS...
           BITCH!



anthonypants posted:

You mean like VirtualBox's Seamless Mode?
https://www.youtube.com/watch?v=eQr8iI0yZH4
Looks pretty close to what I mean, except I want to use real computers on a network instead of in a vm.

Adbot
ADBOT LOVES YOU

evol262
Nov 30, 2010
#!/usr/bin/perl

peepsalot posted:

Looks pretty close to what I mean, except I want to use real computers on a network instead of in a vm.

Doesn't VNC already do this? Many VNC clients don't lock the cursor to the window.

Watermelon Daiquiri posted:

I made a server for plex/transcoding/downloading and since I didn't want to buy a redundant computer for streaming to my main TV, I used Ubuntu so everything I needed could be installed without bending over backwards trying to get everything to work in Linux's special little way. Unfortunately, that happened anyways. I got a motherboard with on board Bluetooth since I didn't want to have to find an adaptor that would work well and I read that the bt that came with Intels wireless modules worked well enough. I tried a couple distros and versions like Steamos just because. The bt worked just fine in those, but when I got around to finalizing everything in Ubuntu server 16, the tot suddenly stopped working. The bt manager in the setting just shows "no bluetooth found" even though listing things in terminal showS the os sees the adapters Ive tried with it. . Ive attempted to install various bt-related programs, but nothing has worked. I could always reinstall Ubuntu, but I'm afraid (read: paranoid) of losing the plex settings & metadata I have as well as the lvms I have my plex media on. Has anyone run into something like this and or know how I can get bluetooth working again? If I must I could reinstall the OS, but Id like to know everything would be safe.

I mean, upgrading ubuntu...It almost always breaks everything.

Does bluetoothctl show anything? hcitool scan? You may need to switch from hid to hci (there's a udev utility to do this). Does rfkill show it blocked?

Super Slash posted:

If I remember from this morning /proc/cmdline couldn't find the directory, the last thing I tried was a rescue CD to regenerate initramfs with the difference this time using dracut -v. I was then able to boot fine but my files were missing particularly everything in /opt/, I tried rsyncing the lot from my first VM which didn't quite work out so I ended up tearing the whole thing down and installing from scratch anyway.

I probably did everything rear end backwards since I don't properly know enough, but reinstalling was simple enough once I got Confluence back up since I could just upload a backup.

dracut should definitely not remove /opt. Are you sure /opt wasn't on a secondary disk you didn't migrate?

/proc/cmdline just shows you what arguments were used to boot the kernel. It should always be there.

The question would have been:

What did /proc/cmdline have as the root= directive? Does that device exist? If not, what's actually in /dev/mapper or "lvm lvs" (the `lvs` alias won't work in dracut). The device was probably renamed, but you could have just changed root= to the new name and booted...

peepsalot
Apr 24, 2007

        PEEP THIS...
           BITCH!



evol262 posted:

Doesn't VNC already do this? Many VNC clients don't lock the cursor to the window.
When I have used VNC before it was for viewing the entire desktop from a remote computer.

Here is my use case for clarification: I have a desktop with dual monitors, and a laptop sitting on the desk next to these monitors. I have synergy installed, so if I move my cursor beyond the edge of my desktop screen, it goes over to the laptop screen. This is pretty nice and useful, but due to the layout of my desk etc, laptop screen I have to turn my head farther than is comfortable for extended periods of time. I would like to be able to drag some of the application windows from my laptop running linux over to my desktop monitors, which are on windows.

So in this case I don't want to share the entire desktop, the laptop isn't remote, i don't need to duplicate its entire UI, I only want to drag one or two windows over to my large monitors which are directly in front of my face, without having to do some physical dock nonsense, and having the laptop hijack the entire monitor.

xzzy
Mar 5, 2009

wakey wakey to
this bowl of tasty


Yams Fan

peepsalot posted:

So in this case I don't want to share the entire desktop, the laptop isn't remote, i don't need to duplicate its entire UI, I only want to drag one or two windows over to my large monitors which are directly in front of my face, without having to do some physical dock nonsense, and having the laptop hijack the entire monitor.

I can't see how something like that would ever be possible. To make something like that happen you would either need a protocol to live migrate a running process from one machine to another, or some kind of X11-style protocol that allows changing the display of a running application.

