* Query label values and names are now fetched from the store.

A time range is now required by the /api/prom/label with a sane default (6 hours from now).

* fix http querystring and update doc

* update vendor

* rebased

* move Dockerfile multi-arch to build and ignore that folder

* docker-driver fix

* fix(loki): pass missing config error to user

Missing config errors are handled at the library level. Our own check mitigated
this and causes loki to SEGFAULT later on

* feat(loki): default config file in container

The container provides a default config file. Use it by default

* Revert "fix(loki): pass missing config error to user"

This reverts commit b2744fcf427b2b386dce5e999c855ae5c899e4e2, because loki it
assumed loki was incapable of running without config, which is not the case.

Adds an entirely new Dockerfile to the repository root which is capable of:

- building promtail and loki from the same image
- based on `scratch`, so that the final image can be executed on every supported
  GOARCH

Please note that this Dockerfile should always be built using
BuildKit (`buildctl` or `DOCKER_BUILDKIT=1` in recent versions) for maximum
performance, as BuildKit's DAG allows for smart skipping of uneeded stages.

These changes were proposed in #659

Renames the `prometheus_rules` key to `prometheusRules` to comply to the
spec (https://github.com/monitoring-mixins/docs/blob/master/design.pdf)

BREAKING `prometheus_rules` is not available anymore

When a more or less invalid config (e.g. `null`) is supplied, the log level prop
of the config receives an invalid value which causes the logger to panic on
re-init.

Prevents this by checking for a nil log-level and notifies the user in case

* fluent-plugin: Mark as multi-workers ready

* fluent-plugin: Add info on multi-worker usage to README

Fixes #715

removing entries.go as we are no longer using the custom Histogram for the log_entries_bytes histogram

changing the log length histogram to be a normal histogram with only a `path` label, also removing the label values when no longer tailing the file.  Following the same pattern we used for readBytes and totalBytes.

fix logcli code src path

Remove useless config option

Signed-off-by: Xiang Dai <764524258@qq.com>

Signed-off-by: Xiang Dai <764524258@qq.com>

* Improvements in live tailing of logs

Instead of polling for new logs, a grpc stream is opened between ingester and querier to get live logs
Querier reconnects to disconnected or newly added ingesters

* Added more comments to code for live log tailing

* Some code refactoring in live log tailing

* handling delayfor in logcli, max delayfor to be 5 seconds

* some changes in tail response in live tailing

* Fixed issue with stopping ingesters gracefully when live tailing is being used

* Added tests for tailer in querier

* Live tailing made logql compatible, some code refactoring suggested in PR

Signed-off-by: Xiang Dai <764524258@qq.com>

updating the loki dashboards to use the metrics from the new gateway

Removing the pre-allocation of a buffer when serializing blocks, in most of my empirical testing we were using 9-10k out of the 32k allocated leaving about 2/3 of the buffer allocated and unused times the number of blocks per chunk, times the number of chunks kept in memory.  This was adding up to a fair amount of allocated but unused space.

component/integrations: rename fluent plugin

Signed-off-by: Steven Sheehy <ssheehy@firescope.com>

* fix docker plugin ci

* use binary cache

* adds first version of docker driver

* without logrus and fixes some linter issue

* fix the driver and start a build system

* adds swarm label discovery

* Add documentation and more targets

* make the linter happy 

* indent config.json

* with circleci steps for master and branch

* Review Feebacks

* Helm chart integration testing

Signed-off-by: Steven Sheehy <ssheehy@firescope.com>

* Don't check helm version on every push

Signed-off-by: Steven Sheehy <ssheehy@firescope.com>

* Remove chart upgrade testing

Signed-off-by: Steven Sheehy <ssheehy@firescope.com>