--username option usage

[son9701]

I want to use --useroption but I don't know how to use it.
Do I need to set it up on the server before the client can use it?

and Is there a reason why I can't see the --username option with -h in a Linux environment?

[davidBar-On]

--username is available only if an OpenSSL library was detected during the iperf3 build. If you built iperf3 yourself search for "ssl" in ./config.log (created when your run ./configure) for a related error/warning messages.

If OpenSSL library library is available, you may try building iperf3 using ./configure --with-openssl=<root of the OpenSSL directory>.

[son9701]

Now build with openssl and you will see that option
I have a few more questions

1. --rsa-public-key-path public.pem --username @@ It starts when any value is entered. Is there a way to set the username and password?
2. Can I use --username only with --rsa-public-key-path public.pem ?
3. The option is a client option, but the server cannot set username and password?
   Kindly explain please

[davidBar-On]

On the client side, the user name is set by the--username and the password is set either by the IPERF3_PASSWORD environment variable, or iperf3 will prompt for it. When using --username, --rsa-public-key-path must be set also.

On the server side, the following two parameters must be set for the authorization to work:

1. --rsa-private-key-path - the private key path.
2. --authorized-users-path - the path to the file that includes the user names and their password. This Oracle page has a good explanation about this file format and an example of generating it (see both the options descriptions and the EXAMPLES section).

[son9701]

Thanks for the kind explanation, and i have one more question.

I tried to create a key on the server and test it as explained on the Oracle page, but at least any path after --authorized-users-path is executed. is this normal?

And when I try to execute it with the values ​​set by S_ USER and S_ PASSWD, iperf cannot be executed with authorization fail. How can I solve this?

[davidBar-On]

I assume the "a" file is the "credentials.csv" file in the Oracle example. Is this correct? (Note that they have a typo there: the "cat credentials.csv" should be "cat >credentials.csv".)

Did you create the corresponding "public.pem" file and set the --username and --rsa-public-key-path when you run the client? Note that the username and its password should be the same as you defined for the "a" file.

[son9701]

The file named "a" does not exist, but it worked and left a question.

and when > S_USER=mario S_PASSWD=rossi was typed, file named S_USER=mario is created. Can I modify and use this file?

plus PEM file was created in order openssl genrsa -des3 -out private.pem 2048, openssl rsa -in private.pem -outform PEM -pubout -out public.pem

[davidBar-On]

Seems that you have not created and used the authorized users file correctly. You should first create the file. In the Oracle example the username is "mario" and the password is "rossi". Per that example, first create the password hash:

> S_USER=mario S_PASSWD=rossi
> echo -n "{$S_USER}$S_PASSWD" | sha256sum | awk '{ print $1 }'

Then create the authorized users file "credentials.csv", where "bf7a49a846d44b454a5d11e7acfaf13d138bbe0b7483aa3e050879700572709b" is the password hash from the above:

> cat >credentials.csv
# file format: username,sha256
mario,bf7a49a846d44b454a5d11e7acfaf13d138bbe0b7483aa3e050879700572709b

Now, run the server using --authorized-users-path credentials.csv and the client using `--username mario". Answer "rossi" the client's prompt for password.

[son9701]

Thanks, it's solved.