Date: Mon, 24 Oct 2022 21:02:22 +0000 From: bugzilla-noreply@freebsd.org To: bugs@FreeBSD.org Subject: [Bug 267315] dangling jail references Message-ID: <bug-267315-227@https.bugs.freebsd.org/bugzilla/>
next in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D267315 Bug ID: 267315 Summary: dangling jail references Product: Base System Version: CURRENT Hardware: Any OS: Any Status: New Severity: Affects Only Me Priority: --- Component: kern Assignee: bugs@FreeBSD.org Reporter: markj@FreeBSD.org Some operations in jails cause the jail to get stuck in the DYING state, holding onto memory resources. This makes it difficult to write regression tests using jails, particularly with nested jails. With nested jails, I *think* the fact that the child gets stuck prevents the parent from auto-destroying itself when the last process exits. But this means that te= st cleanup doesn't work even if the test passes. I can reproduce one case easily: markj@biggie> sudo jail -c name=3Dtest vnet command=3Dsh # ifconfig lo0 up # ^D At this point the jail is supposed to go away since it isn't persistent and doesn't contain any processes. But jls shows that it's still there: 20 / test DYING 30 If I don't UP lo0, then it works as expected. Here's a more involved example: markj@biggie> sudo jail -c name=3Dtest children.max=3D1 vnet command=3Dsh # ifconfig lo0 up # ifconfig lo0 inet 127.0.0.1 # jail -c name=3Dtest ip4.addr=3D127.0.0.1 command=3Dsh # exit # jls -vd JID Hostname Path Name State CPUSetID IP Address(es) So far so good, but then if I try to create a connection: # jail -c name=3Dtest ip4.addr=3D127.0.0.1 command=3Dsh # nc 127.0.0.1 8080 # echo $? 1 # exit # jls -vd JID Hostname Path Name State CPUSetID IP Address(es) 32 / test DYING 39 127.0.0.1 Then upon exiting the parent jail, it remains in the ACTIVE state, so I hav= e to manually destroy it. --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-267315-227>