[MDEV-19277] Add status variable that gets incremented if connection is aborted prior to authentication Created: 2019-04-18 Updated: 2020-08-25 Resolved: 2019-05-14 |
|
| Status: | Closed |
| Project: | MariaDB Server |
| Component/s: | Authentication and Privilege System, Variables |
| Fix Version/s: | 10.4.5 |
| Type: | Task | Priority: | Major |
| Reporter: | Geoff Montee (Inactive) | Assignee: | Oleksandr Byelkin |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Issue Links: |
|
||||||||||||||||
| Description |
|
If a connection is aborted prior to authentication, then the only status variable that gets incremented is Aborted_connects. The Aborted_connects status variable gets incremented for a lot of reasons though, so there is no status variable that can be used to determine how many connections have gotten aborted prior to authentication. You can reproduce this by doing something like using telnet to connect to the MariaDB port, and then killing the telnet process:
The only status variable that is incremented from this is Aborted_connects:
|
| Comments |
| Comment by Oleksandr Byelkin [ 2019-05-09 ] |
|
Is Aborted_nonauth_connects a good name? |
| Comment by Oleksandr Byelkin [ 2019-05-09 ] |
|
I do not think that we should change value and meaning of Aborted_connects , but just add new variable which will count aborted with no authentication additionally? |
| Comment by Geoff Montee (Inactive) [ 2019-05-09 ] |
|
What do you think about Aborted_connects_preauth? Aborted_nonauth_connects would work too though. I agree with you about the behavior of Aborted_connects. |
| Comment by Oleksandr Byelkin [ 2019-05-09 ] |
|
OK, it will be Aborted_connects_preauth, thanks for helping with the name |
| Comment by Oleksandr Byelkin [ 2019-05-09 ] |
|
OK, the last question, does it matter if the connection closed with error or closed without an error? The connection can be closed because of error (probably timeout belong here), because was killed or because actually connection closed, but we are only interested if it was closed unauthorized, correct? |
| Comment by Geoff Montee (Inactive) [ 2019-05-10 ] |
|
I don't think it should matter if the connection was closed with or without an error. I think this status variable should only be concerned with whether the connection was closed prior to performing authentication--regardless of whether an error was involved. |
| Comment by Oleksandr Byelkin [ 2019-05-14 ] |
|
the same patch with |
| Comment by Vladislav Vaintroub [ 2019-05-14 ] |
|
Looks good to me, ok to push |