I mean it would be a really cool feature, but I can't imagine anyone ever putting development time into it because not many people would ever ask for something like that.

The closest you'll be able to get to something like that with software available right this minute would be VNC, RDP, or maybe DLNA.

peepsalot
Apr 24, 2007

        PEEP THIS...
           BITCH!



It would be possible just the same way that VNC is possible. I found this by the way http://shared-app-vnc.sourceforge.net/ which is designed for bit different use case but the result seems similar to what I want.

I have not had a chance to try it yet, it doesn't seem too far fetched to me that, for example synergy app could detect when the mouse crosses display borders while dragging a window , and when that happens it would just trigger this special sort of single-app-vnc to connect, and maybe optionally hide the window on the originating system during this process, to help with the visual indication that the window has "moved" to another monitor.

Also is it really such a crazy nerd corner case that no one else would want to be able to use their monitor seamlessly work with applications from more than one computer at a time? I think a lot of people use crappy docking solutions and extra monitors for laptops, being able to use a monitor for laptop + desktop simultaneously just seems like a no brainer, like why aren't we doing this pro-rear end future poo poo today.

Varkk
Apr 17, 2004




Hey don't forget X11 can do something similar. You can't move the window from one machine to the other but you could launch the app to send the display to another X server.

xzzy
Mar 5, 2009

wakey wakey to
this bowl of tasty


Yams Fan

I'm sure someone's thought of it before, that's why tools like screen/tmux and VNC exist.

But either it wasn't as easy to implement as you're hoping, or it just never got sufficient interest to see serious development.

evol262
Nov 30, 2010
#!/usr/bin/perl

peepsalot posted:

When I have used VNC before it was for viewing the entire desktop from a remote computer.

Here is my use case for clarification: I have a desktop with dual monitors, and a laptop sitting on the desk next to these monitors. I have synergy installed, so if I move my cursor beyond the edge of my desktop screen, it goes over to the laptop screen. This is pretty nice and useful, but due to the layout of my desk etc, laptop screen I have to turn my head farther than is comfortable for extended periods of time. I would like to be able to drag some of the application windows from my laptop running linux over to my desktop monitors, which are on windows.

So in this case I don't want to share the entire desktop, the laptop isn't remote, i don't need to duplicate its entire UI, I only want to drag one or two windows over to my large monitors which are directly in front of my face, without having to do some physical dock nonsense, and having the laptop hijack the entire monitor.

Ah. No, this doesn't exist.

There are a lot of technical reasons why not (process locality, drawing instructions, what happens if you have the window hallway across the divider, etc).

What people do is... Open the application on the powerful system and use it there. This also applies/applied to cases where artists or engineers have one "work" computer running Windows, and one "WORK" computer running OSX or a UNIX variant. And they've done this for decades. Because there is no practical way to write the application you describe.

Thermopyle
Jul 1, 2003

...the stupid are cocksure while the intelligent are full of doubt. —Bertrand Russell


I'm thinking about running my own OpenVPN server on a VPS (probably digitalocean) for all my traffic on my home network as well as my mobile devices. Mostly just because I want to try it out and it's not something I've messed with before.

I've read that doing this can introduce some amount of latency. Assuming the VPS has reasonable bandwidth/ram/cpu, what kind of ballpark latency hit am I looking at? Obviously there is the extra hop to the VPS, but I assume that OpenVPN processing itself adds some amount of latency. A few ms? 10ms? 100ms? 1000ms?

xzzy
Mar 5, 2009

wakey wakey to
this bowl of tasty


Yams Fan

I don't do a VPN but I do run a squid proxy on a digital ocean box and it's fine, using the 1core/1gb configuration. It definitely adds latency because of the extra hop, but it's good enough for normal browsing. Like my SA bookmarks page loads in about 950ms without the tunnel, and it's 1100ms with the tunnel. It only hurts when viewing enormous imgur albums, the link saturates pretty quick.

The biggest problem with using digital ocean as a VPN is you'll hit google's catchpa all the drat time. I'm guessing digital ocean's less savory customers run all kinds of lovely bots and SEO bullshit and get the ip range flagged for verification constantly.

To get around that, I run a "primary" squid proxy on a raspberry pi at home.

mystes
May 31, 2006



peepsalot posted:

When I have used VNC before it was for viewing the entire desktop from a remote computer.

