Re: Update broken depencies

[jarmo]

Tue, 21 Feb 2017 13:39:31 +0800
Ed Greshko  kirjoitti:


> 
> run dnf again but specify --best to see what is blocking for you.

Thanks

It was system-config-firewall-base-1.2.29-12.fc21.noarch Which blocked
update...

Jarmo
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Need help with mounting an old logical volume

[Fulko Hew]

To make a long story short, my laptop screen died, so I bought a new laptop.
I installed Fedora 25.  Then I wanted to copy my old stuff off of the old
drive
(Fedora 23) so I took the drive out and put it into a USB container and
plugged
it in.

The first problem is that its got a logical volume,
and its name is duplicated with the current name ('fedora').

Searching the web I find that I have to rename the old volume to something
unique.

a) I used 'pvs' to determine the device names and to see the sizes,
b) I then used 'vgs -d vg_all' to find the UUID that matched the sizes
   of the device I needed to alter.
c) Having the UUID, I then used 'vgrename UUID fedora_old'
d) Then I needed 'ls /dev/fedora_old' to see the 'sub-device' called
'root00'
e) That let me 'mount /dev/fedora_old/root00 mydir'

That was good enough to allow me to copy some of my data off,
but it wasn't showing anything in /home/fulko.

My next step was to plug an external monitor into that old laptop
and reboot, and try to FTP the remainder of my data from /home/fulko.

That didn't work because the volume name was now changed.
So I had to change it back.

I booted an F23 live CD, used 'vgs' to find the UUID, and the  'vgrename'
to rename it back to 'fedora'.

Rebooting with the drive, now fails with the following error messages:

   ...
   Failed to start File System Check on /dev/mapper/fedora-root00
   See 'systemctl status systemd-fsck-root.service' for details.
   Dependency failed for /sysroot.
   Dependency failed for Initrd Root File System
   Dependency failed for Reload Configuration from the Real Root.
   ...

and then goes into emergency mode.
Following the instructions in those error messages, I used 'sytemctl'
to see what I could see, and it says:


/dev/mapper/fedora-root00: Inode 1317765, end of extent exceeds allowed
value
(logical block 128, physical block 7479168, len 15)
/dev/mapper/fedora-root00: UNEXPECTED INCONSISTENCY: run FsCK manually.
   (i.e. without -a or -p options) fsck failed with error code 4.
Running request emergency.target/start/replace
...


Now what?
I'm afraid of running fsck on it to resolve whatever problems it may have,
(because I'm afraid of screwing it up even more)
and on the other hand, what did I do wrong to get problems in the place?

Comments and suggestions will be more than welcome.
Thanks in advance.

Fulko
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

compiz problems

[Jon LaBadie]

Running Fedora 25 with a Mate desktop environment.  I ran
the same on F24 and still do on another system.  I would
prefer to use compiz as my window manager.  However it is
unusable.  If I move a window, images trail the movement
and remain when I stop.  Popup menus don't disappear when
I make a selection.  Killing a window does not cause its
image to disappear.

All the residual garbage is cleaned up if I change
workspaces and change back again.  But the desktop
wallpaper is gone too.

I don't run any of compiz' fancy features, wobbly
windows, etc.

Any idea what might cause this problem?

Jon
-- 
Jon H. LaBadie  jo...@jgcomp.com
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: Importing OpenVPN. ovpn file in Fedora 25

[Ed Greshko]

On 02/21/17 12:43, Stephen Morris wrote:
> Thanks Ed. After sending the email I went to that site and their site
> has changed. I downloaded the .ovpn files their links were pointing at
> for Singapore and Miami and imported them into Networkmanager which
> set everything up, including pointing the CA Certificates entry at an
> external location URL to get the certificate, which is a different
> location for each server and is pointing at a .pem file. I found via
> this method that their Australian servers still don't work and neither
> does their Indian Server, and their Hong Kong server wants a pin in
> order to connect.


Well, you certainly are seeing things different from what I'm seeing.

