3bfcbffc37
Fix typos and Grammar issues
3 years ago
659bf26d7f
Better name for command-line flag
3 years ago
d74c05c2db
Support disabling in-memory init of on-disk
3 years ago
a4dc56e679
Ignore own join address if passed
...
Fixes https://github.com/rqlite/rqlite/issues/713 .
3 years ago
0a31c90b06
Ignore disco ID if there is preexisting state
...
Fixes https://github.com/rqlite/rqlite/issues/347
3 years ago
0b621cbec8
More logging tweaks
3 years ago
2e7bf1c881
Even friendlier log message when start-up is done
3 years ago
38a16849ca
Record compiler in logs and status
3 years ago
25f4c70d04
Update URL in logog
3 years ago
cdd83ce9e9
Start the HTTP server before doing any joins
...
There is no point waiting for join operations to complete before
starting the HTTP server. In the event the join fails it just introduces
delays due to join-retrys. The HTTP API service is meant to be robust in
the face of a non-ready underlying Store anyway.
3 years ago
dd71c07c73
ABL
3 years ago
1ea29e108a
Exit if on-disk-path is set, but on-disk is not
3 years ago
9965f520a8
Add a performance-related guide
3 years ago
105a894fe5
Support explicit SQLite path at Store level
3 years ago
81a234abd0
Command line option for SQLite on-disk file path
3 years ago
9aecafedbe
Cluster client resolves advertised local address
...
Also, improve HTTP stats.
3 years ago
a8d436e944
Improve stats
3 years ago
3ff1b7092a
Merge branch 'master' of github.com:rqlite/rqlite
3 years ago
f71bfd183d
Update CHANGELOG
3 years ago
f1a7e7e8ec
Use locally-enabled cluster client in HTTP service
3 years ago
a4638ad514
Set Raft Adv once and for all
3 years ago
071112aef7
Pass Store to Cluster service
3 years ago
796ef147f7
Hook in explicit Cluster Dialer
3 years ago
606b45051b
Fix help typo
3 years ago
9ed32972b1
Start adding FK option support
3 years ago
b8d72f3fc4
New DB and FSM indexes to track state
3 years ago
e0da96dd55
Remove support for specifying SQLite DSN
...
This is a breaking change. However this feature hasn't been tested and
allows end-users to break the system too easily. Low-level control over
the SQLite database is better done with PRAGMA commands where possible.
3 years ago
f0f8bf0388
Change what-was too noisy logging
3 years ago
4591e2df35
Don't ignore -join even if previous state exists
3 years ago
232449b113
Correct help text for -node-id
3 years ago
4c43049f9c
Start cluster service as early as possible
3 years ago
585a204f8b
Add more startup logging
3 years ago
088a06bfc9
Try simple way to share build information
3 years ago
697132a6ee
Remove metadata from Join operations
...
It's no longer needed.
3 years ago
0d533dfa4c
Some cleanup
3 years ago
f939d33fe5
Remove Metadata concept from system
...
System testing passes.
3 years ago
2cef5679b7
rqlited starts and stops fine
3 years ago
30d4d63ebb
System tests pass
3 years ago
38b2abca84
Remove older TCP transport code
...
Still unclear about whether tn.Open() is required with new mux.
3 years ago
d237a13a59
Use muxed connection for Raft internode comms
3 years ago
98a7b3d30e
Minor cleanup for mux code
3 years ago
662daa74cb
Add CA cert support which was erroneously removed
...
This was clearly a mistake due to a previous code change.
3 years ago
2a55f98af7
Tweak help message for source IP
...
https://github.com/rqlite/rqlite/issues/774
4 years ago
03e616d1f7
fix: Optimize the Conversation from O\'Toole
4 years ago
3ba5e31fb2
feat: Support to specify the source IP when you are in a complex network environment
4 years ago
2c52d9726c
Log launch command
4 years ago
c782b43f2d
Build on-disk databases in-memory first
...
With this change, rqlite nodes running in "on-disk" mode build the database in memory first, and move it to the disk just before the Raft system starts. This means that on-disk nodes now initialize almost as quickly as in-memory nodes.
4 years ago
c4ee9cafd9
Disable TLS v1.0 and v1.1 by default
4 years ago
5295a0e6b1
Trivial cleanups
4 years ago
b27cdfbf4d
Better control over waiting for leader
4 years ago
6670a22850
Actually check error codes during Store startup
...