Here is my use case for clarification: I have a desktop with dual monitors, and a laptop sitting on the desk next to these monitors. I have synergy installed, so if I move my cursor beyond the edge of my desktop screen, it goes over to the laptop screen. This is pretty nice and useful, but due to the layout of my desk etc, laptop screen I have to turn my head farther than is comfortable for extended periods of time. I would like to be able to drag some of the application windows from my laptop running linux over to my desktop monitors, which are on windows.

So in this case I don't want to share the entire desktop, the laptop isn't remote, i don't need to duplicate its entire UI, I only want to drag one or two windows over to my large monitors which are directly in front of my face, without having to do some physical dock nonsense, and having the laptop hijack the entire monitor.
This would probably be doable now by running the programs through something like Xpra, migrating them to a vnc server x server, and launching a vnc viewer. Kind of ugly, though.

It might also be possible to have a compositor be able to hide a window locally in the final, composited desktop but still render, capture and stream its content over the network. Remote desktop software seems to be moving in the direction of capturing from the video card anyway but it's taking a long time to catch on.

HPL
Aug 28, 2002

Worst case scenario.

Teamviewer can do the multiple monitors thing. One day, just for kicks, I set up a Chromebook as a second monitor for my Windows PC.

mystes
May 31, 2006



HPL posted:

Teamviewer can do the multiple monitors thing. One day, just for kicks, I set up a Chromebook as a second monitor for my Windows PC.
Actually if that's what you want it at least used to be possible to combine two completely separate x servers into one, so you could join your normal display with a vnc server that you would connect to from another computer and drag windows between them.

Saukkis
May 16, 2003

Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

SpaceDesk also can do something similar to what you want, but unfortunately it only works the other way, you could extend the Windows desktop to the Linux computer. Other alternatives also are designed for extending Windows desktop to other devices.

http://alternativeto.net/software/maxivista/

apropos man
Sep 5, 2016

You get a hundred and forty one thousand years and you're out in eight!

So today I installed a spare mechanical hard drive in my desktop machine. There's already two SSD in it: one with Ubuntu and one with Windows 10.

I booted into Ubuntu and created a new gpt table on the mechanical drive, then did full disk encryption with cryptsetup, mounted the crypt and made an ext4 fs on it. Then I copied a bunch of stuff to it and rebooted Ubuntu to make sure it was still working. All is well.

I then booted into Windows and the drive doesn't show up in Explorer, even though I've got 'show empty drives' ticked. This didn't worry me too much, as Windows is probably just hiding a disk it cannot determine the format of.

I decide to have a look at my drives in Disk Management and a popup appears saying 'You need to initialise this disk before you can use it' and prompting me to click OK to write a new partition table to it.

I don't want to Windows bash in here, but sweet baby jesus, is it any wonder that people accidentally click stuff and lose data?

Naturally, I've closed Disk Management and left the new volume alone.

xzzy
Mar 5, 2009

wakey wakey to
this bowl of tasty


Yams Fan

The powershell disk management tool is a lot more capable, but if it's not formatted with something windows understands there's not much point.

Walked
Apr 14, 2003



So I'm trying to expand my knowledge to include a deeper understanding of Linux, coming from an extensive windows background.

Which distribution should I be using in a lab setting with the intention of expanding my skillset for actual career value.

Is RHEL/CentOS the way to go? Ubuntu server? I'm more familiar with Ubuntu but it's rather have the hands on with whatever is going to bring the most value to the table

Keito
Jul 21, 2005

WHAT DO I CHOOSE ?


apropos man posted:

I decide to have a look at my drives in Disk Management and a popup appears saying 'You need to initialise this disk before you can use it' and prompting me to click OK to write a new partition table to it.

Someone setting up dm-crypt hopefully knows better than to blindly press OK.

Poor Mac users with their HFS+ disks, on the other hand...


Microsoft develops Windows with the assumption that it's the king of the hill and everyone else should cater to their fancy, so it's hardly surprising. Windows can't read your disk's file system? Better reinitialize it in a format we are in control of, pronto!

apropos man
Sep 5, 2016

You get a hundred and forty one thousand years and you're out in eight!

Keito posted:

Someone setting up dm-crypt hopefully knows better than to blindly press OK.

Poor Mac users with their HFS+ disks, on the other hand...


