On 13/01/2025 19:57, Rémy Maucherat wrote:
On Mon, Jan 13, 2025 at 8:53 PM Rémy Maucherat <r...@apache.org> wrote:
On Mon, Jan 13, 2025 at 8:39 PM Mark Thomas <ma...@apache.org> wrote:
On 13/01/2025 08:10, Rémy Maucherat wrote:
On Thu, Jan 9, 2025 at 4:31 PM Mark Thomas <ma...@apache.org> wrote:
On 09/01/2025 14:53, Rémy Maucherat wrote:
On Thu, Jan 9, 2025 at 3:17 PM Mark Thomas <ma...@apache.org> wrote:
<snip/>
My current plan is to create InputBuffer with bb set to a zero length
ByteBuffer and have recycle (re)set it to a zero length ByteBuffer. That
avoids the NPEs, avoids retaining references unnecessarily but (so far)
doesn't break anything.
I have some more testing to do but hope to have something to commit by
the end of today.
That sounds like a good plan, hopefully with no hidden issues.
The only issue I found so far was the with SavedRequestInputFilter which
should have been fixed by the commit at the start of this thread.
Since this code is likely to be executed on every request I'm going to
take things relatively slowly in terms of back-ports. My current plan is:
- test on the platforms I normally test for releases
- wait for at least one Gump, BuildBot and GitHub action run
Assuming all is well, I'll back-port to 11.0.x. Back-ports to 10.1.x and
9.0.x will wait at least one release.
The patch series seems broken. Revert and retry later ?
What makes you say that? I'm still working through the backlog that
built up over the holidays.
The testsuite does not work anymore here (I had the impression there
were more failures, but seems not, so it's likely only a 9.0 backport
issue):
https://github.com/apache/tomcat/actions/runs/12744454326
Found these also from the full testsuite:
https://nightlies.apache.org/tomcat/tomcat-12.0.x/logs/317/TEST-jakarta.servlet.http.TestHttpServletDoHeadValidWrite0.NIO.txt
I'm not seeing any of those locally although I have seen rare failures
in those tests for a while.
I wonder if this change has just exposed an underlying issue?
I can add looking at this to my TODO list.
Mark
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org