famosss
dc4037925d
fix: Build DeletedVids before reset dn's children ( #3530 )
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
chrislu
26dbc6c905
move to https://github.com/seaweedfs/seaweedfs
2 years ago
chrislu
bb01b68fa0
refactor
2 years ago
chrislu
68065128b8
add dc and rack
3 years ago
chrislu
3828b8ce87
"github.com/chrislusf/raft" => "github.com/seaweedfs/raft"
2 years ago
chrislu
24291e23eb
refactor
2 years ago
chrislu
9f20d3ebd1
add dc and rack
3 years ago
Konstantin Lebedev
6c20a3b622
avoid set currentMaster k8s svc.local discoveruy service domains
https://github.com/chrislusf/seaweedfs/issues/2589
3 years ago
chrislu
444ac21050
go fmt
3 years ago
guol-fnst
b12944f9c6
fix naming convention
notify volume server of duplicate directoris
improve searching efficiency
3 years ago
guol-fnst
8fab39e775
rename UUID file
fix typo
move locationUUID into DiskLocation
3 years ago
guol-fnst
de6aa9cce8
avoid duplicated volume directory
3 years ago
chrislu
94635e9b5c
filer: add filer group
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
c9952759c4
metrics master is leader
3 years ago
Konstantin Lebedev
28efe31524
new master metrics
3 years ago
Chris Lu
330d1fde7f
send peers info to filers
3 years ago
Chris Lu
84bb8e7365
send peers info to filers
3 years ago
Chris Lu
4b9c42996a
refactor grpc API
3 years ago
Chris Lu
5ea86ef1da
Revert "master: rename grpc function KeepConnected() to SubscribeVolumeLocationUpdates()"
This reverts commit af71ae11aa
.
3 years ago
Chris Lu
af71ae11aa
master: rename grpc function KeepConnected() to SubscribeVolumeLocationUpdates()
3 years ago
Chris Lu
5160eb08f7
shell: optionally read filer address from master
3 years ago
Chris Lu
2789d10342
go fmt
3 years ago
Chris Lu
e5fc35ed0c
change server address from string to a type
3 years ago
Chris Lu
7591336a22
log format
3 years ago
Chris Lu
5496d68f6a
increase counter only if not early terminated
3 years ago
Chris Lu
0128239c0f
handle ipv6 addresses
3 years ago
Chris Lu
006c01a519
fix format
3 years ago
Chris Lu
65af3cf4df
master: disconnect only the phantom volume server
fix https://github.com/chrislusf/seaweedfs/issues/2311
3 years ago
Chris Lu
8126ab4b5d
rename
3 years ago
Chris Lu
5469019852
adjust data type
3 years ago
Chris Lu
ee6c67682c
minor
4 years ago
Chris Lu
5d931eff27
avoid possible nil
fix https://github.com/chrislusf/seaweedfs/issues/1928
The nil was because of `dn.Parent().UnlinkChildNode(dn.Id())` in topo.UnRegisterDataNode() function, when the dn leaves the cluster.
4 years ago
Chris Lu
f8446b42ab
this can compile now!!!
4 years ago
Chris Lu
d156c74ec0
volume server set volume type and heartbeat to the master
4 years ago
Konstantin Lebedev
1eec5c8d5d
gen pb
4 years ago
Konstantin Lebedev
dc26012a3b
initial
4 years ago
Chris Lu
da4edf3651
master: check peers for existing leader before starting a leader election
fix https://github.com/chrislusf/seaweedfs/issues/1509
4 years ago
James Hartig
91e4eca1e9
Fix deadlock with KeepConnected and SendHeartbeat
There's the potential where we're writing to a clientConn and it goes away
and we're stuck keeping a read lock on clientChansLock. This causes
KeepConnected to not be able to remove the client since it requires a write
lock on clientChansLock. This ends up backing up SendHeartbeat because it
can't get a read lock.
4 years ago
Chris Lu
41d508edfd
go fmt
4 years ago
Chris Lu
be54eeb364
passing value by checkWithMaster() instead
4 years ago
Chris Lu
97733c3dff
typo
4 years ago
Chris Lu
a41588279a
change log level 5 to 4
4 years ago
Chris Lu
bee0d7e5eb
lower log priority for noisy heartbeat
4 years ago
Chris Lu
93ea0801ea
volume: the variable for the master node may be stale?
related to https://github.com/chrislusf/seaweedfs/issues/1414
4 years ago
Chris Lu
31583b5d6c
master and volume server: avoid race condition
The volume server may disconnect and reconnect to the same master.
The master's unregistration may happen after the reconnection. Thus the volume server will disappear.
5 years ago
Chris Lu
69eda558ef
reduce the connections
fix https://github.com/chrislusf/seaweedfs/issues/1277
this avoids sending error to wdclients, and clients retry the connections
5 years ago
Chris Lu
826f96b6f3
fix unreachable code
5 years ago