[Bug 3666] sshd crash

[bugzilla-daemon at mindrot.org]

https://bugzilla.mindrot.org/show_bug.cgi?id=3666

--- Comment #2 from Pawan Badganchi <pawan.badganchi at kpit.com> ---
(In reply to Darren Tucker from comment #1)
> (In reply to Pawan Badganchi from comment #0)
> > This is regarding sshd crash.
> 
> What makes you think this is a crash?
> 
> > sshd crash is being observed during openssh ptest running everytime
> > and it's reproducible.
> > Currently we are maintaining dunfell version openssh_8.2p1.
> 
> This seems to refer to a version of Yocto, and ptest is some kind of
> package test?
> 
> > Below is the error message we get. How can i resolve this issue?
> > Could you please help here?
> > 
> > sshd[27514]: Connection from UNKNOWN port 65535 on UNKNOWN port 65535
> > sshd[27514]: debug1: Local version string SSH-2.0-OpenSSH_8.2
> > sshd[27514]: error: kex_exchange_identification: Connection closed
> > by remote host
> 
> This means that something (presumably caused by your ptest) to
> invoke sshd in a way that did not involve a TCP connection (eg
> running sshd -i) then immediately disconnecting.
> 
> >  debug1: inetd sockets after dupping: 3, 4
> >  bmt sshd[27580]: Connection from UNKNOWN port 65535 on UNKNOWN port
> > 65535
> >  debug1: Local version string SSH-2.0-OpenSSH_8.2
> >  sshd[27580]: error: kex_exchange_identification: client sent
> > invalid protocol identifier ""
> 
> This is more or less the same, with the difference being that
> whatever it was sent an empty initial protocol identifier (probably
> a newline).
> 
> If your tests are invoking sshd to test it, they should probably use
> -e to write its logs to stderr or -E to write it to a file rather
> than letting it write to syslog.


> What makes you think this is a crash?

>> We are checking post test crash for openssh, so there this issue is observed.

> This seems to refer to a version of Yocto, and ptest is some kind of
> package test?

>> Yes this is version of YOCTO and kind of package test.

> This means that something (presumably caused by your ptest) to
> invoke sshd in a way that did not involve a TCP connection (eg
> running sshd -i) then immediately disconnecting.

>> May be i need to check the logs, thanks for your input. Will get back if need help.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.
You are watching someone on the CC list of the bug.