RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 1145245 - Libreswan appears to start with systemd before all the NICs are up and running.
Summary: Libreswan appears to start with systemd before all the NICs are up and running.
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: libreswan
Version: 7.1
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: rc
: ---
Assignee: Paul Wouters
QA Contact: Jaroslav Aster
URL:
Whiteboard:
Depends On: 1144831
Blocks:
TreeView+ depends on / blocked
 
Reported: 2014-09-22 15:45 UTC by Paul Wouters
Modified: 2015-03-05 10:22 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of: 1144831
Environment:
Last Closed: 2015-03-05 10:22:51 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2015:0431 0 normal SHIPPED_LIVE libreswan bug fix and enhancement update 2015-03-05 14:52:35 UTC

Description Paul Wouters 2014-09-22 15:45:21 UTC 
+++ This bug was initially created as a clone of Bug #1144831 +++

Description of problem:

When I use systemd to start libreswan 3.10 and earlier versions that come with Fedora 20, the tunnel does not come up.  But after I login, doing systemctl restart ipsec by hand fires everything up as expected.  So I know my tunnel config is good.  

I have an rc.local file that calls a script named rc.firewall, where I setup a bunch of iptables rules.  The last few lines currently look like this:

echo "Starting IPSEC from here since it looks like timing issues mess it up at boot time."
sleep 10
ip addr show > /firewall-scripts/before.txt
systemctl restart ipsec
ip addr show > /firewall-scripts/after.txt

exit

As you can see, sleeping 10 seconds before starting ipsec does the trick.  This suggests a timing issue with systemd at boot time.


Version-Release number of selected component (if applicable):
Libreswan 3.10 and 3.8

How reproducible:
Always

Steps to Reproduce:
1.  Build an F20 system with multiple NICs
2.  Set up an ipsec tunnel to somewhere
3.  systemctl enable ipsec
4.  Boot - ipsec starts before the NICs are ready and the tunnel does not come up.

Actual results:
IPSEC starts up before the NICs are ready and the tunnel does not come up.  Work around the problem by starting the tunnel in a custom script after sleeping an appropriate length of time.

Expected results:
IPSEC should start properly when called from systemd.

Additional info:

The relevant NICs are named enp2s0 and enp3s0.  NIC enp2s0 is public, enp3s0 is private.  

Waiting 10 seconds during startup seems to work.  While sleeping for only 5 seconds, I noticed during troubleshooting that NIC enp2s0 somehow took on the IP Address that NIC enp3s0 should have, but with a /32 mask.  The only possible way this could happen is some kind of timing interaction with ipsec startup.  So I captured data before and after starting ipsec, as you can see from the script extract above.  

First, when sleeping only 5 seconds - note NIC enp3s0 is not yet up. I dummied up the public IP Address on NIC enp2s0.

[root@superior-fw firewall-scripts]# more before2.txt
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:26 brd ff:ff:ff:ff:ff:ff
    inet 1.2.3.4/29 brd 1.2.3.7 scope global enp2s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e26/64 scope link
       valid_lft forever preferred_lft forever
3: enp3s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:c0:6e:27 brd ff:ff:ff:ff:ff:ff
4: enp5s4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e7 brd ff:ff:ff:ff:ff:ff
5: .
   .
   . (Other non-relevant NICs cut.)

And here is what things look like after starting ipsec.  Note enp3s0 is a little further along in getting started with a temporary state of DOWN.  

[root@superior-fw firewall-scripts]# more after2.txt
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:26 brd ff:ff:ff:ff:ff:ff
    inet 209.23.151.114/29 brd 209.23.151.119 scope global enp2s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e26/64 scope link
       valid_lft forever preferred_lft forever
3: enp3s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state DOWN group default q
len 1000
    link/ether 00:30:18:c0:6e:27 brd ff:ff:ff:ff:ff:ff
4: enp5s4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e7 brd ff:ff:ff:ff:ff:ff
5: .
   .
   .
After logging in and doing ip addr show by hand, I noticed NIC enp2s0 had an additional IP Address of 172.21.5.100/32.  I don't have a capture of that. 

Now here is what things look like before and after starting ipsec when sleeping 10 seconds.  

[root@superior-fw firewall-scripts]# more before.txt
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:26 brd ff:ff:ff:ff:ff:ff
    inet 1.2.3.4/29 brd 1.2.3.7 scope global enp2s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e26/64 scope link
       valid_lft forever preferred_lft forever
3: enp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:27 brd ff:ff:ff:ff:ff:ff
    inet6 fe80::230:18ff:fec0:6e27/64 scope link tentative
       valid_lft forever preferred_lft forever
4: enp5s4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e7 brd ff:ff:ff:ff:ff:ff
5: .
   .
   .

