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.
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
Thanks Paul. I'm out of town for a few days but hope to try it tonight. Would it be possible to post a copy of the whole unit file so I can just paste it in? Just makes my testing a little bit easier thanks - Greg
Welcome to the Twilight Zone. I think we're suddenly chasing a different problem now. Long story short, the timing seems better, but now I can't make ipsec on my left side set up the route back to the right side to save my life. I did not touch any tunnel parameters. The misbehaving system is on the left side. I dummied up the first 2 octets of public IP Addresses below. Here is what I did to /usr/lib/systemd/system/ipsec.service when I inserted that line you mentioned. [root@superior-fw system]# diff ipsec.service ipsec.service-original 3c3 < ##After=network.target --- > After=network.target 5d4 < After=network-online.target [root@superior-fw system]# Here is the complete unit file with the change put in: root@superior-fw system]# more ipsec.service [Unit] Description=Internet Key Exchange (IKE) Protocol Daemon for IPsec ##After=network.target #After=remote-fs.target After=network-online.target [Service] Type=simple Restart=always # backwards compatible with plutorestartoncrash=no #RestartPreventExitStatus=137 143 SIGTERM SIGKILL EnvironmentFile=-/etc/sysconfig/pluto #Environment=IPSEC_SBINDIR=/usr/sbin #Environment=IPSEC_EXECDIR=/usr/sbin/ipsec #PIDFile=/var/run/pluto/pluto.pid # ExecStartPre=/usr/libexec/ipsec/addconn --config /etc/ipsec.conf --checkconfig ExecStartPre=/usr/libexec/ipsec/_stackmanager start ExecStartPre=/usr/sbin/ipsec --checknss ExecStart=/bin/sh -c 'eval `/usr/libexec/ipsec/pluto --config /etc/ipsec.conf --nofork $PLUTO_OPTIO NS`' ExecStop=/usr/libexec/ipsec/whack --shutdown ExecStopPost=/sbin/ip xfrm policy flush ExecStopPost=/sbin/ip xfrm state flush ExecReload=/usr/libexec/ipsec/whack --listen [Install] WantedBy=multi-user.target[root@superior-fw ~] Let's try a reboot on my left side and see what happens. Doing /var/log/secure -f on the right side, I see: Sep 24 04:00:05 lme-fw2 pluto[2534]: "Superior-Everywhere" #695: STATE_QUICK_R2: IPsec SA established tunnel mode {ESP=>0x4d85b7d0 <0xb3b21e51 xfrm=AES_128-HMAC_SHA1 NATOA=none NATD=none DPD=none} This looks good. But going to the left side where the rubber meets the road - this is the result of a bunch of troubleshooting: [root@superior-fw ~]# ping 192.168.3.4 PING 192.168.3.4 (192.168.3.4) 56(84) bytes of data. ^C --- 192.168.3.4 ping statistics --- 5 packets transmitted, 0 received, 100% packet loss, time 3999ms [root@superior-fw ~]# ip route show default via 1.2.151.113 dev enp2s0 10.10.10.0/24 dev enp5s7 proto kernel scope link src 10.10.10.88 169.254.0.0/16 dev enp2s0 scope link metric 1002 169.254.0.0/16 dev enp3s0 scope link metric 1003 169.254.0.0/16 dev enp5s7 scope link metric 1006 172.21.5.0/24 dev enp3s0 proto kernel scope link src 172.21.5.100 1.2.151.112/29 dev enp2s0 proto kernel scope link src 209.23.151.114 And to work around the problem: [root@superior-fw ~]# ip route add 192.168.0.0/16 via 1.2.151.113 dev enp2s0 src 172.21.5.100 mtu 1400 [root@superior-fw ~]# ping 192.168.3.4 PING 192.168.3.4 (192.168.3.4) 56(84) bytes of data. 64 bytes from 192.168.3.4: icmp_seq=1 ttl=31 time=34.2 ms 64 bytes from 192.168.3.4: icmp_seq=2 ttl=31 time=33.3 ms 64 bytes from 192.168.3.4: icmp_seq=3 ttl=31 time=33.8 ms ^C --- 192.168.3.4 ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 2002ms rtt min/avg/max/mdev = 33.372/33.808/34.239/0.384 ms [root@superior-fw ~]# I have to add the route back to the right side by hand. That's