Error in InitializeSecurityContext: 'The buffers supplied to a function was too small

[uljana12]

This error was observed when using waffle for Single Sign On using
Kerberos.

Free buffers additionally on each iteration of the InitializeSecurityContext.

[dblock]

Thanks. I'll check it out when I have a PC.

There's a merge conflict. I can resolve it for you when I do the merge. If you want to do it, merge from upstream:

git add upstream git@github.com:dblock/waffle.git
git pull upstream master
# resolve conflicts
git add .
git commit -m "Merged from upstream."
git push origin master

You could also update CHANGELOG describing the problem that you fixed.

[uljana12]

I would prefer if you do it. Thanks.

On Tue, Apr 9, 2013 at 5:11 PM, Daniel Doubrovkine (dB.) <
notifications@github.com> wrote:

Thanks. I'll check it out when I have a PC.

There's a merge conflict. I can resolve it for you when I do the merge. If
you want to do it, merge from upstream:

git add upstream git@github.com:dblock/waffle.git
git pull upstream master

resolve conflicts

git add .
git commit -m "Merged from upstream."
git push origin master

You could also update CHANGELOG describing the problem that you fixed.

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/58#issuecomment-16118566
.

[dblock]

I tried to integrate this fix.

The buffer is being passed from the outside into the initialize function. Freeing something passed from the outside seems just wrong. Who is not disposing of this buffer properly? The fix should probably be there?

[uljana12]

I am not sure. But you might be right about it, I was never sure about the
fix it is something that worked for us.

On Sat, Apr 13, 2013 at 3:51 PM, Daniel Doubrovkine (dB.) <
notifications@github.com> wrote:

I tried to integrate this fix.

The buffer is being passed from the outside into the initialize function.
Freeing something passed from the outside seems just wrong. Who is not
disposing of this buffer properly? The fix should probably be there?

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/58#issuecomment-16333440
.

[dblock]

Do you have a stack trace from a failed request by any chance?

[uljana12]

No that's the point we don't have anything, we don't have the stack trace
because we couldn't reproduce the error on our test system, and we didn't
have the possibility to debug.
The error was on our customer's system, and we had no access to it, but we
had to fix the error urgently.
The solution was my first guess which somehow worked.

On Mon, Apr 15, 2013 at 1:47 PM, Daniel Doubrovkine (dB.) <
notifications@github.com> wrote:

Do you have a stack trace from a failed request by any chance?

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/58#issuecomment-16380161
.

[dblock]

I think this is caused by java-native-access/jna#261