https://bz.apache.org/bugzilla/show_bug.cgi?id=65776
Mark Thomas <ma...@apache.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
--- Comment #12 from Mark Thomas <ma...@apache.org> ---
There are multiple issues being report here so i will try and address each in
turn.
The original issue reported by Johhny Lim
- The log message is correct - you've hit the OS bug
- The issue appears to affect mutliple (most?, all?) Linux distributions
This should be considered resolved as "working as designed".
The issue reported by 2022-01-14 03:39:38 GMT
- Can be avoided by configuration see comment #6
- Modifying the existing check is not practical see comment #7
This should be considered resolved as "Won't fix"
The issue on Windows reported by Brooke T. Hedrick
- The memory allocation error running with WSL (comment #11) indicates a
different issue
- I have ported the original test case to Windows and I have not yet been able
to recreate the error
I have a few more tests I want to run. I want to try and tweak the test case
and the Windows network config to try and increase the load on accept().
I do have a couple of questions regarding the Windows system where this issue
was observed:
- Bare metal or VM?
- If VM, which hypervisor?
- What sort of load (in terms of new connections per second) was the system
under at the time?
Moving status to NEEDINFO to reflect the status of this one remaining
unresolved issue.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
