Commit Graph

276 Commits

Author SHA1 Message Date
Nathan Brown df7c7eec4a
Get out faster on nil udpAddr (#449) 2021-04-26 20:21:47 -05:00
Nathan Brown 6f37280e8e
Fully close tunnels when CloseAllTunnels is called (#448) 2021-04-26 10:42:24 -05:00
Nathan Brown a0735dd7d5
Add locking around ssh conns to avoid concurrent map access on reload (#447) 2021-04-23 14:43:16 -05:00
Nathan Brown 1deb5d98e8
Fix tun funcs for ios and android (#446) 2021-04-22 15:23:40 -05:00
Nathan Brown a1ee521d79
Fix a failed return in an error case (#445) 2021-04-17 18:47:31 -05:00
brad-defined 7859140711
Only set serveDns if the host is also configured to be a lighthouse. (#433) 2021-04-16 13:33:56 -05:00
brad-defined 17106f83a0
Ensure the Nebula device exists before attempting to bind to the Nebula IP (#375) 2021-04-16 10:34:28 -05:00
Nathan Brown ab08be1e3e
Don't panic on a nil response from the lighthouse (#442) 2021-04-15 09:12:21 -05:00
Nathan Brown 710df6a876
Refactor remotes and handshaking to give every address a fair shot (#437) 2021-04-14 13:50:09 -05:00
John Maguire 20bef975cd
Remove obsolete systemd unit settings (take 2) (#438) 2021-04-07 12:02:40 -05:00
Nathan Brown 480036fbc8
Remove unused structs in hostmap.go (#430) 2021-04-01 22:07:11 -05:00
Nathan Brown 1499be3e40
Fix name resolution for host names in config (#431) 2021-04-01 21:48:41 -05:00
Nathan Brown 64d8e5aa96
More LH cleanup (#429) 2021-04-01 10:23:31 -05:00
Nathan Brown 75f7bda0a4
Lighthouse performance pass (#418) 2021-03-31 17:32:02 -05:00
Nathan Brown e7e55618ff
Include bad backets in the good handshake test (#428) 2021-03-31 13:36:10 -05:00
Nathan Brown 0c2e5973e1
Simple lie test (#427) 2021-03-31 10:26:35 -05:00
Nathan Brown 830d6d4639
Start of end to end testing with a good handshake between two nodes (#425) 2021-03-29 14:29:20 -05:00
Nathan Brown 883e09a392
Don't use a global ca pool (#426) 2021-03-29 12:10:19 -05:00
Wade Simmons 4603b5b2dd
fix PromoteEvery check (#424)
This check was accidentally typo'd in #396 from `%` to `&`. Restore the
correct functionality here (we want to do the check every "PromoteEvery"
count packets).
2021-03-26 15:01:05 -04:00
Wade Simmons a71541fb0b
export build version as a prometheus label (#405)
This is how Prometheus recommends you do it, and how they do it
themselves in their client. This makes it easy to see which versions you
have deployed in your fleet, and query over it too.
2021-03-26 14:16:35 -04:00
Nathan Brown 3ea7e1b75f
Don't use a global logger (#423) 2021-03-26 09:46:30 -05:00
Nathan Brown 7a9f9dbded
Don't craft buffers if we don't need them (#416) 2021-03-22 18:25:06 -05:00
Nathan Brown 7073d204a8
IPv6 support for outside (udp) (#369) 2021-03-18 20:37:24 -05:00
Joe Doss 9e94442ce7
Add fedora dist files. (#413) 2021-03-18 12:33:43 -07:00
Joe Doss 13471f5792
Remove obsolete systemd unit settings. (#412) 2021-03-18 12:29:36 -07:00
Thomas Roten ea07a89cc8
Ensure mutex is unlocked when adding remote IP. (#406)
Currently, if you use the remote allow list config, as soon as you attempt to create a tunnel to a node that has a blocked IP address, a mutex is locked and never unlocked. This happens even if the node has an allowed remote IP address in addition to the blocked remote IP address.

This pull request ensures that the lighthouse mutex is unlocked whenever we attempt to add a remote IP.
2021-03-16 12:41:35 -04:00
Ryan Huber 3aaaea6309
don't allow a useless handshake with yourself (#402)
* don't allow a useless handshake with yourself

* remove helper
2021-03-15 12:58:23 -07:00
Wade Simmons 5506da3de9
Fix selection of UDP remote to use during stage2 (#404)
The change for #401 incorrectly called HostInfo.ForcePromoteBest in
stage2, when we really we want to pick the remote that we received the
response from.
2021-03-12 21:43:24 -05:00
Wade Simmons 6c55d67f18
Refactor handshake_ix (#401)
There are some subtle race conditions with the previous handshake_ix implementation, mostly around collisions with localIndexId. This change refactors it so that we have a "commit" phase during the handshake where we grab the lock for the hostmap and ensure that we have a unique local index before storing it. We also now avoid using the pending hostmap at all for receiving stage1 packets, since we have everything we need to just store the completed handshake.

