Network issues with new jails/plugins

[Nick R]

I am running 9.10-STABLE.

I currently have 2 plugins (transmission and plex) running perfectly, as well as a standard jail running nzbget which is also running perfectly. I am not doing anything fancy with them, they just worked right away.

Weird thing now is if I start a 4th jail or install a new plugin (say for example crashplan), then I can't ping it and I get errors in the console such as:
Jul 12 17:23:22 nas2 warden: Jail created at /mnt/tank/jails/crashtest
Jul 12 17:23:22 nas2 devd: Executing '/etc/pccard_ether epair0a start'
Jul 12 17:23:22 nas2 devd: Executing '/etc/pccard_ether epair0b start'
Jul 12 17:23:22 nas2 rtsold[1271]: <rtsock_input_ifannounce> interface epair0b removed
Jul 12 17:23:22 nas2 devd: Executing '/etc/rc.d/dhclient quietstart epair0a'
Jul 12 17:23:22 nas2 epair0a: Ethernet address: 02:ff:20:00:05:0a
Jul 12 17:23:22 nas2 epair0b: Ethernet address: 02:ff:70:00:09:0b
Jul 12 17:23:22 nas2 kernel: epair0a: link state changed to UP
Jul 12 17:23:22 nas2 kernel: epair0a: link state changed to UP
Jul 12 17:23:22 nas2 kernel: epair0b: link state changed to UP
Jul 12 17:23:22 nas2 kernel: epair0b: link state changed to UP
Jul 12 17:23:22 nas2 kernel: epair0a: promiscuous mode enabled
Jul 12 17:23:22 nas2 kernel: ng_ether_ifnet_arrival_event: can't re-name node epair0b
Jul 12 17:23:22 nas2 kernel: ng_ether_ifnet_arrival_event: can't re-name node epair0b
Jul 12 16:23:25 nas2 manage.py: [freeadmin.navtree:577] Couldn't retrieve http://192.168.1.5/plugins/crashplan/4/_s/treemenu: timed out
Jul 12 16:23:26 nas2 manage.py: [plugins.utils:91] Couldn't retrieve http://192.168.1.5/plugins/crashplan/4/_s/status: No JSON object could be decoded

If I use FreeNAS's webgui to console into the jail, I am able to ping out and dns resolves perfectly as well. When I do ping out, I can ping the jail from outside. If I stop pinging from inside to outside, then it stops working again. Figuring this could be an ARP or duplicate IP issue, I double-checked everything and there is no conflict. I even tried changing the IP address of the jail, and it exhibits the same behaviour.

I looks like I can have a maximum of 3 jails. Anyone ever seen this issue?

Interfaces look like this and epair0 is the one not working:

xn1: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=1<RXCSUM>
ether 2a:59:52:16:1e:5b
inet 192.168.4.2 netmask 0xffffff00 broadcast 192.168.4.255
nd6 options=9<PERFORMNUD,IFDISABLED>
media: Ethernet manual
status: active
bridge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
ether 02:21:a1:62:cd:00
nd6 options=1<PERFORMNUD>
id 00:00:00:00:00:00 priority 32768 hellotime 2 fwddelay 15
maxage 20 holdcnt 6 proto rstp maxaddr 2000 timeout 1200
root id 00:00:00:00:00:00 priority 32768 ifcost 0 port 0
member: epair0a flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 5 priority 128 path cost 2000
member: epair3a flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 8 priority 128 path cost 2000
member: epair2a flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 7 priority 128 path cost 2000
member: epair1a flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 6 priority 128 path cost 2000
member: xn1 flags=143<LEARNING,DISCOVER,AUTOEDGE,AUTOPTP>
ifmaxaddr 0 port 3 priority 128 path cost 2000000
epair1a: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=8<VLAN_MTU>
ether 02:ff:20:00:06:0a
nd6 options=1<PERFORMNUD>
media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
status: active
epair2a: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=8<VLAN_MTU>
ether 02:ff:20:00:07:0a
nd6 options=1<PERFORMNUD>
media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
status: active
epair3a: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=8<VLAN_MTU>
ether 02:ff:20:00:08:0a
nd6 options=1<PERFORMNUD>
media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
status: active
epair0a: flags=8943<UP,BROADCAST,RUNNING,PROMISC,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=8<VLAN_MTU>
ether 02:ff:20:00:05:0a
nd6 options=1<PERFORMNUD>
media: Ethernet 10Gbase-T (10Gbase-T <full-duplex>)
status: active

 

[dlavigne]

That's interesting. This is on the latest STABLE update (201606270534)?

