2008/3/21, Dan Fandrich wrote:

> That's pretty bizarre.

I know its quite strange. But.. its happening.

For example on 2008-03-21 and 2008-03-19 autobuild "Red Hat 7.2 Linux
2.4 i686 gcc 2.96 (no valgrind)" did not exhibit the problem. But the
same system on 2008-03-20 failed test 624 with "SSH public key
authentication failed: Invalid signature for supplied public key, or
bad username/public key combination" while all the other 6XX tests
succeeded that same day.

> Note that someone has posted a couple of logs for
> the "good" and "bad" scenarios as part of a bug report:
>
>http://sourceforge.net/tracker/index.php?func=detail&aid=1832491&group_id=125852&atid=703942

Yes, it seems it could be the same, now four months old, bug report.

The posted logs are of no help for me. And in any case, I'm not fluent
in libssh2 internals to know which state variable or whatever is not
being properly cleared or set. So I'm unable to further pursue this
issue.

-- 
-=[Yang]=-