the only way I can get the right side to answer anything. Below worked one time but did not work the second time I tried it. On the right side, do this: ipsec auto --delete Superior-Everywhere ipsec auto --add Superior-Everywhere And on the left side do this: systemctl stop ipsec systemctl start ipsec This worked the first time I tried it, but not the second time. (Edit later - it probably worked the first time because I added the route by hand and stopping and restarting ipsec did not get rid of the route.) I also tried deleting and adding the tunnel on the left side. Here is a snippet from /var/log/secure on the left side when I tried that: Sep 24 04:01:44 superior-fw pluto[2253]: "Superior-Everywhere" #1: deleting state (STATE_MAIN_I4) Sep 24 04:01:44 superior-fw pluto[2253]: "Superior-Everywhere": unroute-client output: /usr/libexec/ipsec/_updown.netkey: doroute "ip route del 192.168.0.0/16 via 1.2.151.113 dev enp2s0 mtu 1400PLUTO_ADDTIME=0 " failed (Error: argument "1400PLUTO_ADDTIME=0" is wrong: "mtu" value is invalid) Hmmm.... Restarting ipsec on my left side, I see this in /var/log/secure: Sep 24 04:23:13 superior-fw pluto[2677]: "Superior-Everywhere": route-client output: /usr/libexec/ipsec/_updown.netkey: doroute "ip route replace 192.168.0.0/16 via 1.2.151.113 dev enp2s0 mtu 1400PLUTO_ADDTIME=0 src 172.21.5.100" failed (Error: argument "1400PLUTO_ADDTIME=0" is wrong: "mtu" value is invalid) Why is my MTU suddenly invalid? This works at 5 other sites and worked at this site up until right now. And it stays broken even when I put the above unit file back the way it was and reboot. Here is what this tunnel looks like on the left side: [root@superior-fw firewall-scripts]# cd /etc/ipsec.d [root@superior-fw ipsec.d]# more superior-ipsec.conf # /etc/ipsec.d/superior-ipsec.conf - IPsec configuration file # The HOME (HQ) office is always on the right. ("Make yerself RIGHT at home!", # while the other branch sites have LEFT home.) # ##version 2.0 # conforms to second version of ipsec.conf specification # Commented out because version is declared in ipsec.conf # basic configuration conn Superior-Everywhere type=tunnel # # Left security gateway, subnet behind it, next hop toward right. # also=superior # # Right security gateway, subnet behind it, next hop toward left. # also=hq mtu=1400 auto=start include /etc/ipsec.d/sites.conf And the relevant portion of sites.conf with details about every site. I have a copy of this at every site. [root@superior-fw ipsec.d]# more sites.conf # /etc/ipsec.d/sites.conf - IPsec configuration file describing each site. # The HOME office is always on the right. ("Make yerself RIGHT at home!", # while the other branch sites have LEFT home.) ##version 2.0 # conforms to second version of ipsec.conf specification # basic configuration conn hq right=3.4.248.50 rightnexthop=3.4.248.49 rightsubnet=192.168.0.0/16 rightsourceip=192.168.3.5 rightid= # rsakey AQPL4uclF Generated 2014-0918 rightrsasigkey=0vjcGc2O+7q9... (Other site descriptions removed.) conn superior left=1.2.151.114 leftnexthop=1.2.151.113 leftsubnet=172.21.5.0/24 leftsourceip=172.21.5.100 leftid= # rsakey AQPpA2gXk leftrsasigkey=0sAQ... [root@superior-fw ipsec.d]# [root@superior-fw ipsec.d]# ipsec version Linux Libreswan 3.10 (netkey) on 3.16.2-201.fc20.x86_64 [root@superior-fw ipsec.d]# more /etc/issue Fedora release 20 (Heisenbug) Kernel \r on an \m (\l) [root@superior-fw ipsec.d]# And here is the right side: # /etc/ipsec.d/rochester-ipsec.conf - IPsec configuration file. # The HOME (HQ) office is always on the right. ("Make yerself RIGHT at home!", # while the other branch sites have LEFT home.) # (Connections to other sites removed.) conn