The "server response time" for git clone operations varies depending on client network bandwidth

[Ravi GH]

I was doing a load test and found that the "server response time" for a baseline clone operations of 30 MB repo is not constant.

Clone time detail Before load Test :

time git clone ssh://git@mycompayurl.com:7999/project1/loadtest_repo.git

Cloning into 'loadtest_repo...

remote: Enumerating objects: 26260, done.

remote: Counting objects: 100% (26260/26260), done.

remote: Compressing objects: 100% (9024/9024), done.

remote: Total 26260 (delta 17450), reused 24565 (delta 16555)/s

Receiving objects: 100% (26260/26260), 29.16 MiB | 188.00 KiB/s, done.

Resolving deltas: 100% (17450/17450), done.

real 2m43.301s user 0m1.829s sys 0m1.306s

The clone time was 2m43 Sec ( on User Laptop)

The server response time was 135957 (2.26 m)

Log trace:  

172.19.x.x | ssh | o*CVW0H6x1000x18849x15 | ravrao | 2020-04-02 16:42:23,265 | SSH - git-upload-pack '/project1/loadtest_repo.git' | "SSH-2.0-OpenSSH_7.9" | 0 | 4973 | 30623556 | cache:miss, clone, protocol:1, ssh:user:id:18160 | 135957 | 1j0141y |

Question : 

 in above log trace 135957 is the server response time, this value will be printed only after "Receiving objects: 100% " step completed at the client /user end.

My observation is if the client download speed decrease , the total clone time /server response Time increases.

is that mean, the server response time also considers client  n/w bandwidth and then prints the time take to complete the clone operation?

Server Response time means only server internal process time OR the time taken to process the objects + delivering the objects to the enduser request and then prints the server response time?

is Server Response time workflow to prints the value in application log is same for HTTPS & SSH?

can someone share the documents on "Server Response time" available in tomcat /BB application log.

Thanks & Regards,

ravi

[Gonchik Tsymzhitov]

Hi!

If we will have the same repo, I can try to calculate