Documentation ¶
Index ¶
- type Devops
- func (d *Devops) GenerateEmptyQuery() query.Query
- func (d *Devops) GroupByOrderByLimit(qi query.Query)
- func (d *Devops) GroupByTime(qi query.Query, nHosts, numMetrics int, timeRange time.Duration)
- func (d *Devops) GroupByTimeAndPrimaryTag(qi query.Query, numMetrics int)
- func (d *Devops) HighCPUForHosts(qi query.Query, nHosts int)
- func (d *Devops) LastPointPerHost(qi query.Query)
- func (d *Devops) MaxAllCPU(qi query.Query, nHosts int)
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type Devops ¶
Devops produces SiriDB-specific queries for all the devops query types.
func (*Devops) GenerateEmptyQuery ¶
GenerateEmptyQuery returns an empty query.SiriDB
func (*Devops) GroupByOrderByLimit ¶
GroupByOrderByLimit populates a query.Query that has a time WHERE clause, that groups by a truncated date, orders by that date, and takes a limit:
select max(1m) from `usage_user` between time - 5m and 'roundedTime' merge as 'max usage user of the last 5 aggregate readings' using max(1)
func (*Devops) GroupByTime ¶
GroupByTime selects the MAX for numMetrics metrics in the group `cpu`, per minute for nhosts hosts, e.g. in pseudo-SQL:
In case of 1 metric: select max(1m) from (`groupHost1` | ...) & `groupMetric1` between 'time1' and 'time2' merge as 'max METRIC for (HOST | ... ) using max(1)
In case of multiple metrics NOTE: it is not possible to merge multiple hosts per metric for a list of metrics, only for one metric. In this case all series for the provided hosts and metrics are returned:
select max(1m) from (`groupHost1` | ...) & (`groupMetric1` | ...) between 'time1' and 'time2'
func (*Devops) GroupByTimeAndPrimaryTag ¶
GroupByTimeAndPrimaryTag selects the AVG of numMetrics metrics in the group `cpu` per device per hour for a day, e.g. in pseudo-SQL:
select mean(1h) from (`groupMetric1` | ...) between 'time1' and 'time2'
func (*Devops) HighCPUForHosts ¶
HighCPUForHosts populates a query that gets CPU metrics when the CPU has high usage between a time period for a number of hosts (if 0, it will search all hosts), e.g. in pseudo-SQL:
NOTE: It is not possible to return the other cpu metrics when e.g. usage_user has reached a threshold. Here only the metric "usage_user" is returned.
nHosts=0: select filter(> 90) from `usage_user` between 'time1' and 'time2' nHosts>0: select filter(> 90) from `usage_user` & (`groupHost1` | ...) between 'time1' and 'time2'
func (*Devops) LastPointPerHost ¶
LastPointPerHost finds the last value of every time serie within the CPU group.
select last() from `cpu`
func (*Devops) MaxAllCPU ¶
MaxAllCPU selects the MAX of all metrics in the group `cpu` per hour for nhosts hosts, e.g. in pseudo-SQL:
NOTE: it is not possible to merge multiple hosts per metric for a list of metrics, only for one metric. In this case all series for the provided hosts and all `cpu` metrics are returned):
select max(1h) from (`groupHost1` | ...) & `cpu` between 'time1' and 'time2'