The openVPN config file for Singapore (gw1.sin1.slickvpn.com) contains

# host/port of vpn server
remote gw1.sin1.slickvpn.com 443 udp
# prompt for authentication
auth-user-pass
# equivalent to pull, tls-client
client
# redirect all outgoing traffic to the vpn gateway
redirect-gateway
# verify the server certificate for authenticity
remote-cert-tls server
cipher AES-256-CBC
proto udp
dev tun
keepalive 10 120
nobind
persist-key
persist-tun
# ssl certificate / key used for tls
#ca certs/ca.crt

-BEGIN CERTIFICATE-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-END CERTIFICATE-


I see no pointer towards an external URL.  The ca CERT is in the file
itself.  Also the Miami file contains the same CA.cert as does the file
SlickVPNcrt referenced in the Ubuntu setup.

I also could see no difference in the Hong Kong, Miami, or Singapore
setups other than the name of the openVPN server.  Nothing to indicate a
PIN of any sort is needed.

[egreshko@meimei ~]$ ll gw1.*
-rw-rw-r--. 1 egreshko egreshko 1634 Feb 21 14:03 gw1.hkg2.slickvpn.com.ovpn
-rw-rw-r--. 1 egreshko egreshko 1634 Feb 21 13:44 gw1.mia4.slickvpn.com.ovpn
-rw-rw-r--. 1 egreshko egreshko 1634 Feb 20 15:50 gw1.sin1.slickvpn.com.ovpn

[egreshko@meimei ~]$ diff gw1.hkg2.slickvpn.com.ovpn
gw1.mia4.slickvpn.com.ovpn
2c2
< remote gw1.hkg2.slickvpn.com 443 udp
---
> remote gw1.mia4.slickvpn.com 443 udp

[egreshko@meimei ~]$ diff gw1.mia4.slickvpn.com.ovpn
gw1.sin1.slickvpn.com.ovpn
2c2
< remote gw1.mia4.slickvpn.com 443 udp
---
> remote gw1.sin1.slickvpn.com 443 udp


I wouldn't compare what is being done on the Windows side since they
seem to have their own client.

In any event, since you are their customer you may want to contact them
for support. 

-- 
Fedora Users List - The place to go to get others to do the work for you
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: Update broken depencies

[Ed Greshko]

On 02/21/17 13:23, jarmo wrote:
> Can't update because of broken depencies.
>
> 
>  Package  Arkkitehtuuri Versio
>  AsennuslähdeKoko
> 
> Upgrading:
>  iptables x86_641.6.0-3.fc25  
>  updates416 k
>  iptables-libsx86_641.6.0-3.fc25  
>  updates 87 k
> Asennetaan uudelleen:
>  iptables-servicesx86_641.6.0-2.fc25  
>  fedora  54 k
> Skipping packages with broken dependencies:
>  iptables-servicesx86_641.6.0-3.fc25  
>  updates 53 k
>
> Transaktion yhteenveto
> 
> Upgrade  2 Packages
> Skip 1 Package
>
> Anyone else?

No...

[egreshko@meimei ~]$ rpm -qa | grep iptables
iptables-libs-1.6.0-3.fc25.x86_64
iptables-1.6.0-3.fc25.x86_64
iptables-services-1.6.0-3.fc25.x86_64

run dnf again but specify --best to see what is blocking for you.


-- 
Fedora Users List - The place to go to get others to do the work for you
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Update broken depencies

[jarmo]

Can't update because of broken depencies.


 Package  Arkkitehtuuri Versio  
   AsennuslähdeKoko

Upgrading:
 iptables x86_641.6.0-3.fc25
   updates416 k
 iptables-libsx86_641.6.0-3.fc25
   updates 87 k
Asennetaan uudelleen:
 iptables-servicesx86_641.6.0-2.fc25
   fedora  54 k
Skipping packages with broken dependencies:
 iptables-servicesx86_641.6.0-3.fc25
   updates 53 k

Transaktion yhteenveto

Upgrade  2 Packages
Skip 1 Package

Anyone else?