This was a bug.
4 years ago
f07c3673c5
Log at start-up first and last Raft log indexes
4 years ago
499df692be
Exit if arguments are passed after data directory
...
Fixes issue #718 .
4 years ago
a40338b113
Tweak log output
4 years ago
d5322a1909
Ignore join request if node already part of cluster ( #711 )
...
rqlite used to work like this, but suffered a regression due to a change in how Hashicorp Raft worked. The manner it changed in was not public, so relying on it was always fragile.
4 years ago
6575de779d
Use Protobuf for encoding Raft Log commands
...
This PR changes Raft Log Entry encoding from JSON to Protobuf. Furthermore, larger Raft commands (which can result from batching SQL statements, or individually long SQL statements) are compressed before encoding.
This primary reason for this change is to reduce IO load since that is one of the largest performance bottlenecks. It will also reduce internode traffic.
Legacy JSON-encoded commands are still handled by this code, so this change is backwards-compatible with previous releases in the v5 series.
4 years ago
b6ebdf3ab1
Support setting Leader Lease timeout
4 years ago
fbdbee3e25
Include API proto in node meta
4 years ago
3f0182c8e8
Make num join attempts and interval configurable
4 years ago
487c0e3f8c
Set Snapshot Interval to non-zero value
...
Snapshotting was *never* functional due to the zero interval value.
5 years ago
536cac4fdf
Allow log level control for Raft module
...
Default log level is INFO.
5 years ago
668720cbb6
Remove obsolete SQLite const
5 years ago
ef44c199aa
Update main.go
5 years ago
cc65e36fcc
Remove obsolete const
5 years ago
81f5f6faf1
Add support for non-voting nodes
...
A non-voting node doesn't participate in Raft consensus, but does
subscribe to the committed log entries originating with the leader.
This means a non-voting node keeps up-to-date with the state machine,
without impacting write-latency. These non-voting nodes can provide
read scalability for the cluster.
5 years ago
3b9aca29c2
Add features list to status output
...
This will make it easier for client libraries to check for features in a given server, in a standard way.
5 years ago
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
853133ecb2
By default use Raft network address as node ID
5 years ago
8a0e1ce8f0
Move to Hashicorp Raft v1
5 years ago
9499a7f92d
Revert changes to Disco client
5 years ago
de379a883c
When joining try HTTPS if HTTP fails
5 years ago
f01e6b1b0a
Server -http-ca-cert and -node-ca-cert options
...
The -http-ca-cert and -node-ca-cert options allow the user to specify
trusted X.509 root CA certificates as an alternative to the
-http-no-verify and -node-no-verify options. This behavior is analogous
to the rqlite client -ca-cert option.
5 years ago
2e91858e1e
Allow Raft election timeout to be set
6 years ago
6937a377ae
Allow control of shutdown-on-remove
...
Defaults to leaving Raft run if the node is removed.
6 years ago
c1e1509b93
Correct rqlite GoDoc
7 years ago
3fcc5d36f6
Refactor credential store instantiation
...
This will make it easier for other services to use the credential store.
7 years ago
aa998db89e
More GoDoc
7 years ago
1a11536495
Include Go runtime version when issued -version
7 years ago
c8fb4dc9cf
Fix HTTP x509 key help
7 years ago
d753a8f7be
HTTP service now supports registered statuses
...
Mux is first service to register status for serving.
7 years ago
75cec9a9c1
Tweak start-up logo
...
Strictly speaking the system is rqlite, not rqlited. The latter is just
the daemon.
7 years ago
ebfa38d726
Move to clearer command-line options
7 years ago
5e2718d7f7
Better TLS-related logging
7 years ago
a0753d9968
Unit test TLS mux
7 years ago
8f8894bd0e
Complete node-to-node encryption
7 years ago
b96220f1eb
Better commenting
7 years ago
ba53bde877
Add creation TLS-enabled listener
7 years ago
ece0994057
Command-line options for secure internode comms
7 years ago
6e664b9797
Log Go version at startup
8 years ago
57495a8457
Support multiple join addresses
8 years ago
299f5dee7e
Tweak rqlite description
8 years ago
fb92953a88
const out the name 'rqlited'
8 years ago
4fad2d22df
Tweak join-related logging
8 years ago
159da1b5e5
Check if join allowed before determining addresses
8 years ago
b1ddec679d
Don't name variable after package
8 years ago
0ae7425678
Disco service is capitalized in logs
8 years ago
acaa014b49
Merge pull request #286 from rqlite/fix_http_help
...