And after starting ipsec.  Note that NIC enp3s0 now has its proper IP Address.

[root@superior-fw firewall-scripts]# more after.txt
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:26 brd ff:ff:ff:ff:ff:ff
    inet 1.2.3.4/29 brd 1.2.3.7 scope global enp2s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e26/64 scope link
       valid_lft forever preferred_lft forever
3: enp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen
1000
    link/ether 00:30:18:c0:6e:27 brd ff:ff:ff:ff:ff:ff
    inet 172.21.5.100/24 brd 172.21.5.255 scope global enp3s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e27/64 scope link
       valid_lft forever preferred_lft forever
4: enp5s4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e7 brd ff:ff:ff:ff:ff:ff
5: .
   .
   .


And here is the current picture of all NICs when I login and look at them by hand.  This time showing everything, even the irrelevant ones so you can have a complete picture for what this system looks like.

[root@superior-fw firewall-scripts]# ip addr show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:30:18:c0:6e:26 brd ff:ff:ff:ff:ff:ff
    inet 1.2.3.4/29 brd 1.2.3.7 scope global enp2s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e26/64 scope link
       valid_lft forever preferred_lft forever
3: enp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:30:18:c0:6e:27 brd ff:ff:ff:ff:ff:ff
    inet 172.21.5.100/24 brd 172.21.5.255 scope global enp3s0
       valid_lft forever preferred_lft forever
    inet6 fe80::230:18ff:fec0:6e27/64 scope link
       valid_lft forever preferred_lft forever
4: enp5s4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e7 brd ff:ff:ff:ff:ff:ff
5: enp5s6: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e8 brd ff:ff:ff:ff:ff:ff
6: enp5s7: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state DOWN group default qlen 1000
    link/ether 00:30:18:ad:e7:e9 brd ff:ff:ff:ff:ff:ff
    inet 10.10.10.88/24 brd 10.10.10.255 scope global enp5s7
       valid_lft forever preferred_lft forever
7: wlp1s0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
    link/ether b4:b6:76:b6:39:1f brd ff:ff:ff:ff:ff:ff
8: ip_vti0@NONE: <NOARP> mtu 1332 qdisc noop state DOWN group default
    link/ipip 0.0.0.0 brd 0.0.0.0
[root@superior-fw firewall-scripts]#


Here are the relevant ifcfg-nnn files.  I dummied up the public IP Addresses.  

[root@superior-fw network-scripts]# more ifcfg-enp2s0
TYPE=Ethernet
BOOTPROTO=static
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_PEERDNS=yes
IPV6_PEERROUTES=yes
IPV6_FAILURE_FATAL=no
NAME=enp2s0
UUID=0f07ae8d-4974-4f72-af77-548cea8a0cf5
ONBOOT=yes
HWADDR=00:30:18:C0:6E:26
PEERDNS=yes
PEERROUTES=yes
BROADCAST=1.2.3.7
IPADDR=1.2.3.4
NETMASK=255.255.255.248
NETWORK=1.2.3.0
[root@superior-fw network-scripts]#
[root@superior-fw network-scripts]#
[root@superior-fw network-scripts]# more ifcfg-enp3s0
TYPE=Ethernet
BOOTPROTO=static
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_PEERDNS=yes
IPV6_PEERROUTES=yes
IPV6_FAILURE_FATAL=no
NAME=enp3s0
UUID=a81b3bae-9e9d-4c7b-9e16-fc83f7fa2b56
ONBOOT=yes
HWADDR=00:30:18:C0:6E:27
PEERDNS=yes
PEERROUTES=yes
BROADCAST=172.21.5.255
IPADDR=172.21.5.100
NETMASK=255.255.255.0
NETWORK=172.21.5.0
[root@superior-fw network-scripts]#


The only difference when waiting only 5 seconds is, NIC enp2s0 also ends up with an additional IP Address of 172.21.5.100/32.  

All this has to be a consequence of ipsec starting too soon with systemd.

--- Additional comment from Paul Wouters on 2014-09-22 11:40:52 EDT ---

Can you try this change to the systemd unit file and see if it fixes your issue:

diff --git a/initsystems/systemd/ipsec.service.in b/initsystems/systemd/ipsec.service.in
index 082ef79..9edcbd4 100644
--- a/initsystems/systemd/ipsec.service.in
+++ b/initsystems/systemd/ipsec.service.in
@@ -1,7 +1,6 @@
 [Unit]
 Description=Internet Key Exchange (IKE) Protocol Daemon for IPsec
-After=network.target
-#After=remote-fs.target
+After=network-online.target
 
 [Service]
 Type=simple
Comment 7 errata-xmlrpc 2015-03-05 10:22:51 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHBA-2015-0431.html
Note You need to log in before you can comment on or make changes to this bug.