-
Notifications
You must be signed in to change notification settings - Fork 33
Expand file tree
/
Copy pathvalues-usdfprod-prompt-processing.yaml
More file actions
135 lines (120 loc) · 4.63 KB
/
values-usdfprod-prompt-processing.yaml
File metadata and controls
135 lines (120 loc) · 4.63 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
prompt-keda:
image:
pullPolicy: IfNotPresent
tag: 6.5.0
worker:
# TODO: need to adjust this once we know how leaky the LSSTCam pipeline is
# restart: 7
# Expect much less memory usage/leakage for Single-Frame only
restart: 30
instrument:
pipelines:
# IMPORTANT: don't use flow-style mappings (i.e., {}) in pipelines specs
# if the result (including any comments) is longer than 72 characters.
# The config will get corrupted after template substitution.
# Block-style mappings can have lines of any length.
main: |-
- survey: BLOCK-365 # FBS SV Field Survey
pipelines:
# TODO: enable ApPipe only when we're ready (resources, alerts, etc.)
- ${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/SingleFrame.yaml
- ${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr.yaml
- survey: BLOCK-T427 # Daytime checkout
pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr-cal.yaml']
- survey: BLOCK-351 # Bright star test
pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr-cal.yaml']
- survey: BLOCK-T451 # Initial alignment
pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr-cal.yaml']
# Hexapod tests
- survey: BLOCK-T413
pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr-cal.yaml']
- survey: BLOCK-T414
pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Isr-cal.yaml']
# Manual observations during SV blocks?
- survey: BLOCK-T407
pipelines: []
- survey: BLOCK-T364
pipelines: []
- survey: BLOCK-T365
pipelines: []
- survey: BLOCK-T366
pipelines: []
- survey: BLOCK-T367
pipelines: []
- survey: BLOCK-T368
pipelines: []
- survey: BLOCK-T369
pipelines: []
- survey: BLOCK-T373
pipelines: []
- survey: BLOCK-T460
pipelines: []
- survey: BLOCK-T461
pipelines: []
# Produces nextVisits but not images
- {survey: "BLOCK-T454", pipelines: []}
# Miscellaneous scripts, not always images
- {survey: "", pipelines: []}
# Ignore unknown events
- {pipelines: []}
preprocessing: |-
# TODO: run if and only if ApPipe is a possibility
# - survey: BLOCK-365
# pipelines: ['${PROMPT_PROCESSING_DIR}/pipelines/LSSTCam/Preprocessing.yaml']
- {survey: "", pipelines: []}
# Don't preprocess anything unknown
- {pipelines: []}
# TODO: need to adjust this based on observed slew accuracy
preloadPadding: 50
calibRepo: s3://rubin-summit-users
s3:
imageBucket: rubin-summit
endpointUrl: https://sdfembs3.sdf.slac.stanford.edu
raw_microservice: http://172.24.5.158:8080/presence
imageNotifications:
kafkaClusterAddress: prompt-processing-2-kafka-bootstrap.kafka:9092
topic: rubin-summit-notification
# TODO: need to adjust this based on observed nextVisit lead time (depends on scheduler) and PP prep time
# The shorter this is, the less capacity is wasted on canceled visits
imageTimeout: 120
apdb:
config: s3://rubin-summit-users/apdb_config/cassandra/pp_apdb_lsstcam.yaml
alerts:
username: kafka-admin
server: usdf-alert-stream-dev.lsst.cloud:9094
topic: lsst-alerts
# TODO: may need to override for debugging
logLevel: timer.lsst.activator=DEBUG timer.lsst.daf.butler=DEBUG lsst.diaPipe=VERBOSE lsst.rbClassify=VERBOSE lsst.daf.butler=VERBOSE
sasquatch:
# TODO: production Sasquatch not yet ready
endpointUrl: https://usdf-rsp-dev.slac.stanford.edu/sasquatch-rest-proxy
namespace: lsst.prompt.prod
auth_env: false
keda:
minReplicaCount: 3
# TODO: this is scaled for SingleFrame, with 25% margin
maxReplicaCount: 1100
# TODO: may need to override for debugging
# failedJobsHistoryLimit: 100
# TODO: may need to reduce if we don't have enough capacity
# redisStreams:
# expiration: 600
initializer:
# 6 retries is not enough time to fix, e.g., DB permissions problems
retries: 15
# Keep around for first few days so we can check in the morning PDT
cleanup_delay: 21600 # 6 hours
podAnnotations: {
edu.stanford.slac.sdf.project/usdf-embargo: "true"
}
podAnnotations: {
edu.stanford.slac.sdf.project/usdf-embargo: "true"
}
affinity:
nodeAffinity:
preferredDuringSchedulingIgnoredDuringExecution:
- weight: 10
preference:
matchExpressions:
- key: node-role.kubernetes.io/prompt-processing
operator: Exists