Superior-Everywhere type=tunnel # # Left security gateway, subnet behind it, next hop toward right. # also=superior # # Right security gateway, subnet behind it, next hop toward left. # also=hq mtu=1400 auto=start include /etc/ipsec.d/sites.conf (Sites.conf is the same here as in all the left sides.) [root@lme-fw2 ipsec.d]# ipsec version Linux Libreswan 3.8 (netkey) on 3.16.2-200.fc20.x86_64 [root@lme-fw2 ipsec.d]# more /etc/issue Fedora release 20 (Heisenbug) Kernel \r on an \m (\l) The left side in Superior is running version 3.10 (see above). Looks like its MTU problems started on Sept. 21. So the MTU problem is not new tonight, I just found it tonight. [root@superior-fw ipsec.d]# grep mtu /var/log/secure Sep 21 04:59:22 superior-fw pluto[1673]: "Superior-Everywhere": route-client output: /usr/libexec/ipsec/_updown.netkey: doroute "ip route replace 192.168.0.0/16 via 209.23.151.113 dev enp2s0 mtu 1400PLUTO_ADDTIME=0 src 172.21.5.100" failed (Error: argument "1400PLUTO_ADDTIME=0" is wrong: "mtu" value is invalid) . . . (lots of others) The right side is running version 3.8. No MTU problems on the right side. [root@lme-fw2 ipsec.d]# grep mtu /var/log/secure [root@lme-fw2 ipsec.d]#
confirmed bug. Fix commited upstream. I'll check if upstream will release 3.11 today, and either upgrade to that, or call a new build with 3.10 and a patch.
Thanks Paul. Is the confirmed bug the startup timing thing or the MTU thing or both? No sense doing a special patch - I have a workaround for now and I can test 3.11 when it comes out the other end. thanks - Greg
(In reply to Paul Wouters from comment #1) > Can you try this change to the systemd unit file and see if it fixes your > issue: I can confirm that the equivalent in e.g. /etc/systemd/system/ipsec.service.d/network-online.target.conf was enough to fix the interface issue: [Unit] After=network-online.target Is libreswan expected to pay attention to a netlink socket and/or periodically scan for new/changed interfaces? The above is sufficient to bring up a tunnel shortly after boot, but feels like an incomplete hack... -Rob
libreswan-3.11-1.fc21 has been submitted as an update for Fedora 21. https://admin.fedoraproject.org/updates/libreswan-3.11-1.fc21
libreswan-3.11-1.fc20 has been submitted as an update for Fedora 20. https://admin.fedoraproject.org/updates/libreswan-3.11-1.fc20
libreswan-3.11-1.fc19 has been submitted as an update for Fedora 19. https://admin.fedoraproject.org/updates/libreswan-3.11-1.fc19
Package libreswan-3.11-1.fc21: * should fix your issue, * was pushed to the Fedora 21 testing repository, * should be available at your local mirror within two days. Update it with: # su -c 'yum update --enablerepo=updates-testing libreswan-3.11-1.fc21' as soon as you are able to. Please go to the following url: https://admin.fedoraproject.org/updates/FEDORA-2014-13597/libreswan-3.11-1.fc21 then log in and leave karma (feedback).
I just updated my problem site to 3.11-1.fc20 and restarted. All seems to work well now. It starts when it's supposed to and the MTU problem doesn't happen anymore. Leaving good karma in https://admin.fedoraproject.org/updates/FEDORA-2014-13597/libreswan-3.11-1.fc21 - Greg
thanks for the feedback! Feel free to close this bug :)
libreswan-3.11-1.fc20 has been pushed to the Fedora 20 stable repository. If problems still persist, please make note of it in this bug report.
libreswan-3.11-1.fc19 has been pushed to the Fedora 19 stable repository. If problems still persist, please make note of it in this bug report.
libreswan-3.11-1.fc21 has been pushed to the Fedora 21 stable repository. If problems still persist, please make note of it in this bug report.