Tweak help output
8 years ago
4b9220c107
More join logging
8 years ago
abd24c75ec
Tweak help output
8 years ago
144514f370
Tweak main log message
8 years ago
218610822c
Merge pull request #283 from rqlite/fix_main_log
...
main code should also log to stderr
8 years ago
4d145f8ac5
main code should also log to stderr
8 years ago
9d39d59a6c
Fix join failure log message
8 years ago
cf39350d62
Start fixing Discovery Redirects
...
The Go HTTP client does some automatic redirect handling, and the code needs to deal with this.
8 years ago
7810c3470a
Complete integration with Discovery Service
8 years ago
a61f05985b
Supply HTTP advertised address to disco service
8 years ago
35528928e2
Integrate cluster.Join with main
8 years ago
8f9a26ed7b
Add initial, more complex, join functions
...
Also, remove questionable check of join and disco.
8 years ago
b0abb64d62
Sart integration of disco service
8 years ago
caf1562a77
Add GOOS and GOARCH to --version output
8 years ago
e6d9a9dda8
Add build branch to -version output
8 years ago
851f486ff8
Make waiting for logs optional
8 years ago
03899af510
Set params on store before opening
8 years ago
059ab0757b
Allow Store to wait for initial logs to be applied
...
Better fix for issue #260 .
8 years ago
5b0cc7bb98
Log GOOS and GOARCH at startup
8 years ago
ae0d119d83
Allow Raft Apply timeout to be configurable
8 years ago
982fa54030
Not all queries begin with SELECT so pull check
...
Common Table Expressions can begin with WITH. This check was always dubious.
8 years ago
326ca757d7
add an API for using custom raft.PeerStore/log.Logger
...
This is useful in case the server needs to store other metadata (e.g. auth
data) along side the peer list.
The logging bit is handy in case something has its own logging framework
that it wants to use.
Signed-off-by: Tycho Andersen <tycho.andersen@canonical.com>
8 years ago
39ebf50cb7
Remove explicit control of FK constraints
...
It's not clear whether it's the right thing to do, and FK constraints
can still be controlled via PRAGMA commands.
8 years ago
c691b04dca
Tweak usage output
8 years ago
4cf39b626e
Tweak logo
8 years ago
d7833c66c4
Set build variables default at declaration time
8 years ago
9ad9c593fb
Remove unreachable code
8 years ago
aa599a34e3
Use HTTP status codes, not magic numbers
8 years ago
4a74e439e5
Join redirects now fully functional
8 years ago
e11aa14b6e
Support memory profiling
8 years ago
c40fff4d46
Support controlling FK constraints at startup
...
This was possible previously, but would need to be set everytime on
startup via the API. This change allows it to set at startup AND enables
foreign constraint checking by default.
8 years ago
2982058a81
Support setting Raft heartbeat timeout
8 years ago
8a05236768
Support configurable snapshot thresholds
8 years ago
6f4814a4c3
Support pprof information over HTTP
8 years ago
786a8b0464
Used resolved address when joining
...
It seems Hashicorp Raft only works in IP addresses, and joining with a
hostname caused the remove logic to cause all nodes to leave when one
node was instructed to leave.
8 years ago
8975a4744d
Dump full body on join error
8 years ago
2c334d4f93
Allow SELECT-verification to be disabled
...
This is not recommended as it leaves the system vulnerable to changes
via the "query" endpoint.
8 years ago
1c6ef1d84f
Use Raft advertise address, if exists, during join
9 years ago
4a5402fc9a
Allow HTTP advertise address to be set
9 years ago
f16b6ba55b
Allow Raft address advertise ( #115 )
...
* Use resolved Raft address as API peer key
* Allow Raft advertise address to be set
* Better log message for mux
* CHANGELOG updates
* Unit test mux layer address advertise
9 years ago
c1e2ab90e4
Switch to in-mem database by default ( #113 )
...
With low-latency networks, this results in a noticeable performance
increase.
9 years ago
f115f6fbef
Version is now 3
9 years ago
a35b877ba6
Publish peer before starting http service
...
This ensures the node is not accessible until it's completed its housework.
9 years ago
78f652fcc2
Enable leader redirection
...
Not quite complete.
