#datacenter id is different

1 messages · Page 1 of 1 (latest)

wanton bison
#

@unkempt gull That should not happen. Could you share output of bin/harvest doctor --print and also the sample of metrics where datacenter is changing

unkempt gull
wanton bison
#

which collector are you using?

unkempt gull
#

ZAPI

wanton bison
#

and output of bin/harvest doctor --print

unkempt gull
#

/opt/harvest # bin/harvest doctor --print
Admin:
Tools:
Exporters:
prometheus:
exporter: Prometheus
local_http_addr: 0.0.0.0
port: 12990
prometheus1:
exporter: Prometheus
port_range: 13000-14000
Defaults:
collectors:
- Zapi
- ZapiPerf
- Ems
use_insecure_tls: false
Pollers:
unix:
datacenter: local
addr: -REDACTED-
collectors:
- Unix
exporters:
- prometheus

#

the o/p is the from the harvest pod

wanton bison
#

okay. Checking on my side

#

hmm i don't think if raw_capacity is a metric we expose by default. Have you made any changes to disk template?

unkempt gull
#

yes I did

wanton bison
#

okay could you share

unkempt gull
#

`
name: Disk
query: storage-disk-get-iter
object: disk

counters:
storage-disk-info:
- ^^disk-uid
- ^^disk-name => disk
- disk-inventory-info:
- bytes-per-sector => bytes_per_sector
- right-size-sectors => right_size_sectors
- disk-ownership-info:
- ^home-node-name => node
- ^owner-node-name => owner_node

plugins:

  • MetricAgent:
    compute_metric:
    - raw_capacity MULTIPLY storage-disk-info.disk-inventory-info.bytes-per-sector storage-disk-info.disk-inventory-info.right-size-sectors

export_options:
instance_keys:
- node
- disk
instance_labels:
- owner_node
`

wanton bison
#

Could you share the correct config

#

okay i suspect you may have multiple pods running with different datacenter name pointing to same cluster. datacenter field is running just fine for me.