Jarmo
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: Importing OpenVPN. ovpn file in Fedora 25

[Stephen Morris]

On 20/02/2017 18:34, Ed Greshko wrote:

On 02/20/17 12:53, Stephen Morris wrote:

 I'm using the Slickvpn provider for my vpn, which also uses
Openvpn. Under windows they provide a client which make connecting to
any of their servers, and changing between servers, and auto switching
between servers, child's play.
 They don't provide a client for linux but instead their web site
provides a link to instructions on how to set up Openvpn under Linux,
but it is aimed directly at Ubuntu. In order to get the vpn to work I
had to download a crt file that was configured for their server in
Singapore (as being the closest location to me), but there was no
keyfile. When I downloaded the .crt file I stored that file in my
downloads directory, and then in the vpn definition in networkmanager
I pointed the 'CA Certificates' option to the .crt file in my
downloaded location. Given you comments above, is that not how I
should be setting things up?

I looked at the Slickvpn site

 From what I could tell, all you should need do is download the openVPN
file for each site you want to connect with from this page
https://www.slickvpn.com/locations/.  Then change the name of the file
from *.bin to *.ovpn and use the Import function of NetworkManager.
When prompted to save the certificates you should say yes.

After the connection is imported you would then go to the connection and
choose "Edit".  There you'd add your UN/PW and say OK.  After that you
should be able to connect.  Looking at the openVPN config file it is
using simple UN/PW authentication.   I don't see any requirements for
additional keys.

The Connection Type is show as Password.  This is different than my VPN
provider which has a Connection Type of Password with Certificates (TLS)
and the openVPN config supplied by them includes a CA, and User certs
plus a private key and tls_auth key.
Thanks Ed. After sending the email I went to that site and their site 
has changed. I downloaded the .ovpn files their links were pointing at 
for Singapore and Miami and imported them into Networkmanager which set 
everything up, including pointing the CA Certificates entry at an 
external location URL to get the certificate, which is a different 
location for each server and is pointing at a .pem file. I found via 
this method that their Australian servers still don't work and neither 
does their Indian Server, and their Hong Kong server wants a pin in 
order to connect. From their windows client they have servers that are 
Core Servers and one's that are not, and it seems to be difficult to 
connect to their servers which are not Core Servers. When using their 
Windows Client and pointing it at a definition that is a group of 
servers, its auto selection seems to only select Core Servers and fails 
if the list doesn't include a Core Server. Unfortunately their location 
site doesn't indicate which are Core Servers and which aren't, unless 
the links they provide are only for Core Servers, as I noticed from 
their map that Miami has two servers but they only provide a link to 
download the .ovpn file for one of them.




 As a side issue to this, which may be the provider I am using,
when I use the vpn my network performance goes through the floor. I
have spoken to the provider about it and they suggested I try a couple
of server in the US, but their performance is worse. I get the same
issue under both Linux and Windows. The provider recommends that I not
use their servers in Australia because network costs are so expensive.
Do you notice similar performance issues with your setup, and if not,
how do I identify whether it is a provider issue or my setup?

What I would do is go to a site such as http://www.speedtest.net and in
the settings menu pick a destination server at the same location as your
VPN end point.  Then I would run the speed test with and without being
connected to the VPN.  I would run each test multiple times and take an
average.  I did this with one of my endpoints in Seattle.  With the VPN
it was slower but only by about 5%.

It is not clear that their servers support it...but you can also try
turning LZO compression on in the "Advanced" settings for the
connection.  I see a slight increase in my speed tests with this turned
on.  YMMV depending on the type of data being transferred.
I've turned on LZO compression to see what happens. I'm on cable and 
I've found that speedtest.net tends to be a bit hit or miss as to 
whether it connects or not, plus my ISP recommends to not use it for 
cable speed tests. I tried a site called ozbroadband.com which test 
speed rating by downloading an image file of a user selected size, and 
if I'm not connecting via the vpn I am able to download a larger file 
than if I am connecting via vpn. I'm on a 100Mbps link and when I speed 
tested yesterday without the vpn I tested at a line speed of 80.52 Mbps, 
with the Singapore vpn connection it drops to 35.9 Mbps and with the 
Miami vpn connection it drops to 4.5 

Re: question about packages

[Jon LaBadie]

On Mon, Feb 20, 2017 at 09:50:51PM +0100, john.david.r.smith wrote:
> hi.
> i can install the package 'vim' via `sudo dnf install vim`
> but there is no package called vim in the list returned by `dnf search vim`.
> 
> is there a way to get the real package name given the name vim?
> (i want a generic way to resolve such a name).
> - John

Two things come to mind.

  dnf list 'vim*' ordnf list '*vim*'

  dnf provides /usr/bin/vim

jon
-- 
Jon H. LaBadie  jo...@jgcomp.com
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Laptop, KVM, Auto bridge.. wifi and ethernet?

