Konstantin Lebedev
|
6199d9b923
|
rand start garbageThreshold and reset metric ReplicaPlacementMismatch (#3396)
* rand start garbageThreshold
reset metrics MasterReplicaPlacementMismatch
* vacuum through sleep
|
2 years ago |
Konstantin Lebedev
|
4d08393b7c
|
filer prefer volume server in same data center (#3405)
* initial prefer same data center
https://github.com/seaweedfs/seaweedfs/issues/3404
* GetDataCenter
* prefer same data center for ReplicationSource
* GetDataCenterId
* remove glog
|
2 years ago |
Konstantin Lebedev
|
2a5629dd28
|
align
|
2 years ago |
Konstantin Lebedev
|
df9cc31a0a
|
refactor
|
2 years ago |
Konstantin Lebedev
|
0a5d3c3cf6
|
rename to localWrite and replicatedWrite
|
2 years ago |
Konstantin Lebedev
|
fc65122766
|
rename to LoadAvg_1M
|
2 years ago |
Konstantin Lebedev
|
ab50f21ad9
|
Detailed metrics VolumeServerRequestHistogram for writing to disk and replication
|
2 years ago |
Konstantin Lebedev
|
5209ebbeef
|
remove percent
|
2 years ago |
Konstantin Lebedev
|
78cbd8002f
|
revert Sleep
|
2 years ago |
Konstantin Lebedev
|
cd5c7ad052
|
move to github.com/seaweedfs/seaweedfs
|
2 years ago |
Konstantin Lebedev
|
c0d92f61a1
|
comment
|
2 years ago |
Konstantin Lebedev
|
1d29f67c02
|
revert disk stats
|
2 years ago |
chrislu
|
26dbc6c905
|
move to https://github.com/seaweedfs/seaweedfs
|
2 years ago |
Konstantin Lebedev
|
2f0dda384d
|
vacuum show LA
|
2 years ago |
chrislu
|
3828b8ce87
|
"github.com/chrislusf/raft" => "github.com/seaweedfs/raft"
|
2 years ago |
ningfd
|
58f2dd6740
|
change topology mapping definition, improve spelling
|
2 years ago |
ningfd
|
6f882eb354
|
optimiz: master ui will render data in order
|
2 years ago |
chrislu
|
48382676d2
|
fix filtering by volume id
|
2 years ago |
chrislu
|
a79f5bd0d5
|
adjust used size reporting
|
3 years ago |
chrislu
|
3142e77eca
|
mount: stats report physical size instead of logical size
|
3 years ago |
chrislu
|
96496d5286
|
master: broadcast new volume locations to clients to avoid possible racing condition
fix https://github.com/chrislusf/seaweedfs/issues/3220
|
3 years ago |
zzq09494
|
9df5ad5309
|
fix: vacuum create a lot of connections quickly
|
3 years ago |
Konstantin Lebedev
|
fb57aa431c
|
stats master_replica_placement_mismatch
|
3 years ago |
guol-fnst
|
b12944f9c6
|
fix naming convention
notify volume server of duplicate directoris
improve searching efficiency
|
3 years ago |
guol-fnst
|
de6aa9cce8
|
avoid duplicated volume directory
|
3 years ago |
qiang.zeng
|
06cb337b82
|
fix typo
|
3 years ago |
chrislu
|
00c1dfec4f
|
go fmt
|
3 years ago |
Konstantin Lebedev
|
36c5a59ed8
|
add help
|
3 years ago |
Konstantin Lebedev
|
1e35b4929f
|
shell vacuum volume by collection and volume id
|
3 years ago |
chrislu
|
4524ceae6b
|
less verbose logs
fix https://github.com/chrislusf/seaweedfs/discussions/2925#discussioncomment-2584085
|
3 years ago |
shibinbin
|
c20e1edd99
|
fix: master lose some volumes
|
3 years ago |
chrislu
|
bc888226fc
|
erasure coding: tracking encoded/decoded volumes
If an EC shard is created but not spread to other servers, the masterclient would think this shard is not located here.
|
3 years ago |
Konstantin Lebedev
|
14dd971890
|
hashicorp raft with state machine
|
3 years ago |
Konstantin Lebedev
|
c514710b7b
|
initial add hashicorp raft
|
3 years ago |
chrislu
|
ae558fa073
|
log reasons volumes became unwritable
|
3 years ago |
chrislu
|
57c6eddd22
|
avoid possible deadlock if volume layout is used in some logs
|
3 years ago |
Konstantin Lebedev
|
9ea09cc41c
|
healthz check to avoid drain pod with last replicas
|
3 years ago |
Konstantin Lebedev
|
0ed76a0556
|
clearly
|
3 years ago |
Konstantin Lebedev
|
36013f63ed
|
https://github.com/chrislusf/seaweedfs/issues/2648
|
3 years ago |
chrislu
|
433fde4b18
|
move error to a separate file
This file contains metric names for all errors
The naming convention is ErrorSomeThing = "error.some.thing"
|
3 years ago |
Konstantin Lebedev
|
3f4e17aa24
|
error metrics for filer and store
|
3 years ago |
chrislu
|
9f9ef1340c
|
use streaming mode for long poll grpc calls
streaming mode would create separate grpc connections for each call.
this is to ensure the long poll connections are properly closed.
|
3 years ago |
banjiaojuhao
|
dda6b90d25
|
assign fileId according to DataNode with empty DataCenter and Rack
|
3 years ago |
chrislu
|
5eacff9d4f
|
log message adds server name
address https://github.com/chrislusf/seaweedfs/issues/2514#issuecomment-995925733
|
3 years ago |
Chris Lu
|
3be3c17f59
|
volume vacuum: avoid timeout with streaming progress report
fix https://github.com/chrislusf/seaweedfs/issues/2396
|
3 years ago |
Chris Lu
|
e4830bd93d
|
go fmt
|
3 years ago |
Chris Lu
|
332d49432d
|
reduce concurrent volume grow requests
|
3 years ago |
Chris Lu
|
96119eab00
|
refactor
|
3 years ago |
Chris Lu
|
8a66306064
|
calculate disk usage in case of race condition
related to https://github.com/chrislusf/seaweedfs/issues/2357
|
3 years ago |
Chris Lu
|
a067deaabc
|
avoid possible modified location list
fix issue 1 of https://github.com/chrislusf/seaweedfs/issues/2345
|
3 years ago |