Microsoft develops Windows with the assumption that it's the king of the hill and everyone else should cater to their fancy, so it's hardly surprising. Windows can't read your disk's file system? Better reinitialize it in a format we are in control of, pronto!

That's what I was thinking. It's the arrogant way that Windows suggests to nuke the drive. It'd probably do the same to a portable HDD, too. Leave your dm-crypt drive lying around next to a casual Windows users PC? Kablamm!

Walked posted:

So I'm trying to expand my knowledge to include a deeper understanding of Linux, coming from an extensive windows background.

Which distribution should I be using in a lab setting with the intention of expanding my skillset for actual career value.

Is RHEL/CentOS the way to go? Ubuntu server? I'm more familiar with Ubuntu but it's rather have the hands on with whatever is going to bring the most value to the table

Someone on here mentioned using a distro that's appropriate to the work you're trying to do or what you're trying to achieve. I think it was evol262. I'd agree with that. I like Fedora on my laptop because it has SELinux for security and my laptop is the most used computer in the house. I end up browsing all sorts of websites on it. I use Ubuntu on my sever because if I want to set up mail forwarding or do other exotic stuff (well, exotic for me) it's well documented online and I can achieve my goal. I also use Ubuntu on my workstation PC at the moment, but that could change if I decide to try something else.

RFC2324
Jun 7, 2012

Http 418


Walked posted:

So I'm trying to expand my knowledge to include a deeper understanding of Linux, coming from an extensive windows background.

Which distribution should I be using in a lab setting with the intention of expanding my skillset for actual career value.

Is RHEL/CentOS the way to go? Ubuntu server? I'm more familiar with Ubuntu but it's rather have the hands on with whatever is going to bring the most value to the table

CentOS/RHEL is the most common business distro, Ubuntu is probably the easiest 'it just works' for desktop.

I say go gentoo so you can understand all of your configurations, and the pain of compiling everything from scratch.

Saukkis
May 16, 2003

Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

apropos man posted:

That's what I was thinking. It's the arrogant way that Windows suggests to nuke the drive. It'd probably do the same to a portable HDD, too. Leave your dm-crypt drive lying around next to a casual Windows users PC? Kablamm!

You can probably avoid this problem if you create a partition, give it a suitable partition type and encrypt the contents. This should make it possible for Windows to identify the partition and know that it doesn't understand the contents. This at least should work with DOS partition table, but I haven't tried it with GPT.

If I understood your first post correctly, you first created a GPT partition table on the drive, then encrypted the whole drive, which sounds pointless to me. The encryption would destroy the GPT and because of it Windows would see it as empty drive.

apropos man
Sep 5, 2016

You get a hundred and forty one thousand years and you're out in eight!

It's not a problem. What struck me as bad was the insistence of Windows to create a new partition table, just because it didn't seem to recognise what was on it. I haven't got a problem with Windows not recognising dm-crypt partitions, but it should at least warn to the effect that all data will be lost.

Saukkis posted:

If I understood your first post correctly, you first created a GPT partition table on the drive, then encrypted the whole drive, which sounds pointless to me. The encryption would destroy the GPT and because of it Windows would see it as empty drive.

I created a new gpt at the very start in order to blitz whatever partitions were on the drive beforehand. It had been sitting around for a while. I don't know if dm-crypt writes over the partition table, but I think it doesn't.

Docjowles
Apr 9, 2009



Is anyone in here a ninja with rsyslog? I have a fairly specific question and either it's not possible, or I can't find the right Google terms.

We run rsyslog on all of our servers, configured to stream logs to a central logging server. They get placed in a directory structure like /year/month/day/hostname/. Standard stuff. There's some scripts that run on that server and do work to parse and act on them. Sometimes for whatever reason, replication breaks, and logs aren't copied over, which breaks said scripts when the files they want aren't there.

Is there a way to get rsyslog to go back and re-read the files and lines it missed the first time? I can't simply rsync them over, because rsyslog prepends the normal syslog header to the lines it receives. And also renames them. So the scripts expect a different name and format than the raw source logs use.

I tried adding an "imfile" config pointed at the old files, but nothing happened. Possibly user error, of course.

If there's no good way to do this, I'll screw around with Logstash (which we already use in the traditional way) or a custom script. But was hoping it was something rsyslog could do out of the box.

