InfluxDB state history

Contents

InfluxDB state history

Allows to store item states history in InfluxDB, v1 and v2 protocols are supported.

The service provides unified database EAPI.

Note

Time convention used in EVA ICS time-frames to fill data:

S for seconds (e.g. 5S for 5 seconds)
T for minutes
H for hours
D for days
W for weeks
A to get automatic number of records (e.g. 5A for 5 exactly records)

Time convention used in EVA ICS to specify start/end of a timeframe:

If there are dedicated parameters for start/end, they are filled separately
If there is a single parameter only, it is filled either as START:END or as START only (END is automatically set to the current time)

where values must be:

UNIX timestamps only (if a parameter is strictly specified as a number)
Date in human-readable format (RFC3339 recommended)
Using the same notation as for filling. E.g. setting start=30T sets time-frame start to 30 minutes before now.

START value is always mandatory. If END value is required to be set to the current time, it can be omitted.

Setup

Use the template EVA_DIR/share/svc-tpl/svc-tpl-db-influx.yml:

# Event-to-InfluxDB database service
command: svc/eva-db-influx
workers: 1
bus:
  path: var/bus.ipc
config:
  api_version: 1 # 1 or 2
  ## v1 basic auth
  # username: user
  # password: pass
  ## v2 token auth
  # token: sometoken
  url: http://localhost:8086
  db: eva
  ## alternative CA file (.der for DER, any other extension for PEM)
  #tls_ca: path/to/crt
  ## required for v2
  # org: myorg
  # event buffer time-to-live, seconds
  buf_ttl_sec: null
  # periodically submit all item states
  interval: null
  # do not submit remote disconnected items (useful for zfrepl or similar)
  skip_disconnected: false
  # ignore real-time events
  ignore_events: false
  # event queue size
  queue_size: 8192
  # panic on critical errors in X seconds after happened (0 - panic immediately)
  panic_in: 0
  # influx client pool size (default = workers)
  #clients: 2
  # item OIDs / OID masks
  oids:
    - "#"
  oids_exclude: []
user: nobody

Create the service using eva-shell:

eva svc create eva.db.i1 /opt/eva4/share/svc-tpl/svc-tpl-db-influx.yml

or using the bus CLI client:

cd /opt/eva4
cat DEPLOY.yml | ./bin/yml2mp | \
    ./sbin/bus ./var/bus.ipc rpc call eva.core svc.deploy -

(see eva.core::svc.deploy for more info)

EAPI methods

See EAPI commons for the common information about the bus, types, errors and RPC calls.

state_history

Description	Gets item state history
Parameters	required
Returns	State history payload

Parameters
Name	Type	Description	Required
i	String	Item OID	yes
t_start	f64	Beginning timestamp (default: last 24 hours)	no
t_end	f64	Ending timestamp (default: now)	no
fill	String	Fill (nS/T/H/D/W e.g. 10T for 10-minute)	no
precision	u32	Round values to digits after commma	no
limit	u32	Limit records to	no
xopts	Map<String, String>	Extra: vfn=fn for value grouping (d: mean), rp=X for retention policy	no
compact	bool	Pack data in arrays according to type	no

Return payload example:

[
    {
        "status": 1,
        "t": 1652059860.0424938,
        "value": 15
    },
    {
        "status": 1,
        "t": 1652059865.045223,
        "value": 15
    },
    {
        "status": 1,
        "t": 1652059870.0452943,
        "value": 15
    },
    {
        "status": 1,
        "t": 1652059875.0443518,
        "value": 15
    }
]

state_log

Description	Gets item state log
Parameters	required
Returns	State log payload (includes OIDs, as other svcs may support get-by-mask)

Parameters
Name	Type	Description	Required
i	String	Item OID	yes
t_start	f64	Beginning timestamp (default: last 24 hours)	no
t_end	f64	Ending timestamp (default: now)	no
limit	u32	Limit records to	no
xopts	Map<String, String>	Extra: rp=X for retention policy	no

Return payload example:

[
    {
        "oid": "sensor:tests/temp",
        "status": 1,
        "t": 1652060175.0443184,
        "value": 15
    },
    {
        "oid": "sensor:tests/temp",
        "status": 1,
        "t": 1652060180.046056,
        "value": 15
    },
    {
        "oid": "sensor:tests/temp",
        "status": 1,
        "t": 1652060185.0454304,
        "value": 15
    }
]

state_push

Description	push item states into db, (payload: single item state or list)
Parameters	none
Returns	nothing

Retention policies

In InfluxDB v1, retention policies can be created as the following:

CREATE RETENTION POLICY "daily" ON "eva" DURATION 1D REPLICATION 1
CREATE CONTINUOUS QUERY "downsampled_env_temp1_30m" ON "eva" BEGIN
  SELECT mode(status) as "status",mean(value) as value
  INTO "daily"."sensor:env/temp1"
  FROM "sensor:env/temp1"
  GROUP BY time(30m)
END

To process all items with the same downsampled rate, set the continuous query to:

SELECT mode(status) as status, mean(value) as value
INTO "daily".:MEASUREMENT
FROM /.*/ WHERE time > now() - 1d
GROUP BY time(30m);