other: 0.213
semantic: 0.117
device: 0.109
graphic: 0.078
PID: 0.072
performance: 0.064
socket: 0.058
debug: 0.053
network: 0.053
files: 0.042
boot: 0.037
vnc: 0.037
KVM: 0.035
permissions: 0.033
debug: 0.195
other: 0.147
socket: 0.124
network: 0.099
PID: 0.081
semantic: 0.070
files: 0.068
device: 0.058
performance: 0.042
boot: 0.030
graphic: 0.028
permissions: 0.026
vnc: 0.017
KVM: 0.016

Serial communication between VMs problematic

Hello,

I want to setup serial communication between VM hosts but I have found it quite difficult...:

...because when trying unix sockets:

- host A has serial device as unix socket (bind)
- host B has serial device as client of unix socket
- host A is down thus not unix socket does exist
- host B can't be started because cannot read the socket:

error: Failed to start domain opd1s02
error: internal error Process exited while reading console log output: char device redirected to /dev/pts/0
connect(unix:/tmp/test.sock): Connection refused
chardev: opening backend "socket" failed

Can that work like the cable is not plugged in? So host B can start and when the socket would exist it would connect to it?

...and when using pty and host device combination one cannot predict pty device under /dev/pts, it would be nice if would be possible to define exact device name.

Tested on Fedora 14.

The QEMU project is currently considering to move its bug tracking to another system. For this we need to know which bugs are still valid and which could be closed already. Thus we are setting older bugs to "Incomplete" now.
If you still think this bug report here is valid, then please switch the state back to "New" within the next 60 days, otherwise this report will be marked as "Expired". Or mark it as "Fix Released" if the problem has been solved with a newer version of QEMU already. Thank you and sorry for the inconvenience.


[Expired for QEMU because there has been no activity for 60 days.]