193 Commits

Author SHA1 Message Date
Ingo Oppermann
317b30c14c
Merge branch 'vod' into vod-auto-remove-node 2024-10-31 16:03:03 +01:00
Ingo Oppermann
bfb54ca177
Add GPU usage in cluster about API endpoint 2024-10-31 14:32:18 +01:00
Ingo Oppermann
e2def57a6f
Update API docs 2024-10-24 16:51:14 +02:00
Ingo Oppermann
2dbe5b5685
Add GPU support 2024-10-24 15:08:26 +02:00
Ingo Oppermann
df30a6b8e3
Replace timer-based SMA with a timer-less implementation 2024-10-23 11:08:13 +02:00
Ingo Oppermann
719449a4c8
Implement own byte buffer type 2024-10-10 16:35:39 +02:00
Ingo Oppermann
a581f1dbc2
User buffer pool where appropriate 2024-10-10 15:09:50 +02:00
Ingo Oppermann
0f6d7949c4
Fix deadlock in cluster shutdown 2024-09-17 15:08:11 +02:00
Ingo Oppermann
705c7fa946
Introduce budget for process relocation 2024-09-17 11:55:31 +02:00
Ingo Oppermann
0327edcaf3
Update openapi specification 2024-09-05 13:58:59 +02:00
Ingo Oppermann
bebef61e55
Add /v3/cluster/events endpoint to gather events from all nodes 2024-08-22 13:40:38 +02:00
Ingo Oppermann
0b1601542d
Wait for follower and leader loops to finish 2024-08-19 15:22:24 +02:00
Ingo Oppermann
1650b17e05
Simply return error as-is, check process list length 2024-08-19 15:21:24 +02:00
Ingo Oppermann
fa199736e0
Merge branch 'vod' into vod-auto-remove-node 2024-08-19 10:55:43 +02:00
Ingo Oppermann
d6d39f162a
Adding a TODO 2024-07-26 12:38:21 +02:00
Ingo Oppermann
d391e274d7
Fix wrong memory limit, add total memory, add cpu and memory consumed by core itself to node resources 2024-07-25 21:13:49 +02:00
Ingo Oppermann
0a74470d38
Don't mark processes as errNotEnoughResourcesForDeployment when budget has been used up 2024-07-24 12:54:45 +02:00
Ingo Oppermann
54b1fe8e86
Dump casbin, replace with own policy enforcer 2024-07-23 15:54:09 +02:00
Ingo Oppermann
879819f10f
Retrieve current process from leader, clone metadata, introduce new state 'deploying' 2024-07-22 16:58:57 +02:00
Ingo Oppermann
9e52f19a66
Introduce synchronize budget, experimental 2024-07-22 09:25:23 +02:00
Ingo Oppermann
308f008969
Only compare configs if the process will get replaced 2024-07-19 16:00:45 +02:00
Ingo Oppermann
72883d18d4
Remove bottlenecks in process handling, still some rough edges 2024-07-18 17:16:49 +02:00
Ingo Oppermann
4d0eed092e
Return error from ClusterProcessList, remove ProcessFindNodeID 2024-07-17 16:50:39 +02:00
Ingo Oppermann
db564de1f1
Use store.ProcessGetNode function 2024-07-17 16:49:09 +02:00
Ingo Oppermann
e12fb0be52
Fix cluster shutdown, limit parallel opstack worker 2024-07-17 16:48:33 +02:00
Ingo Oppermann
3df1075548
Add ProcessGetNode function 2024-07-17 16:47:00 +02:00
Ingo Oppermann
88739e3f7f
Cosmetics 2024-07-17 16:45:33 +02:00
Ingo Oppermann
3d78122053
Fix crash when updating unavailable node 2024-07-16 08:13:15 +02:00
Ingo Oppermann
7e90bb87ce
Allow to import report history for a process 2024-07-10 16:46:49 +02:00
Ingo Oppermann
787948080b
Fix name and use of CORE_CLUSTER_RECOVER_TIMEOUT_SEC 2024-07-09 14:44:34 +02:00
Ingo Oppermann
e306d20d55
Fix use of proxy 2024-07-09 14:14:47 +02:00
Ingo Oppermann
8dcda07fc1
Merge branch 'vod' into vod-auto-remove-node 2024-07-09 14:08:57 +02:00
Ingo Oppermann
480dbb7f53
Refactor cluster node code 2024-07-09 12:26:02 +02:00
Ingo Oppermann
28603aab98
Incorporate process throttling into deploy decision, fix bug in rebalance, parallelize opstack 2024-06-26 17:03:42 +02:00
Ingo Oppermann
ca177becfa
Fix tests 2024-06-24 17:37:04 +02:00
Ingo Oppermann
c032cdf5c7
Add API for setting node status, respect it in leader tasks 2024-06-24 16:50:15 +02:00
Ingo Oppermann
166e313642
Fix tests 2024-06-19 15:38:42 +02:00
Ingo Oppermann
a9d6b1ec49
Add API endpoints for relocating processes 2024-06-19 15:28:30 +02:00
Ingo Oppermann
de6a267fd4
Add operations to relocate processes 2024-06-18 16:50:59 +02:00
Ingo Oppermann
cc75be6372
Fix punctuation in log messages 2024-06-13 16:57:25 +02:00
Ingo Oppermann
8290d007b4
Fix minimal raft uptime requirement 2024-06-13 16:05:15 +02:00
Ingo Oppermann
0bf371807a
Add autorecovery, add CORE_CLUSTER_NODE_RECOVER_TIMEOUT_SECONDS 2024-06-13 15:55:18 +02:00
Ingo Oppermann
f5d9725a48
Return proper HTTP status on leave 2024-06-12 15:08:07 +02:00
Ingo Oppermann
a1b1609e73
Return proper HTTP status on leave 2024-06-12 15:05:05 +02:00
Ingo Oppermann
2c17572027
Drop dead nodes after CORE_CLUSTER_NODE_RECOVER_TIMEOUT_SEC 2024-06-12 14:42:27 +02:00
Ingo Oppermann
1a64fddbb1
Allow cluster leave endpoint to remove any node in the cluster 2024-06-07 11:28:54 +02:00
Ingo Oppermann
03da97217b
Use faster JSON and gzip packages 2024-04-24 15:42:11 +02:00
Ingo Oppermann
423e0453de
Remove NodeID from ClusterAbout 2024-04-10 16:58:34 +02:00
Ingo Oppermann
4bc08d2b1e
Lift some message from the cluster to info 2024-04-10 10:39:34 +02:00
Ingo Oppermann
7a4875f818
Promote core version in cluster about 2024-04-09 15:57:17 +02:00