 Adam Ierymenko
|
3740b83f63
Don't back up sqlite db if it hasn't changed to prevent constant thrashing on inactive controllers.
|
9 سال پیش |
|
Adam Ierymenko
|
b2d048aa0e
Make Dictionary templatable so it can be used where we want a higher capacity.
|
9 سال پیش |
|
Adam Ierymenko
|
20d4dada40
Refactor controller for new merged format.
|
9 سال پیش |
|
Adam Ierymenko
|
734cbb2f1e
Controller modifications for default route are ready to test. Will require slight changes in ZeroTier Central when it goes live.
|
9 سال پیش |
|
Adam Ierymenko
|
acbe8ad398
More controller work, and some RedHat fixes.
|
9 سال پیش |
|
Adam Ierymenko
|
9898066b47
Remove some deprecated stuff in controller -- not done yet.
|
9 سال پیش |
|
Adam Ierymenko
|
8b9519f0af
Simplify a bunch of NetworkConfig stuff by eliminating accessors, also makes network controller easier to refactor.
|
9 سال پیش |
|
Adam Ierymenko
|
9b59bcd995
Clean controller circuit test memory.
|
9 سال پیش |
|
Adam Ierymenko
|
78c1d9006a
flood protection fix
|
9 سال پیش |
|
Adam Ierymenko
|
2e04dc03f2
Logging to NodeHistory, SQL queries.
|
9 سال پیش |
|
Adam Ierymenko
|
f8eb6b0067
Add NodeHistory table on sqlite controller.
|
9 سال پیش |
|
Adam Ierymenko
|
9cb4bbe2b8
Save test results for circuit tests in memory and then cancel the test and send the results when the test is queried later. This way you can POST a test and then come GET the result at the appointed time.
|
9 سال پیش |
|
Adam Ierymenko
|
7903f24a8f
Create periodic backup copies of controller.db in network controller from the main process itself to facilitate easier and safer backups of controller.db.
|
9 سال پیش |
|
Adam Ierymenko
|
a95fa379cc
Circuit tests basically work but need some tweaks, and fix some issues found with valgrind.
|
9 سال پیش |
|
Adam Ierymenko
|
a3876353ca
Abiltiy to post a test via the controller web API, and parsing of CIRCUIT_TEST_REPORT messages.
|
9 سال پیش |
|
Adam Ierymenko
|
7394ec6f6a
Prep in controller code to run tests.
|
9 سال پیش |
|
Adam Ierymenko
|
610ab0750c
Drop Sqlite-based Log table for now and switch to an in-memory log for recent activity. Log table gets too big on busy nodes. Should probably support push of events to some kind of event system later.
|
10 سال پیش |
|
Adam Ierymenko
|
cd005341c5
Extra statement to clean up Members -- cascade did not seem to work, possibly due to dual key.
|
10 سال پیش |
|
Adam Ierymenko
|
bebe3d7cfa
Fix deadlock in test mode.
|
10 سال پیش |
|
Adam Ierymenko
|
1f7a41cff8
Fix to allowing identity to be populated if not present.
|
10 سال پیش |
|
Adam Ierymenko
|
0d386f1c31
Add a bit of useful testing instrumentation to SqliteNetworkController.
|
10 سال پیش |
|
Adam Ierymenko
|
d57ea671d7
Add version to log.
|
10 سال پیش |
|
Adam Ierymenko
|
b3516c599b
Add a rate limiting circuit breaker to the network controller to prevent flooding attacks and race conditions.
|
10 سال پیش |
|
Adam Ierymenko
|
e2a2993b18
Add a Log table to log queries for debugging and security logging. No JSON API support for querying the log yet, but will probably come via /network/###/member/###/log/... or something.
|
10 سال پیش |
|
Adam Ierymenko
|
649a12472b
Report controllerInstanceId in all objects so that controller resets can be easily detected by whatever is using the service.
|
10 سال پیش |
|
Adam Ierymenko
|
5515909c1e
Add a concept of an "instanceId" to the controller, which the OnePoint can use to determine whether it is the same running database instance it already knows.
|
10 سال پیش |
|
Adam Ierymenko
|
99969b186b
Add a concept of a member revision counter to networks. This can be used to select all members that have been added or changed since a given point.
|
10 سال پیش |
|
Adam Ierymenko
|
30e4a188d0
ipLocalRoutes now exposed via network objects in JSON controller API, and documentation changes.
|
10 سال پیش |
|
Adam Ierymenko
|
5c9411a671
Untested -- modifications to support IP ranges instead of ip/mask for IP assignment pools, also add portId to Rule for future use.
|
10 سال پیش |
|
Adam Ierymenko
|
57c7992c78
GitHub issue #191 - kill intra-network multicast rate limits (which were not well supported or easily configurable anyway) -- this is really left over from the old collaborative multicast propagation algorithm. New algorithm (in for a while) has been sender-side replication in which sender "pays" all bandwidth, which intrinsically limits multicast.
|
10 سال پیش |