Co-authored-by: Nate Brown <nbrown.us@gmail.com>
Co-authored-by: Ryan Huber <rhuber@gmail.com>
Co-authored-by: forfuncsake <drussell@slack-corp.com>
2021-03-12 14:16:25 -05:00
Wade Simmons 64d8035d09
fix race in getOrHandshake (#400)
We missed this race with #396 (and I think this is also the crash in
issue #226). We need to lock a little higher in the getOrHandshake
method, before we reset hostinfo.ConnectionInfo. Previously, two
routines could enter this section and confuse the handshake process.

This could result in the other side sending a recv_error that also has
a race with setting hostinfo.ConnectionInfo back to nil. So we make sure
to grab the lock in handleRecvError as well.

Neither of these code paths are in the hot path (handling packets
between two hosts over an active tunnel) so there should be no
performance concerns.
2021-03-09 09:27:02 -05:00
Ryan Huber 73a5ed90b2
Do not allow someone to run a nebula lighthouse with an ephemeral port (#399)
* Do not allow someone to run a nebula lighthouse with an ephemeral port

* derp - we discover the port so we have to check the config setting

* No context needed for this error

* gofmt yourself

* Revert "gofmt yourself"

This reverts commit c01423498e3792f7acd69d7e691dce1edad81bcb.

* Revert "No context needed for this error"

This reverts commit 6792af6846d1200c564a4ad601a637535dd56c5b.

* snip snap snip snap
2021-03-08 12:42:06 -08:00
Wade Simmons d604270966
Fix most known data races (#396)
This change fixes all of the known data races that `make smoke-docker-race` finds, except for one.

Most of these races are around the handshake phase for a hostinfo, so we add a RWLock to the hostinfo and Lock during each of the handshake stages.

Some of the other races are around consistently using `atomic` around the `messageCounter` field. To make this harder to mess up, I have renamed the field to `atomicMessageCounter` (I also removed the unnecessary extra pointer deference as we can just point directly to the struct field).

The last remaining data race is around reading `ConnectionInfo.ready`, which is a boolean that is only written to once when the handshake has finished. Due to it being in the hot path for packets and the rare case that this could actually be an issue, holding off on fixing that one for now.

here is the results of `make smoke-docker-race`:

before:

    lighthouse1: Found 2 data race(s)
    host2:       Found 36 data race(s)
    host3:       Found 17 data race(s)
    host4:       Found 31 data race(s)

after:

    host2: Found 1 data race(s)
    host4: Found 1 data race(s)

Fixes: #147
Fixes: #226
Fixes: #283
Fixes: #316
2021-03-05 21:18:33 -05:00
Nathan Brown 29c5f31f90
Add a check in the makefile to ensure a minimum version of go is installed (#383) 2021-03-02 13:29:05 -06:00
Nathan Brown b6234abfb3
Add a way to trigger punch backs via lighthouse (#394) 2021-03-01 19:06:01 -06:00
Wade Simmons 2a4beb41b9
Routine-local conntrack cache (#391)
Previously, every packet we see gets a lock on the conntrack table and updates it. When running with multiple routines, this can cause heavy lock contention and limit our ability for the threads to run independently. This change caches reads from the conntrack table for a very short period of time to reduce this lock contention. This cache will currently default to disabled unless you are running with multiple routines, in which case the default cache delay will be 1 second. This means that entries in the conntrack table may be up to 1 second out of date and remain in a routine local cache for up to 1 second longer than the global table.

