Dear Debian users,
I am encountering a problem with X2Go session resumption after upgrading one of
my remote systems from Debian Bookworm (12) to Debian Trixie (13). The setup is
as follows:
• Local site: Debian Bookworm with the X2Go client.
• Remote site: Armbian, based on Debian Trixie with only a few different
packages like Kernel, running x2goserver and nxagent on an Odroid‑C4.
New sessions can be created and used normally, and they can be suspended.
However, resuming an existing session fails with the error message:
“The remote proxy closed the connection while negotiating the session. This
may be due to the wrong authentication credentials passed to the server.”
On the remote side, x2golistsessions shows the session state changing from S to
R, and the NX‑agent log (/tmp/.x2go‑user1/C‑.../session.log) indicates that the
options file is present and correct after suspend, but gets truncated to size 0
during the resume attempt.
The remote site then reports in the logfile:
Loop: PANIC! Can't read options from file '/tmp/.x2go‑user1/C‑.../options'.
Error is 4 'Interrupted system call'.
Error: Parsing of NX transport options failed.
Loop: PANIC! Call to bind failed for x11. Error is 98 'Address already in
use'.
Despite this, on a second Odroid‑C4 where the remote system is still running
Debian Bookworm, the very same X2Go setup and client are able to resume the
session without problems.
This suggests that the resume failure is tied to the upgrade to Debian Trixie
on the remote side, and not to the local client or network configuration.
Before I file a bug report against x2goserver or nx‑libs for Trixie, I would
like to know if this pattern is already known, or whether others have seen
similar resume‑state corruption after upgrading to Debian Trixie.
If you have any similar experiences, debugging hints, or configuration
workarounds, I would be very grateful for feedback.
Best regards,
Klaus.
--
Klaus Singvogel
GnuPG-Key-ID: 1024R/5068792D 1994-06-27
