# Connection failed during handshake

**URL:** https://discourse.paraview.org/t/connection-failed-during-handshake/10278
**Category:** ParaView Support
**Created:** [September 8, 2022, 3:16pm UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278 "2022-09-08T15:16:00Z")
**Posts on this page:** 15
**Page:** 1

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 8, 2022, 3:16pm UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/1 "2022-09-08T15:16:00Z")

</div>

I am running pvserver on remote linux machine (paraview v5.8.0)  
but when I try to connect on the windows PC with same version of paraview  
Connection failed during handshake

I can get 3.98.1 and 4.1.0 to connect,  
but then they crash out when I try to load the temp.foam file.

```auto
Waiting for client...
Connection URL: cs://giles-calc-machine:11111
Accepting connection(s): giles-calc-machine:11111
Client connected.
ERROR: In /home/utkarsh/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/Rendering/OpenGL/vtkXOpenGLRenderWindow.cxx, line 530
vtkXOpenGLRenderWindow (0x17593f0): bad X server connection. DISPLAY=ERROR: In /home/utkarsh/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/Rendering/OpenGL/vtkXOpenGLRenderWindow.cxx, line 322
vtkXOpenGLRenderWindow (0x17593f0): bad X server connection. DISPLAY=Segmentation fault (core dumped)

```

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 2:11am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/2 "2022-09-09T02:11:41Z")

</div>

Looks like you have a X configuration issue. Did you set the DISPLAY var on your remote machine ? Is that a Xorg server running ?

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 9:54am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/3 "2022-09-09T09:54:42Z")

</div>

Thanks Matt,

no, I have not set up any DISPLAY yet.

I’m using exactly the same procedure that I use on my laptop (which works fine in pv3 and pv4), but for some reason it doesnt work from my workstation.

Normally I can only connect using pv3 or pv4, but in this case it does connect,

and the pv window comes up, but as soon as I try to load the .foam file it crashes out.

Ideally I would like to be using pv5.8.0, but the handshake doesnt work for that one.

(on any of my machines).

I will try your suggestion about the DISPLAY seeting - what do you use

export DISPLAY=localhost:0.0 (?)

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 9:57am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/4 "2022-09-09T09:57:51Z")

</div>

DISPLAY depends on your machine, you can find it using ps:

```auto
[glow@frollo ~]$ ps -ef | grep Xorg
glow 1039 1038 2 11:16 tty1 00:11:52 /usr/lib/Xorg -nolisten tcp :0 vt1 -keeptty -auth /tmp/serverauth.ToFiCxwsH7
glow 171907 170142 0 17:57 pts/2 00:00:00 grep Xorg

```

In my case, I should use

`DISPLAY=:0`

Best,

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 10:06am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/5 "2022-09-09T10:06:02Z")

</div>

sorry I dont know what the grep command is telling me…

```auto
ubuntu@giles-calc-machine:~/working/bldg-simpleFoam-fstream-cylinder$ ps -ef | grep Xorg
ubuntu 326643 325813 0 10:05 pts/0 00:00:00 grep --color=auto Xorg

```

I just tried setting display and I still get the same coredump, when loading a file.

```auto
No protocol specified
ERROR: In /home/utkarsh/Dashboards/MyTests/NightlyMaster/ParaViewSuperbuild-Release/paraview/src/paraview/VTK/Rendering/OpenGL/vtkXOpenGLRenderWindow.cxx, line 322
vtkXOpenGLRenderWindow (0x14543f0): bad X server connection. DISPLAY=:0

```

Segmentation fault (core dumped)

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 10:16am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/6 "2022-09-09T10:16:15Z")

</div>

Simple, you have no Xorg running on your server. You cant use the OpenGL version of binary without a Xorg server running.

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 10:19am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/7 "2022-09-09T10:19:47Z")

</div>

> [@mwestphal](#):
>
> `ps -ef | grep Xorg`

what I mean is that is works fine when Im connecting via windows laptop to linux machine…

but when connecting from my windows workstation to linux machine, its not working.

so Im assuming something not right with my Windows workstation ?

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 10:21am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/8 "2022-09-09T10:21:39Z")

</div>

I’m confused, please state exactly what you are doing, what is working and what is not.

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 10:28am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/9 "2022-09-09T10:28:19Z")

</div>

ok. I connect windows laptop to Linux (ubuntu) workstation.  
then start pvserver in ubuntu. then start paraview on windows laptop.  
then I can connect (in paraview), and open files, and it works fine. all good.

But the doing same from my windows workstation doesnt work,  
although I can connect (in paraview), and see the files in Linux workstation,  
but as soon as try to open one of them, paraview crashes out with that error message, and coredump.

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 10:36am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/10 "2022-09-09T10:36:39Z")

</div>

I’d start by deleting your local paraview settings.

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 10:42am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/11 "2022-09-09T10:42:51Z")

</div>

sorry - Im not sure what you mean by that…?

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 9, 2022, 10:44am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/12 "2022-09-09T10:44:55Z")

</div>

- Edit → Settings → Restore Defaults  
 ![a](https://discourse.paraview.org/uploads/default/original/2X/6/643243fd9d7387b7abb31ac125e85dd05eb3e588.png)

- restart paraview

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 12:45pm UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/13 "2022-09-09T12:45:39Z")

</div>

that doesnt make much difference. BUT I have just discovered that its only the larger cases that it crashes out on, so maybe its more of a graphics RAM issue. I dont know how much RAM is required for rendering larger meshes, although this case Im using now is only 200k cells and that is crashing out.

---

<div class="post-metadata">

### Author: ![garcfd](https://discourse.paraview.org/letter_avatar_proxy/v4/letter/g/ccd318/32.png) [@garcfd](https://discourse.paraview.org/u/garcfd)
#### Post date: [September 9, 2022, 1:43pm UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/14 "2022-09-09T13:43:20Z")

</div>

Matt,  
I think your were right in the first place - it seems to be a lack of xserver thats the problem.  
so we need to install the osmesa version in 5.11.0 then?

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [September 12, 2022, 5:54am UTC](https://discourse.paraview.org/t/connection-failed-during-handshake/10278/15 "2022-09-12T05:54:09Z")

</div>

egl or osmesa version should do the trick.