Thermopyle
Jul 1, 2003

...the stupid are cocksure while the intelligent are full of doubt. —Bertrand Russell


Thermopyle posted:

I'm thinking about running my own OpenVPN server on a VPS (probably digitalocean) for all my traffic on my home network as well as my mobile devices. Mostly just because I want to try it out and it's not something I've messed with before.

I've read that doing this can introduce some amount of latency. Assuming the VPS has reasonable bandwidth/ram/cpu, what kind of ballpark latency hit am I looking at? Obviously there is the extra hop to the VPS, but I assume that OpenVPN processing itself adds some amount of latency. A few ms? 10ms? 100ms? 1000ms?

For anyone who cares, the answer is "a few ms".

I ping ~40ms to https://www.google.com from my home connection.
I ping ~20ms to my digital ocean VPS running openvpn.
I ping ~63ms to https://www.google.com when using OpenVPN.

If I'm thinking about this correctly, if OpenVPN added 0 latency, than my ping would increase by the latency of the extra hop(s) caused by going to digital ocean. My best estimate is that, on average, adding the two together comes up a few ms short...and I assume that comes from OpenVPN and whatever other layers my packets travel through on my Ubuntu server itself.

Now to see what adding HAVP and Privoxy into the mix does...

LochNessMonster
Feb 3, 2005

I need about three fitty



Thermopyle posted:

For anyone who cares, the answer is "a few ms".

I ping ~40ms to https://www.google.com from my home connection.
I ping ~20ms to my digital ocean VPS running openvpn.
I ping ~63ms to https://www.google.com when using OpenVPN.

If I'm thinking about this correctly, if OpenVPN added 0 latency, than my ping would increase by the latency of the extra hop(s) caused by going to digital ocean. My best estimate is that, on average, adding the two together comes up a few ms short...and I assume that comes from OpenVPN and whatever other layers my packets travel through on my Ubuntu server itself.

Now to see what adding HAVP and Privoxy into the mix does...

Interesting. I've been thinking about doing the same for some time now. I currently vpn to my openvpn docker container on a raspberry pi at home which works fine for now though. It does add some latency but surprisingly nothing major.

VikingofRock
Aug 24, 2008


Slippery Tilde

apropos man posted:


I decide to have a look at my drives in Disk Management and a popup appears saying 'You need to initialise this disk before you can use it' and prompting me to click OK to write a new partition table to it.

I don't want to Windows bash in here, but sweet baby jesus, is it any wonder that people accidentally click stuff and lose data?

Naturally, I've closed Disk Management and left the new volume alone.

This happened to me recently, and what was most surprising to me was that the wording didn't even really make it clear that Windows was going to write a new partition table to the device. I read it as a purely informative pop-up, clicked OK (as in, "I understand"), and then boom the other drive was gone. Luckily all I lost was a 1-day-old backup, but wow that is a bad pop-up.

Odette
Mar 19, 2011



VikingofRock posted:

This happened to me recently, and what was most surprising to me was that the wording didn't even really make it clear that Windows was going to write a new partition table to the device. I read it as a purely informative pop-up, clicked OK (as in, "I understand"), and then boom the other drive was gone. Luckily all I lost was a 1-day-old backup, but wow that is a bad pop-up.

Windows is extremely destructive & damaging to other partitions. I've had Linux boot partitions on both BIOS & UEFI systems completely blown away by Windows without warning. And if you have a Windows 10 partition automounted in your Linux config; any Windows update will most likely turn "fast startup" on, rendering Linux unable to boot.

It's a nightmare to deal with.

RFC2324
Jun 7, 2012

Http 418


Odette posted:

Windows is extremely destructive & damaging to other partitions. I've had Linux boot partitions on both BIOS & UEFI systems completely blown away by Windows without warning. And if you have a Windows 10 partition automounted in your Linux config; any Windows update will most likely turn "fast startup" on, rendering Linux unable to boot.

It's a nightmare to deal with.

you mean mounted at boot, or mounted using automountd?

Odette
Mar 19, 2011



RFC2324 posted:

you mean mounted at boot, or mounted using automountd?

Well, I've got the partition UUID in /etc/fstab, so it would be mounted at boot. Not really sure if there's a better way of doing this.

apropos man
Sep 5, 2016

You get a hundred and forty one thousand years and you're out in eight!