If you would like a dev to take a stab at figuring out what is going on or what the limitation is, create a bug report at bugs.freenas.org that includes a debug (made from System -> Advanced -> Save Debug), and post the issue number here.

 

[Valdhor]

This happened to me after upgrading to the latest 9.10 STABLE 201606270534. I have been following the procedures at https://gist.github.com/jacobblock/4d9270c25548c65300dd for all my jails. I did some troubleshooting and it appears that if I set anything in the sysctls section when creating the jail this particular problem surfaces. pkg update/upgrade fails with a timeout. It's kind of like networking works for a little bit but then stops.

 

[Nick R]

That's interesting. This is on the latest STABLE update (201606270534)?

If you would like a dev to take a stab at figuring out what is going on or what the limitation is, create a bug report at bugs.freenas.org that includes a debug (made from System -> Advanced -> Save Debug), and post the issue number here.

Correct, FreeNAS-9.10-STABLE-201606270534 (dd17351)

I'll do a PR tonight and post it here.

 

[Nick R]

Here is the bug report:
https://bugs.freenas.org/issues/16331

 

[MattN]

Old thread, but I ran into this and found a solution in case anyone searches for this and runs into the same thing.
When creating a new jail, it used the same mac address as an existing jail thus creating a conflict. I incremented the mac address of the new jail that wasn't working and voila, problem solved.

Check the mac address of each jail you're running, and make sure they're different. If not, that's likely the issue.

 

[Michael Sparks]

Old thread, but I ran into this and found a solution in case anyone searches for this and runs into the same thing.
When creating a new jail, it used the same mac address as an existing jail thus creating a conflict. I incremented the mac address of the new jail that wasn't working and voila, problem solved.

Check the mac address of each jail you're running, and make sure they're different. If not, that's likely the issue.

Thanks with latest freenas build to date (FreeNAS-9.10.2 (a476f16)) I ran into this issue when installing a new plugin, your fix resolved my issue!

 

[Dotty]

Geezzz !!,, same thing just happened to me, I just installed 9.10.1-U4 (the latest release right before 9.10.2).
Exactly as the OP stated.
On 9.10.2 the VirtualBox was taken out without any notice or release notes warning, and on 9.10.1-U4 then the jails made by the plugins breaks.

Now Im in the purgatory, cant move to 9.10.2 because of the VB and can't trust 9.10.1-U4 because of the plugin jails.

Come on guys, I know FreeNAS is sort of enthusiast build, and it is ok that new features are not that stable, but dont break stuff backwards.
Maybe the word "stable" should be taken out of the whole versioning system, and just leave everything as "beta".
"Stable" inspire some sort of dependability that seems a little unrealistic by my latest experience with FreeNAS.

 

[pirateghost]

On 9.10.2 the VirtualBox was taken out without any notice or release notes warning, and on 9.10.1-U4 then the jails made by the plugins breaks.

Got any proof to this? I am using 9.10.1-U4 and the plugins and jails work fine. I have not heard of Virtualbox "being removed" in 9.10.2. Pretty sure we would have had an uproar in the forums if this were the case.

 

[Dotty]

If you had an older FreeNAS version with plugins working and then upgraded to 9.10.1-U4 you will not see the issue.
You need to do a fresh install, and then add plugins.

VirtualBox template was removed from 9.10.2,, if you have a box with that version installed, go to Jails | Template, and you wont see VirtualBox there, furthermore, if you install it manually, it looks like it works, but the VMs wont boot.
There are threads about this, here and on Reddit, where people are a little more loose with their comments about FreeNAS.

 

[joebad1]

Same problem here with the latest greatest 9.10.2-U4. In my case, I deleted and re-installed my Transmission Plug-in. It was assigned the same MAC address as my SABnzbd plug-in and caused all sorts of heartache. I manually changed the MAC address on the Transmission plug-in and restarted both Transmission and SABnzbd. All seems to be working fine now. Thanks for the tip!!!