[InvalidPath]

So I run KVM on my laptop (Dell XPS 13 9550) with a Win10 guest.  Id there
any way to create a bridged interface that'd automatically detect and
bridge to ethernet or wireless, whichever one is available?
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: question about packages

[Cheng Rk]

then a question is why don't you (Fedora) make a a vim package since `dnf 
install vim` is such a high demand?___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: question about packages

[stan]

On Mon, 20 Feb 2017 21:50:51 +0100
"john.david.r.smith"  wrote:

> hi.
> i can install the package 'vim' via `sudo dnf install vim` but there
> is no package called vim in the list returned by `dnf search vim`. is
> there a way to get the real package name given the name vim? (i want
> a generic way to resolve such a name).

This will work if the package containing vim is installed on your
system.

locate vim

This gives the location of the vim executable, /usr/bin/vim.  There can
be a lot of output, so you could pipe it to less, locate vim | less

rpm -qf [path to vim from above command] or
rpm -qf /usr/bin/vim

This gives the name of the package that installed that executable.
vim-enhanced-8.0.329-1.fc25.x86_64

$ rpm -q --filesbypkg vim-enhanced-8.0.329-1.fc25.x86_64
vim-enhanced  /etc/profile.d/vim.csh
vim-enhanced  /etc/profile.d/vim.sh
vim-enhanced  /usr/bin/rvim
vim-enhanced  /usr/bin/vim
vim-enhanced  /usr/bin/vimdiff
vim-enhanced  /usr/bin/vimtutor

dnf info vim-enhanced will give general information about the package.

I am not aware of a way to query package contents if the package isn't
installed.  e.g. it is in a repository  Perhaps someone else will give
that answer.
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

question about packages

[john.david.r.smith]

hi.
i can install the package 'vim' via `sudo dnf install vim` but there is no 
package called vim in the list returned by `dnf search vim`.
is there a way to get the real package name given the name vim? (i want a 
generic way to resolve such a name).
- John
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: [OT] bash help

[Jon LaBadie]

On Mon, Feb 20, 2017 at 10:46:23AM +, Andy Blanchard wrote:
> On 20 February 2017 at 09:18, T_POL  wrote:
> >
> > not sure about that but I think the "cd" command executes indeed
> > but it's valid only for the scripts' environment and not for the
> > shell you started the script from.
> 
> *ding* *ding* *ding*  We have a winner!
> 
> Shells execute in their own instance of bash (or whatever) - a
> subshell.  The way to do this is with an alias or function, depending
> on your needs (aliases don't except parameters, functions do), so
> here's the function for "mdcd":
> 
> $ mdcd() {
> # additional checks here, per original post
> mkdir $1
> cd $1
> }
> 
> 
> Alternatively, you can force a script to run in the current shell by sourcing 
> it
> 
> $ . {scriptname}
> 
> Note the "." between the prompt and script (you can also use the more
> literal "source"):
> 
> $ source {scriptname}
> 
> Another option, just for completeness, would be to alias sourcing a
> script, e.g.:
> 
> $ alias mdcd='. {path to script}'
> 

A caveat I think the OP already found out.  Both the sourced
script and the function would be running in the users
interactive shell environment, it must for the cd to be effective.
Thus you do not want to "exit" from either the sourced
script nor from the function as that causes the interactive
shell to exit.  In the function you can use "return "
instead of "exit ".

Jon
-- 
Jon H. LaBadie  jo...@jgcomp.com
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: F25 : Quitting Remmina simply crashes Gnome-Shell

[Hoggins!]

