https://bugzilla.mindrot.org/show_bug.cgi?id=3639
Darren Tucker <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[email protected] --- Comment #1 from Darren Tucker <[email protected]> --- (In reply to JM from comment #0) [...] > 2023-12-02T12:28:41.053381-08:00 host1 audit[3791]: SECCOMP > auid=4294967295 uid=107 gid=65534 ses=4294967295 pid=3791 > comm="sshd" exe="/opt/openssh-9.2p1/sbin/sshd" sig=31 arch=40000028 > syscall=20 compat=1 ip=0xf787080c code=0x0 This looks like a seccomp sandbox violation. The first thing I'd suggest is to try 9.5p1, because there were a couple of bugs fixed in that specifically (bug#3512 and bug#3537). There was also a thing about RNG seeding, but that depended on an interaction between different kernel and libc versions. Failing that, I'd suggest building with -DSANDBOX_SECCOMP_FILTER_DEBUG to get additional debugging on what's failing (but note that this configuration is for debugging only and is not safe for production use) and see what's in sshd's log. FWIW I have an rpi4 with a very similar configuration to what you describe ("Linux 6.1.21-v8+ #1642 SMP PREEMPT Mon Apr 3 17:24:16 BST 2023 aarch64 GNU/Linux") but was not able to reproduce the failure. -- 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. _______________________________________________ openssh-bugs mailing list [email protected] https://lists.mindrot.org/mailman/listinfo/openssh-bugs
