[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20220405102350.61dce151@hermes.local>
Date: Tue, 5 Apr 2022 10:23:50 -0700
From: Stephen Hemminger <stephen@...workplumber.org>
To: netdev@...r.kernel.org
Subject: Fw: [Bug 215793] New: ip netns del is asynchronous??? -- very
unexpected and puzzling
Begin forwarded message:
Date: Sat, 02 Apr 2022 12:14:36 +0000
From: bugzilla-daemon@...nel.org
To: stephen@...workplumber.org
Subject: [Bug 215793] New: ip netns del is asynchronous??? -- very unexpected and puzzling
https://bugzilla.kernel.org/show_bug.cgi?id=215793
Bug ID: 215793
Summary: ip netns del is asynchronous??? -- very unexpected and
puzzling
Product: Networking
Version: 2.5
Kernel Version: 5.10.106
Hardware: All
OS: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: Other
Assignee: stephen@...workplumber.org
Reporter: rm+bko@...anrm.net
Regression: No
Hello,
I have a script like this:
ip netns del test-ns
ip netns add test-ns
ip link add tun-test type gre local any remote 8.8.8.8
ip link set tun-test netns test-ns
As is, the script will fail exactly every other time:
---------------
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
# ./nstest.sh
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
# ./nstest.sh
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
# ./nstest.sh
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
# ./nstest.sh
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
# ./nstest.sh
# ./nstest.sh
RTNETLINK answers: File exists
Cannot find device "tun-test"
---------------
That's because "ip netns del" seems to return without actually finishing to
fully delete the NS. Adding a "sleep 1" after the first line, makes the script
complete successfully every time. This is a very unexpected behavior. I would
ask for a better workaround than "sleep 1", but I believe there should be no
workaround, and the script should reliably work as posted.
Hope you can consider changing this.
Thanks!
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
Powered by blists - more mailing lists