There are options you can put in fstab which would allow Linux to ignore the drive if it can't be mounted. I've never found a particularly definitive set of options to use, though, so a bit trial and error is needed.

I'm assuming that's what is happening here: the Windows drive is still unlocked by Windows, thus preventing Linux from using it.

Docjowles
Apr 9, 2009



Docjowles posted:

Is anyone in here a ninja with rsyslog? I have a fairly specific question and either it's not possible, or I can't find the right Google terms.

We run rsyslog on all of our servers, configured to stream logs to a central logging server. They get placed in a directory structure like /year/month/day/hostname/. Standard stuff. There's some scripts that run on that server and do work to parse and act on them. Sometimes for whatever reason, replication breaks, and logs aren't copied over, which breaks said scripts when the files they want aren't there.

Is there a way to get rsyslog to go back and re-read the files and lines it missed the first time? I can't simply rsync them over, because rsyslog prepends the normal syslog header to the lines it receives. And also renames them. So the scripts expect a different name and format than the raw source logs use.

I tried adding an "imfile" config pointed at the old files, but nothing happened. Possibly user error, of course.

If there's no good way to do this, I'll screw around with Logstash (which we already use in the traditional way) or a custom script. But was hoping it was something rsyslog could do out of the box.

If anyone cares, I got this working using Logstash to replay the old logs. Not ideal, but it works, and I can go on with my life.

gourdcaptain
Nov 16, 2012



Hey, I know this tends torwards a more enterprise/serious Linux thread, but I've got a question on where and if to file a bug.
I'm running Arch Linux (yeah, I know, it does a few things I need that other Linuxes don't do well, I'm used to its issues, and I don't recommend it to other people) and they in an Arch-iest possible way recently pushed a Mesa 13 release canidate to the repositories. I'm not complaining (the extra OpenGL support makes a game I was waiting to run work and it hasn't caused any issues). Anyway, this also means they added a package for the open-source Vulkan implementation for recent Radeon cards (I've got an RX 460 in a machine I just built because the CPU didn't have an integrated one and I needed something and it was cheap). I'd like to mess with this with Dolphin, but anything I try to run for it just crashes immediately with a segfault or says it can't create a Vulkan context.

Should I file a bug against:
A) Arch (even though this is a stock package with no patches)
B) Mesa (even though the driver flat out admits when you run something using it in the terminal it's not yet ready for prime time)
C) Dolphin (just for completeness, this doesn't seem right to me at all, and Vulkan is only available in recent git releases there although that runs fine on the hilariously low end Intel GPU I have in a convertible tablet)
D) Multiple of the above?
E) None of the above?

gourdcaptain fucked around with this message at Oct 31, 2016 around 18:02

anthonypants
May 6, 2007

by Nyc_Tattoo


Dinosaur Gum

gourdcaptain posted:

Hey, I know this tends torwards a more enterprise/serious Linux thread, but I've got a question on where and if to file a bug.
I'm running Arch Linux (yeah, I know, it does a few things I need that other Linuxes don't do well, I'm used to its issues, and I don't recommend it to other people) and they in an Arch-iest possible way recently pushed a Mesa 13 release canidate to the repositories. I'm not complaining (the extra OpenGL support makes a game I was waiting to run work and it hasn't caused any issues). Anyway, this also means they added a package for the open-source Vulkan implementation for recent Radeon cards (I've got an RX 460 in a machine I just built because the CPU didn't have an integrated one and I needed something and it was cheap). I'd like to mess with this with Dolphin, but anything I try to run for it just crashes immediately with a segfault or says it can't create a Vulkan context.

Should I file a bug against:
A) Arch (even though this is a stock package with no patches)
B) Mesa (even though the driver flat out admits when you run something using it in the terminal it's not yet ready for prime time)
C) Dolphin (just for completeness, this doesn't seem right to me at all, and Vulkan is only available in recent git releases there)
D) Multiple of the above?
E) None of the above?
Depends on what your crash/segfault says.

gourdcaptain
Nov 16, 2012



anthonypants posted:

Depends on what your crash/segfault says.

