Logging OpenBSD virtual machine console output

[J65nko]

A shell script to start an OpenBSD VM and log the console input/output to an unique log file
The name of the logs start with the VM name followed by DDMM_HHMMSS.log

18259 Aug  1 04:15 vmlogs/alix_new0731_221336.log
11549 Aug  9 04:39 vmlogs/alix_new0809_043239.log

This name is created with:

LOGFILE="${LOGDIR}/${VMNAME}$(date -j   "+%m%d_%H%M%S").log"

For customizing this format, see date(1) and strftime(3)
Some alternatives shown on the shell command line:

$ date -j   "+%m%d_%H:%M:%S"                                  
0809_05:37:05

$ date -j   "+%m%d_%H_%M_%S" 
0809_05_37_56

$ date -j   "+%b%d_%H_%M_%S"
Aug09_05_43_59

The following snippet from the script shows how you could start the script:

# if variable VMNAME is not defined you will be prompted to select a VM

# USAGE options:

# 1) $ env VMNAME=alix_new ./dovm.sh
# 2) $ export VMNAME=alix_new ; ./dovm.sh
# 3) Define the shell variable VMNAME in the script
# 4) Just run './dovm.sh' and select a stopped VM from a menu
# 5) If you place it in your $HOMEDIR/bin directory: $ dovm.sh

Using the interactive method 4:

$ dovm.sh
  
   ID   PID VCPUS  MAXMEM  CURMEM     TTY        OWNER    STATE NAME
    1     -     1    1.0G       -       - adriaan:wheel  stopped df-us
    2     -     1    1.0G       -       - adriaan:wheel  stopped siralas-clone
    3     -     1    1.0G       -       - adriaan:wheel  stopped hetzner
    4     -     1    500M       -       - adriaan:wheel  stopped alix_new
Select 'ID' number of VM to start : 3
Script started on Tue Aug  9 05:56:44 2022
Connected to /dev/ttypb (speed 115200)
Using drive 0, partition 3.
Loading......
probing: pc0 com0 mem[638K 1022 a20=on]
disk: hd0+
>> OpenBSD/amd64 BOOT 3.53

com0: 115200 baud
switching console to com0
>> OpenBSD/amd64 BOOT 3.53
[snip]
root@df-us[~]shutdown -hp now
Shutdown NOW!
shutdown: [pid 33853]
                                                                               
*** FINAL System shutdown message from root@df-us.utp.xnet ***               
System going down IMMEDIATELY                                                  
                                                                               
                                                                               
root@df-us[~]
System shutdown time has arrived

root@df-us[~]syncing disks... done
vmmci0: powerdown

The operating system has halted.
Please press any key to reboot.

Instead of rebooting we press '~." ( tilde followed by a dot) to exit the console.

[EOT]
Script done, output file is /home/adriaan/vmlogs/hetzner0809_055644.log

You will then see a hint to use col(1) to sanitize 'download progress meter' lines:

    ------------------------------------------------------------------------------
HINT: You can use col(1) to filter out backspaces on messy lines like this:
    ------------------------------------------------------------------------------
    [snip]
    ^MGet/Verify SHA256.sig     0% |                          |     0
    --:-- ETA^MGet/Verify SHA256.sig   100% |**************************|
    2144       00:00 Signature Verified ^MGet/Verify bsd            0%
    |                          |     0       --:-- ETA^MGet/Verify bsd
    8% |**                        |  1792 KB    00:11 ETA^MGet/Verify
    bsd           17% |****
    [snip]

    For example:

    $ col -b <df-us0724_023205.log | less

    [snip]
    Get/Verify SHA256.sig   100% |**************************|  2144       00:00 ETA
    Signature Verified
    Get/Verify bsd          100% |**************************| 22339 KB    00:11 ETA
    Get/Verify bsd.rd       100% |**************************|  4498 KB    00:02 ETA
    [snip]

    Neat huh?

You can disable this hint by setting the variable SHOW_COL_HINT to 0:

#SHOW_COL_HINT=0
SHOW_COL_HINT=1