Philip O'Toole
95a689b3cf
Some improvements thanks to ChatGPT 4
2 years ago
Philip O'Toole
27839b53f7
ABL
2 years ago
Philip O'Toole
60ef0a86b8
Final tweaks
2 years ago
Philip O'Toole
98575d727b
Start adding remote remove node
...
More testing, including unit testing, required.
2 years ago
Philip O'Toole
65cd556549
Fix remote-load error handling
...
Error from remote node was being dropped.
2 years ago
Philip O'Toole
ec0750a4c2
Update service.go
2 years ago
Philip O'Toole
c8d521e628
Fix cluster service with 8-byte lengths
2 years ago
Philip O'Toole
9ff03d3028
Client-level unit testing of Load
2 years ago
Philip O'Toole
16a0b41321
Better errors
2 years ago
Philip O'Toole
1827c6eec5
Compress backups before transmission between nodes
2 years ago
Philip O'Toole
34c9914148
WIP -- unit test passing
2 years ago
Philip O'Toole
e492c5f5f0
More Backup-forwarding WIP
2 years ago
Philip O'Toole
1049e41b6c
WIP before moving to proto backup model
2 years ago
Philip O'Toole
3afa2ce80a
Stick with simple proto
...
The proto was not well designed in the first place, but it's better to
deal with it as-is, and wait until another major release to fix it.
2 years ago
Neal Harrington
e4a810d7ea
Treat error from rpc response as a first class object, delineate error types a bit, use this data to check auth when forward from http
2 years ago
Neal Harrington
c0b1b6c762
write some tests, make them pass
2 years ago
Neal Harrington
8921ee002e
Clean up unncessary conditional. Add explicit auth/ tests for empty credentials/anonymous case
2 years ago
Neal Harrington
7cceb822a4
First things
2 years ago
Philip O'Toole
d9b2e0b2fa
WIP creds forwarding
2 years ago
Philip O'Toole
ca7e6e98ff
Stop using deprecated protobuf package
3 years ago
Philip O'Toole
3f039a9b3a
Don't truncate cluster response length
3 years ago
Philip O'Toole
d90ea2e6b1
Service Node API requests locally if possible
3 years ago
Philip O'Toole
39615983d4
Merge branch 'master' of github.com:rqlite/rqlite into conn-pool-2
3 years ago
Philip O'Toole
711f6664f9
Service can return HTTP API URL
3 years ago
Philip O'Toole
3a70db5150
Long-lived connections in cluster service
...
Needed now that we're using connection pooling. Unit tests pass.
3 years ago
Philip O'Toole
25a51534da
Fix comments
3 years ago
Philip O'Toole
d503db616e
Better modeling with oneof
3 years ago
Philip O'Toole
1f56b5372f
cluster package unit tests passing
3 years ago
Philip O'Toole
796ef147f7
Hook in explicit Cluster Dialer
3 years ago
Philip O'Toole
281c454fc2
Refactor into distinct cluster client type
3 years ago
Philip O'Toole
ab0968644f
More error handling
3 years ago
Philip O'Toole
f101b2afdd
Check conn write errors
3 years ago
Philip O'Toole
4fd5555c41
ABL
3 years ago
Philip O'Toole
497ee77625
More Mux instrumentation
3 years ago
Philip O'Toole
0d533dfa4c
Some cleanup
3 years ago
Philip O'Toole
be7198f550
Go Doc
3 years ago
Philip O'Toole
2cef5679b7
rqlited starts and stops fine
3 years ago
Philip O'Toole
cbc2a686db
Instrument Cluster service
3 years ago
Philip O'Toole
f4639b2caf
Defer conn close
3 years ago
Philip O'Toole
f2947f84c4
Basic Cluster service in place
3 years ago
Philip O'Toole
e934630a58
Basic cluster RPC service
3 years ago
Philip O'Toole
f57ace7da2
Broadcast Store meta via standard consensus
...
With this change the cluster metadata (arbitrary key-value data associated with each node) is now broadcast across the cluster using the standard consensus mechanism. Specifically the use case for this metadata is to allow all nodes know the HTTP API address of all other nodes, for the purpose of redirecting requests to the leader.
This change removed the need for multiplexing two logical connections
over the single Raft TCP connection, which greatly simplifies the
networking code generally.
Original PR https://github.com/rqlite/rqlite/pull/434
5 years ago
Philip O'Toole
e376812907
Add method to Store that returns leader ID
...
With the advent of Raft IDs, this distinction matters.
5 years ago
Philip O Toole
bf2047b45b
'go lint' fixes
8 years ago
Philip O'Toole
fac48629bc
Start initial integration of mux ( #109 )
...
Integrate TCP mux with cluster and store
This change allows any node, including followers, to use the Raft log to make changes to a cluster-wide state.
9 years ago
Philip O Toole
aef17f6a52
Listener is actually a Transport
9 years ago
Philip O Toole
57880acc58
Better names for cluster package files
9 years ago