Instead of calling time.Now() for every packet, this cache system relies on a tick thread that updates the current cache "version" each tick. Every packet we check if the cache version is out of date, and reset the cache if so.
2021-03-01 19:52:17 -05:00
Wade Simmons d232ccbfab
add metrics for the udp sockets using SO_MEMINFO (#390)
Retrieve the current socket stats using SO_MEMINFO and report them as
metrics gauges. If SO_MEMINFO isn't supported, we don't report these metrics.
2021-03-01 19:51:33 -05:00
Nathan Brown ecfb40f29c
Fix osx for mq changes, this does not implement mq on osx (#395) 2021-03-01 16:57:05 -05:00
Wade Simmons 1bae5b2550
more validation in pending hostmap deletes (#344)
We are currently seeing some cases where we are not deleting entries
correctly from the pending hostmap. I believe this is a case of
an inbound timer tick firing and deleting the Hosts map entry for
a newer handshake attempt than intended, thus leaving the old Indexes
entry orphaned. This change adds some extra checking when deleteing from
the Indexes and Hosts maps to ensure we clean everything up correctly.
2021-03-01 12:40:46 -05:00
Wade Simmons 73081d99bc
add `make smoke-docker` (#287)
This makes it easier to use the docker container smoke test that
GitHub actions runs. There is also `make smoke-docker-race` that runs the
smoke test with `-race` enabled.
2021-03-01 11:15:15 -05:00
Tim Rots e7e6a23cde
fix a few typos (#302) 2021-03-01 11:14:34 -05:00
Wade Simmons a0583ebdca
tun_disabled: reply to ICMP Echo Request (#342)
This change allows a server running with `tun.disabled: true` (usually
a lighthouse) to still reply to ICMP EchoRequest packets. This allows
you to "ping" the lighthouse Nebula IP as a quick check to make sure the
tunnel is up, even when running with tun.disabled.

This is still gated by allowing `icmp` packets in the inbound firewall
rules.
2021-03-01 11:09:41 -05:00
Wade Simmons 27d9a67dda
Proper multiqueue support for tun devices (#382)
This change is for Linux only.

Previously, when running with multiple tun.routines, we would only have one file descriptor. This change instead sets IFF_MULTI_QUEUE and opens a file descriptor for each routine. This allows us to process with multiple threads while preventing out of order packet reception issues.

To attempt to distribute the flows across the queues, we try to write to the tun/UDP queue that corresponds with the one we read from. So if we read a packet from tun queue "2", we will write the outgoing encrypted packet to UDP queue "2". Because of the nature of how multi queue works with flows, a given host tunnel will be sticky to a given routine (so if you try to performance benchmark by only using one tunnel between two hosts, you are only going to be using a max of one thread for each direction).

Because this system works much better when we can correlate flows between the tun and udp routines, we are deprecating the undocumented "tun.routines" and "listen.routines" parameters and introducing a new "routines" parameter that sets the value for both. If you use the old undocumented parameters, the max of the values will be used and a warning logged.

Co-authored-by: Nate Brown <nbrown.us@gmail.com>
2021-02-25 15:01:14 -05:00
John Maguire 2bce222550
List possible cipher options in example config (#385) 2021-02-19 21:46:42 -06:00
Wade Simmons 3dd1108099
Go 1.16 and darwin-arm64 (#381)
This commit switches to Go 1.16 and adds a release binary for darwin-arm64.

Fixes: #343
2021-02-17 13:11:57 -05:00
Nathan Brown d4b81f9b8d
Add QR code support to `nebula-cert` (#297) 2021-02-11 18:53:25 -06:00
brad-defined 454bc8a6bb
Check certificate banner during nebula-cert print (#373) 2021-02-05 14:52:32 -06:00
Wade Simmons ce9ad37431
fix regression with LightHouseHandler and punchBack (#346)
The change introduced by #320 incorrectly re-uses the output buffer for
sending punchBack packets. Since we are currently spawning a new
goroutine for each send here, we need to allocate a new buffer each
time. We can come back and optimize this in the future, but for now we
should fix the regression.
2020-11-25 17:49:26 -05:00
Wade Simmons ee7c27093c
add HostMap.RemoteIndexes (#329)
This change adds an index based on HostInfo.remoteIndexId. This allows
us to use HostMap.QueryReverseIndex without having to loop over all
entries in the map (this can be a bottleneck under high traffic
lighthouses).

Without this patch, a high traffic lighthouse server receiving recv_error
packets and lots of handshakes, cpu pprof trace can look like this:

      flat  flat%   sum%        cum   cum%
    2000ms 32.26% 32.26%     3040ms 49.03%  github.com/slackhq/nebula.(*HostMap).QueryReverseIndex
     870ms 14.03% 46.29%     1060ms 17.10%  runtime.mapiternext

Which shows 50% of total cpu time is being spent in QueryReverseIndex.
2020-11-23 14:51:16 -05:00
Wade Simmons 2e7ca027a4
Lighthouse handler optimizations (#320)
We noticed that the number of memory allocations LightHouse.HandleRequest creates for each call can seriously impact performance for high traffic lighthouses. This PR introduces a benchmark in the first commit and then optimizes memory usage by creating a LightHouseHandler struct. This struct allows us to re-use memory between each lighthouse request (one instance per UDP listener go-routine).
2020-11-23 14:50:01 -05:00
mhp 672ce1f0a8
Move slice allocations in connection manager monitor loop (#340)
* Move slice allocations in connection manager monitor loop

* move further out

Co-authored-by: Miran Park <mpark@slack-corp.com>
2020-11-19 15:44:05 -08:00