This is the mail archive of the
mailing list for the Cygwin project.
Issue while running Visual Studio's devenv through ssh
- From: The General <the dot third dot general at gmail dot com>
- To: cygwin at cygwin dot com
- Date: Fri, 15 May 2015 15:08:46 -0400
- Subject: Issue while running Visual Studio's devenv through ssh
- Authentication-results: sourceware.org; auth=none
I have some code that needs to be built on a windows machine (W), but
my primary development/deploy environment is an Ubuntu server (U). The
goal is to fully automate the build/deploy process so that it can be
kicked off from U.
Before I describe the problem let me explain my process.
U is an Ubuntu 12.04.5 VM running on ESXI
W is a Windows 7 VM (Version 6.1.7601) running on ESXI, it has Cygwin
installed (with the openssl and git packages included) and UAC
I am running the following command from U:
ssh firstname.lastname@example.org 'cd
/cygdrive/c/Users/win_user/git_repo/windows_code && ./build.bat'
I have set up public key authentication from U to W, but I also
experience the below issue when I include the '-o
build.bat is stored locally on W and looks like this:
call "C:\Program Files (x86)\Microsoft Visual Studio
devenv /nologo "windows_code.sln" /Build "Release|x64"
The git pull command is pulling changes from a remote git server (G).
I have also set up public key authentication from W to G, disabling
this doesn't seem to have any effect.
After I got the above working I took a snapshot of W. If I reset to
that snapshot and run the ssh command on U everything works. If I
leave W idle for a few hours and try again I get the following from
1>------ Rebuild All started: Project: windows_code, Configuration:
Release x64 ------
1>Stdafx.cpp : fatal error C1902: Program database manager mismatch;
please check your installation
Right now I'm at a loss as to why this will work after a snapshot
reset, but will stop working if the machine is left alone for awhile.
Resetting to the most recent snapshot, rebooting the VM, and then
running the ssh command from U will also generate the above error.
A temporary solution is to manually reset the VM before kicking off
the job, but I'd like to figure out why it stops working after awhile.
Any feedback is much appreciated, and if anyone needs more information
please don't hesitate to ask.
Problem reports: http://cygwin.com/problems.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple