Leo Balduf
89bc479a3b
*: make use of timecache
2017-10-15 20:05:39 +02:00
Leo Balduf
5400a99b75
storage: add Benchmarks for ScrapeSwarm, optimize implementations
2017-08-30 08:54:11 +02:00
Leo Balduf
df7b59e2f3
storage: add PeerLifetime to test config
2017-08-17 23:35:20 +02:00
Leo Balduf
8ed171b0ea
pkg/log: create wrapper around logrus
2017-06-26 20:46:40 +02:00
Leo Balduf
1aa6c86d3f
storage: update config defaults
2017-06-25 19:04:24 +02:00
Jimmy Zelinskie
dab03f52dc
storage: share prometheus models
...
Running of the binary actually caused a panic due to multiple calling of
MustRegister(). This fixes that by sharing models in the storage
package.
2017-06-05 22:07:13 -04:00
Jimmy Zelinskie
c41519e73f
storage: sync memory & memorybysubnet
...
This change is a manual merge of all of the optimizations that have been
added to the memory peer store into the memorybysubnet peer store.
This also fixes some inconsistencies between the two.
2017-06-04 16:42:32 -04:00
Jimmy Zelinskie
7786e1a915
storage: enforce all peer stores are loggable
2017-06-04 16:06:01 -04:00
Jimmy Zelinskie
acf7e3c749
storage: register unique Prometheus metrics names
2017-06-03 15:47:58 -04:00
Jimmy Zelinskie
496cc1a31d
storage: dynamically register drivers
2017-06-03 15:47:58 -04:00
Cedric Charly
ce6706b0d5
storage/memory: set default prometheus_reporting_interval
...
When left blank in the config, the default prometheus_reporting_interval value
defaulted to 0, causing a panic during peer store initialization. This
change sets the default value to 1 if not provided.
Fixes #319
2017-06-03 14:21:39 -05:00
Leo Balduf
a70d6dc036
memory: add metrics for peers
2017-05-29 19:04:40 +02:00
Leo Balduf
143ba54695
storage/memory: remove calls to time.Now()
2017-05-12 20:27:31 +02:00
Leo Balduf
71eb9fb767
storage/config: fix example config, add warning for missing shardCount
2017-05-12 13:16:39 +02:00
Jimmy Zelinskie
cb55019ab8
*: add structured logging
2017-05-07 04:30:22 -04:00
Jimmy Zelinskie
ea0dba3a3d
cmd/chihaya: refactor root run command
...
This change refactors a bunch of the state of execution into its own
object. It also attempts to simplify stopping and adjusts some other
packages to integrate with the stopper interface.
Fixes #309 .
2017-05-01 15:56:38 -04:00
Jimmy Zelinskie
0e07b33827
prometheus: record IP AddressFamily
2017-02-02 19:19:33 -05:00
Leo Balduf
102b032c43
udp: fix ordering of scrapes
2017-01-30 13:07:14 +01:00
Leo Balduf
18d7e5d51b
storage: add prometheus to memory implementation
2017-01-25 18:59:10 +01:00
Leo Balduf
3c098c0703
middleware: add sanitization hook
2017-01-20 20:29:59 +01:00
Leo Balduf
70ceb96313
storage: add TestPeerStore test
2016-11-27 21:43:22 +01:00
Leo Balduf
97444f08a4
memory: add ScrapeSwarm method
2016-09-08 11:29:36 -04:00
Jimmy Zelinskie
fa32839623
replace std log w/ logrus
...
There still needs to be much more logged with the debug level.
2016-09-05 12:10:42 -04:00
Leo Balduf
62f2a095a1
memory: fix garbage collection, add config check
2016-08-17 17:15:54 -04:00
Jimmy Zelinskie
cc6614c474
rename back to chihaya
2016-08-16 22:17:10 -04:00
Jimmy Zelinskie
a553ded043
memory: add max numwant
2016-08-16 22:17:10 -04:00
Jimmy Zelinskie
6fddcb8eea
remove gc from storage, but implement it in memory
2016-08-16 22:17:10 -04:00
Leo Balduf
94dc902b20
comment/lint clean PeerStore
2016-08-16 22:17:10 -04:00
Leo Balduf
e1cf159d9c
separate IPv4 and IPv6 swarms on by shards
2016-08-16 22:17:10 -04:00
Leo Balduf
6a45107193
make benchmarks parallel, fix memory benchmarks
2016-08-16 22:17:10 -04:00
Jimmy Zelinskie
651ed50957
initial benchmarks. warning: /0 bug
2016-08-16 22:17:10 -04:00
Jimmy Zelinskie
c3137508d0
make new storage pkg with memory impl
2016-08-16 22:17:10 -04:00