Well, this is more serious than I thought... same behavior when I close
my VLC player.
And in journalctl, I got not much more than just :

"Process 1681 (gnome-shell) of user 1001 killed by SIGABRT - dumping core"


Le 20/02/2017 à 20:18, Hoggins! a écrit :
> Hey there,
>
> I have a fairly embarrassing problem : just upgraded from F24 to F25
> using 'dnf system-upgrade'. Everything works as smooth as expected...
> except... when I close Remmina OR when I lock my screen while Remmina is
> running, Gnome-Shell crashes and gets me back to the login screen,
> losing everything I had on screen.
>
> Have you experienced this ? What could I provide to help solve this ?
>
> Thanks !
>
> Hoggins!
>
>
>
>
>
> ___
> users mailing list -- users@lists.fedoraproject.org
> To unsubscribe send an email to users-le...@lists.fedoraproject.org




signature.asc
Description: OpenPGP digital signature
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

F25 : Quitting Remmina simply crashes Gnome-Shell

[Hoggins!]

Hey there,

I have a fairly embarrassing problem : just upgraded from F24 to F25
using 'dnf system-upgrade'. Everything works as smooth as expected...
except... when I close Remmina OR when I lock my screen while Remmina is
running, Gnome-Shell crashes and gets me back to the login screen,
losing everything I had on screen.

Have you experienced this ? What could I provide to help solve this ?

Thanks !

Hoggins!





signature.asc
Description: OpenPGP digital signature
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: Personal disputs in bugzilla (bug 1401166 )

[Neil Thompson]

At the very least, I think the package maintainer needs to be spoken to by
fedora people - swearing at bug reporters on the Fedora Bugzilla is
probably not "being excellent to each other".

On 20 February 2017 at 10:52, Clemens Eisserer  wrote:

> Hi there,
>
> I frequently get bitten by bug *1401166*
> .
> The author closed the bug as WONTFIX because of some personal issues he
> had/has with the original reporter.
>
> Could somebody with the permissions please reopen the bug, so other users
> who run into this problem can report?
>
> Thanks, Clemens
>
> ___
> users mailing list -- users@lists.fedoraproject.org
> To unsubscribe send an email to users-le...@lists.fedoraproject.org
>
>
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: [OT] bash help

[Andy Blanchard]

On 20 February 2017 at 09:18, T_POL  wrote:
>
> not sure about that but I think the "cd" command executes indeed
> but it's valid only for the scripts' environment and not for the
> shell you started the script from.

*ding* *ding* *ding*  We have a winner!

Shells execute in their own instance of bash (or whatever) - a
subshell.  The way to do this is with an alias or function, depending
on your needs (aliases don't except parameters, functions do), so
here's the function for "mdcd":

$ mdcd() {
# additional checks here, per original post
mkdir $1
cd $1
}


Alternatively, you can force a script to run in the current shell by sourcing it

$ . {scriptname}

Note the "." between the prompt and script (you can also use the more
literal "source"):

$ source {scriptname}

Another option, just for completeness, would be to alias sourcing a
script, e.g.:

$ alias mdcd='. {path to script}'

-- 
Andy

The only person to have all his work done by Friday was Robinson Crusoe
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: [OT] bash help

[T_POL]

On Sun, 19 Feb 2017 05:22:44 -0800
Mike Wright  wrote:

> Hi all,
> 
> My brain cell ran away from home.  I have an incredibly simple script 
> that doesn't do what I expect.  I use "mkdir DIR; cd DIR" a lot so I'm 
> trying to put it in a script: "~/bin/mdcd".
> 
> After checking that $1 exists:
> 
> dir="$1"
> mkdir -p "$dir"
> cd "$dir"   <-- never executes
> 
> The directory is created so there is no error there.
> 
> Huh? Insight anyone?
> 
> Thanks,
> Mike Wright
> ___
> users mailing list -- users@lists.fedoraproject.org
> To unsubscribe send an email to users-le...@lists.fedoraproject.org


Hi

not sure about that but I think the "cd" command executes indeed 
but it's valid only for the scripts' environment and not for the 
shell you started the script from.