9 years ago
3c02cbdc53
Actually start mux
9 years ago
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
5aefda9983
Move to new rqlite org
9 years ago
e8c92e088c
Don't join unless actually necessary
9 years ago
888e80294d
Actually close response body
9 years ago
58ba3463b5
Always close response body during join
9 years ago
fb5e3c99d3
Bring system-testing into line with new DB layer
9 years ago
75ee23bdd9
Move URL normalization to http package
...
This function may be useful for leader-redirection.
9 years ago
659be46190
Add build time to status output
9 years ago
59cddcc2f7
Correctly indicate no auth enabled
9 years ago
7fd7a9a385
Finish multi-node cluster testing
9 years ago
4c36f8345c
Create standalone auth module
9 years ago
dce5e33016
Display logo on startup
9 years ago
f94a10424c
Check for failing auth on join
9 years ago
c1ad8c8ec1
Allow auth config to be set at command line
9 years ago
3ccb97c98b
Merge pull request #84 from otoolep/https_support
...
Add HTTPS support to API
9 years ago
5fb2b0a658
Less precise version
9 years ago
43195b37cd
Allow node to accept insecure X.509 certs
9 years ago
f23cd8ef9c
Support HTTPS API
...
API can now be secured with X.509 certficate and key.
9 years ago
aee03269db
Tweak help output for -join
9 years ago
6cb929954b
Trivial cleanup of main code
9 years ago
7d8980878b
Allow join address with protocol
...
This is part of the work for issue #82 .
9 years ago
b00084a771
Include commit information in version
9 years ago
09eb50d2bb
Support dumping version via command-line option
9 years ago
fa8c9b20c2
Add build information to status output
9 years ago
eba25cfda4
Allow build version to be set
...
In addition, log it at startup.
9 years ago
076a996a8b
Support expvar data
9 years ago
b705de5240
Faulty leader redirection still disabled
9 years ago
4524d014eb
Leader redirection almost complete
9 years ago
faaaacd2a6
Remove reporting
9 years ago
59052ec0b4
More status output
9 years ago
ce0a2ce269
Usage and comment fixes
9 years ago
593719f86c
Remove obsolete comment
9 years ago
9c02e1edb6
Make it easier to understand DSN
9 years ago
4ff67c2c90
Support in-memory databases
9 years ago
17aea81d75
Add Raft statistics
9 years ago
48b6a469e3
Update main.go
9 years ago
d0b4d6aa0a
RWADME cleanups
9 years ago
a583b70180
Enhance reporting so Hashicorp use is flagged
9 years ago
35803cdf75
Shutdown Store on exit
9 years ago
a6cb950ff4
Allow DSN config at command line
...
It might still need to be richer, so end-users could specify an
in-memory SQLite database. Specifying a DSN would require the user
to supply the full path to the SQLite database. This is OK. However,
the code then needs to be able to parse out the path to the database
so it can remove it before start up.
9 years ago
2ebec9b973
Start adding support for snapshots
...
This shows that passing the database into the Raft module is probably
not going to work, since the database could be opened in two ways -
directly at startup, and be completely restored from the log, or with
a combination of restoring from a snapshot, followed by the remaining
log entries. In both cases the database must be opened using the
requested DSN settings.
A detailed config object for controlling SQLite behaviour, is probably
best, and it should be passed to the Raft store on start up.
Finally, file-level copying of the SQLite file can only take place if
no transaction is in effect. This might be handled by the use of a
RWLock. The write-lock is taken during Execute() and Snapshot, but
the Read lock is taken during Query(). Unfortunately this may reduce
the concurrency of inserts and updates. Perhaps the Backup call on
the Go SQLite library might be better, but it might be slow.
9 years ago
ab71c3bdd2
Execute working now
9 years ago
d67957d16d
Tweak DSN help output
9 years ago
3d277dd8fb
Update help output for DSN override
9 years ago
d70a9cb0ed
Push database into Raft store
...
Allows the DSN to be fully set, if requested.
9 years ago
73d880c93c
Start support for SQLite DSN
...
This isn't complete, since it prepends the Raft directory to the
filename, which would break a DSN.
9 years ago
875bb6ebf6
Raft clustering functional
9 years ago
12aef34625
Start integrating actual SQLite support
...
Builds without error.
9 years ago
d7d46f79c0
Start refactor for Hashicorp
...
Heavily influenced by hraftd. rqlited builds without error.
9 years ago
chermehdi
Philip O'Toole
osxlinux
Zac Medico
Tycho Andersen