Re: git: aa34b1d20e44 - main - vmrun.sh: Add arm64 support
- In reply to: John Baldwin : "Re: git: aa34b1d20e44 - main - vmrun.sh: Add arm64 support"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Sat, 04 May 2024 18:18:24 UTC
On 4 May 2024, at 13:02, John Baldwin wrote: > On 5/3/24 11:01 PM, Lexi Winter wrote: >> John Baldwin: >>> On 5/3/24 7:58 AM, Gleb Smirnoff wrote: >>>> The share/examples/bhyve/vmrun.sh is not really an example, but a good >>>> standalone tool. I use it all the time without any modification, and I >>>> guess many other people do, too. >>>> >>>> Maybe time to move it to some path that is in the default $PATH? >>> >>> I really think we should instead be building a tool like vm-bhyve from ports, >>> either by importing that directly, or perhaps rewriting it in lua and having >>> that eventually replace vmrun.sh. >> >> is this what vmstated[0] is supposed to be? this was mentioned in the >> last status report[1] but it's not entirely clear if this is meant to >> replace vm-bhyve, or be something more low level. >> >> [0] https://github.com/christian-moerz/vmstated >> [1] https://www.freebsd.org/status/report-2024-01-2024-03/#_bhyve_improvements > > I'm unaware of vmstated. Checking its manpage it does not yet seem to have > feature parity with vm-bhyve. Many of the things I think you want a VM > manager to do are also perhaps a bit harder to do in C (e.g. forking and > execing tmux to access the serial console) compared to a scripting language. +1 for vm-bhyve. I used an old copy of vmrun.sh for a long time, but vm-bhyve was a big improvement. Mike > Note that jail management today via jails.conf doesn't require a running > daemon, so that seems a bit curious, though perhaps the goal is to handle > auto-restarting if a VM exits unexpectedly? > > -- > John Baldwin