Unfortunately, as posted, just "cannot initialize a Vulkan context" from Dolphin and a segmentation fault without any other messages related to it from any other program using Vulkan. The Vulkan capabilities viewer works, and that's about it.

xzzy
Mar 5, 2009

wakey wakey to
this bowl of tasty


Yams Fan

Run coredumpctl then, and have fun.

ExcessBLarg!
Aug 31, 2001


gourdcaptain posted:

Should I file a bug against:
I'd probably post a report against Dolphin in hopes of catching the attention of people likely to run into the same issue and hope that there's enough collective traction to figure out where the problem really is.

Otherwise you would generally post a report closest to the most-likely buggy component, but in this case it could be any of Mesa, Radeon, or Dolphin, and the first two are probably full of "it doesn't work" reports with not much information to debug further. Also the Dolphin people are, if they care and are able to reproduce it, in the best position to obtain enough debugging information to make a useful report against Mesa or Radeon.

gourdcaptain
Nov 16, 2012



ExcessBLarg! posted:

I'd probably post a report against Dolphin in hopes of catching the attention of people likely to run into the same issue and hope that there's enough collective traction to figure out where the problem really is.

Otherwise you would generally post a report closest to the most-likely buggy component, but in this case it could be any of Mesa, Radeon, or Dolphin, and the first two are probably full of "it doesn't work" reports with not much information to debug further. Also the Dolphin people are, if they care and are able to reproduce it, in the best position to obtain enough debugging information to make a useful report against Mesa or Radeon.

Thanks. However, since there's a new Mesa release candidate out since I first posted and the final version is due out soon (apparently less than a day), I'm probably going to sit on this until I've got the final release of Mesa 13 because I really don't want to compile Mesa again due to some bad experiences. Although given this Mesa version isn't using a pre-release version of LLVM like my last try at compiling Mesa development versions myself (I needed OpenGL 4.3 for something) it wouldn't be too bad (compiling LLVM SVN builds are the worst, it'll take half an hour, error on one test, and abort), but I'll just give it a few days to get that straightened out when Arch pushes it and then try Dolphin again with it.

EDIT: Side note: Does anyone know a good way to set it so certain applications open on a specific monitor all the time? It wasn't especially bad on my last system where I wanted most things on the primary monitor, but I've got a good half-dozen windows to drag to the secondary monitor (chat clients, e-mail, music player) each boot on the current one.

gourdcaptain fucked around with this message at Oct 31, 2016 around 20:30

Xik
Mar 10, 2011



Dinosaur Gum

gourdcaptain posted:

EDIT: Side note: Does anyone know a good way to set it so certain applications open on a specific monitor all the time? It wasn't especially bad on my last system where I wanted most things on the primary monitor, but I've got a good half-dozen windows to drag to the secondary monitor (chat clients, e-mail, music player) each boot on the current one.

What DE/WM? If you're running i3 you can assign workspaces to specific displays and then configure applications to run on certain workspaces by default.

gourdcaptain
Nov 16, 2012



Xik posted:

What DE/WM? If you're running i3 you can assign workspaces to specific displays and then configure applications to run on certain workspaces by default.

XFCE4. Yeah... I need to think about moving to LxQT or something, it's getting a bit creaky.

Adbot
ADBOT LOVES YOU

DrankSinatra
Aug 25, 2011


I'm trying to figure out what distro to use for my daily work. I realize, to a certain extent that's a dumb question, but whatever. I'm a grad student/programmer, and I spend a solid 75% of my computer time at the command line or in Emacs. I like farting around in Slackware, because it's a Unix-rear end Linux system, and coming at it from the perspective of a dude who does a ton of command line stuff, I like how every component can basically be configured in their respective text file; it feels simple and predictable to me.

On the other hand, dealing with slackbuilds gets old really fast. I have a life to live, and I'd rather not piss it away staring at compiler output.

Is there anything out there with a robust binary package management system that's still that level of dead-simple vanilla text configuration? I thought about just doing a server install of CentOS, and installing the packages I want, but the packages in the default repos are old as hell, and I really don't want to putz with adding a bunch of extra repos. Also it seems like, even at the default setup level, CentOS has a bunch of extra enterprise cruft that I'm not interested in.

Realistically, all I need is the bleeding edge version of my compilers and text editor, a web browser, virtualbox, and Xmonad.

DrankSinatra fucked around with this message at Nov 1, 2016 around 07:22

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply
«750 »