If it's not so, may be my brain cells are gone too.

Ciao
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: [OT] bash help

[Dave Mitchell]

On Sun, Feb 19, 2017 at 02:12:35PM -0800, Joe Zeff wrote:
> On 02/19/2017 05:22 AM, Mike Wright wrote:
> > dir="$1"
> > mkdir -p "$dir"
> > cd "$dir"   <-- never executes


> I'm not sure why this doesn't work but you're doing it the hard way.
> 
> mkdir -p $1
> cd $1
> 
> is much simpler, and might work better.

Unless $1 contains any spaces, in which case the mkdir will create two or
more directories; e.g.:

$ p="a b"
$ mkdir $p
$ ls -l
total 8
drwxrwxr-x. 2 davem davem 4096 Feb 20 09:07 a
drwxrwxr-x. 2 davem davem 4096 Feb 20 09:07 b
$


-- 
"You may not work around any technical limitations in the software"
-- Windows Vista license
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Personal disputs in bugzilla (bug 1401166 )

[Clemens Eisserer]

Hi there,

I frequently get bitten by bug *1401166*
.
The author closed the bug as WONTFIX because of some personal issues he
had/has with the original reporter.

Could somebody with the permissions please reopen the bug, so other users
who run into this problem can report?

Thanks, Clemens
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org

Re: Importing OpenVPN. ovpn file in Fedora 25

[Ed Greshko]

On 02/20/17 12:53, Stephen Morris wrote:
> I'm using the Slickvpn provider for my vpn, which also uses
> Openvpn. Under windows they provide a client which make connecting to
> any of their servers, and changing between servers, and auto switching
> between servers, child's play.
> They don't provide a client for linux but instead their web site
> provides a link to instructions on how to set up Openvpn under Linux,
> but it is aimed directly at Ubuntu. In order to get the vpn to work I
> had to download a crt file that was configured for their server in
> Singapore (as being the closest location to me), but there was no
> keyfile. When I downloaded the .crt file I stored that file in my
> downloads directory, and then in the vpn definition in networkmanager
> I pointed the 'CA Certificates' option to the .crt file in my
> downloaded location. Given you comments above, is that not how I
> should be setting things up?

I looked at the Slickvpn site

From what I could tell, all you should need do is download the openVPN
file for each site you want to connect with from this page
https://www.slickvpn.com/locations/.  Then change the name of the file
from *.bin to *.ovpn and use the Import function of NetworkManager. 
When prompted to save the certificates you should say yes.

After the connection is imported you would then go to the connection and
choose "Edit".  There you'd add your UN/PW and say OK.  After that you
should be able to connect.  Looking at the openVPN config file it is
using simple UN/PW authentication.   I don't see any requirements for
additional keys.

The Connection Type is show as Password.  This is different than my VPN
provider which has a Connection Type of Password with Certificates (TLS)
and the openVPN config supplied by them includes a CA, and User certs
plus a private key and tls_auth key.


> As a side issue to this, which may be the provider I am using,
> when I use the vpn my network performance goes through the floor. I
> have spoken to the provider about it and they suggested I try a couple
> of server in the US, but their performance is worse. I get the same
> issue under both Linux and Windows. The provider recommends that I not
> use their servers in Australia because network costs are so expensive.
> Do you notice similar performance issues with your setup, and if not,
> how do I identify whether it is a provider issue or my setup?

What I would do is go to a site such as http://www.speedtest.net and in
the settings menu pick a destination server at the same location as your
VPN end point.  Then I would run the speed test with and without being
connected to the VPN.  I would run each test multiple times and take an
average.  I did this with one of my endpoints in Seattle.  With the VPN
it was slower but only by about 5%.

It is not clear that their servers support it...but you can also try
turning LZO compression on in the "Advanced" settings for the
connection.  I see a slight increase in my speed tests with this turned
on.  YMMV depending on the type of data being transferred.

 


-- 
Fedora Users List - The place to go to get others to do the work for you
___
users mailing list -- users@lists.fedoraproject.org
To unsubscribe send an email to users-le...@lists.fedoraproject.org