https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
Mark Thomas changed:
What|Removed |Added
Resolution|--- |FIXED
Status|NEW
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #12 from Peter Major ---
Things are looking good so far, haven't seen my websocket requests interrupted
and catalina.out didn't show the NPE either.
Thank you for the quick turnaround.
--
You are receiving this mail because:
You
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #11 from Mark Thomas ---
It is now in 9.0.x. Let us know how you get on.
--
You are receiving this mail because:
You are the assignee for the bug.
-
To
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #10 from Peter Major ---
FYI if you have a PR/branch for this, I can test the fix locally.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #9 from Mark Thomas ---
I have a fix for this locally. Just need to test it a little more.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #8 from Remy Maucherat ---
(In reply to Mark Thomas from comment #7)
> I can't fund any EG discussion to support a requirement that the upgrade
> header is set on the response before calling HttpServletRequest.upgrade().
> I'll see
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #7 from Mark Thomas ---
I can't fund any EG discussion to support a requirement that the upgrade header
is set on the response before calling HttpServletRequest.upgrade(). I'll see if
can find a different way to get the protocol
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #6 from Mark Thomas ---
Hmm. Tomcat's Javadocs for HttpServletRequest.upgrade() have language stating
the headers must be set before the method is called. The Servlet spec API does
not have that language. I wonder where that
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #5 from Peter Major ---
The Upgrade header does eventually show up in the response, but it looks like
it happens after the call to UpgradeProcessorExternal is made (added by an
async servlet). From the browser's point of view I'm
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #4 from Mark Thomas ---
To re-phrase what Rémy said, we can fix the NPE but this is still going to fail
because the response header is missing. You'll just get a nicer error message.
--
You are receiving this mail because:
You
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #3 from Remy Maucherat ---
(In reply to Peter Major from comment #2)
> Sadly I don't have access to the source code (found it when setting up a
> third party reverse proxy product). The reason why upgradeGroupInfo is null,
> is
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #2 from Peter Major ---
Sadly I don't have access to the source code (found it when setting up a third
party reverse proxy product). The reason why upgradeGroupInfo is null, is
because AbstractHttp11Protocol#getUpgradeGroupInfo is
https://bz.apache.org/bugzilla/show_bug.cgi?id=64947
--- Comment #1 from Remy Maucherat ---
Ok, this is new code that allows collecting stats.
I cannot immediately see why the UpgradeGroupInfo ends up null so more research
could be useful [example code maybe ?], but there is a null check for
13 matches
Mail list logo
