# "vtkSocketCommunicator::GetVersion() returns different values on the two connecting processes"
**URL:** https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347
**Category:** ParaView Support
**Tags:** vtk
**Created:** [January 16, 2020, 9:21am UTC](https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347 "2020-01-16T09:21:35Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@XavierStuvw](https://discourse.paraview.org/u/XavierStuvw)
#### Post date: [January 16, 2020, 9:21am UTC](https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347/1 "2020-01-16T09:21:35Z")
What does an error message “vtkSocketCommunicator::GetVersion() returns different values on the two connecting processes” refer to? This post is a branch of [Reverse client-server connection fails with more than one MPI process](https://discourse.paraview.org/t/reverse-client-server-connection-fails-with-more-than-one-mpi-process/3207).
This occurs when I try to set up a reverse connection between client and server within a network and parallel jobs fall back to serial – see above post for the mechanics of the error. Here I am interested in understanding what the trigger of this particular mismatch is, and seeing what I could tune to work around this. For example, whose program does GetVersion() refer to?
At server and client side I work with same architectures, operating systems and compilers: x86\_64, Debian GNU/Linux 10 (buster), gcc (Debian 8.3.0-6) 8.3.0. Paraview is a release build of 5.7.0 out of the Superbuild at [https://gitlab.kitware.com/paraview/paraview-superbuild/blob/master/README.md](https://gitlab.kitware.com/paraview/paraview-superbuild/blob/master/README.md). The repository has been cloned separately on client and server.
There is little and old information about this sort of error on the internet, but I could not get usable cues out of it:
- 2016, issue tracker of VTK: [https://vtk.org/Bug/view.php?id=16094](https://vtk.org/Bug/view.php?id=16094)
- 2013, paraview narchive: [https://paraview.paraview.narkive.com/ogEJvINP/connecting-to-pvserver](https://paraview.paraview.narkive.com/ogEJvINP/connecting-to-pvserver)
- 2009, [https://www.paraview.org/pipermail/paraview/2009-February/011237.html](https://www.paraview.org/pipermail/paraview/2009-February/011237.html)
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [January 16, 2020, 12:15pm UTC](https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347/2 "2020-01-16T12:15:31Z")
[https://gitlab.kitware.com/paraview/paraview/issues/18171](https://gitlab.kitware.com/paraview/paraview/issues/18171)
---
### Author:  [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [January 16, 2020, 12:17pm UTC](https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347/3 "2020-01-16T12:17:13Z")
This error is a red herring from another error and should be improved.
---
### Author:  [@XavierStuvw](https://discourse.paraview.org/u/XavierStuvw)
#### Post date: [January 16, 2020, 1:33pm UTC](https://discourse.paraview.org/t/vtksocketcommunicator-getversion-returns-different-values-on-the-two-connecting-processes/3347/4 "2020-01-16T13:33:59Z")
Thanks for pointing out that the message in itself is an issue. Hopefully it will be fixed soon. Since deciphering that message has no purpose, the [vtkSocketCommunicator errors: "Could not receive tag" and "Endian handshake failed"](https://discourse.paraview.org/t/vtksocketcommunicator-errors-could-not-receive-tag-and-endian-handshake-failed/3350) are then the candidate troublemakers.