Commit graph

4808 commits

Author SHA1 Message Date
Aliaksandr Valialkin
bbfa52bd75
docs: follow-up after 68e56b6fc5 2022-08-17 21:27:24 +03:00
Aliaksandr Valialkin
2de9d42dcc
go.mod: update github.com/VictoriaMetrics/metrics from v1.22.1 to v1.22.2 2022-08-17 21:27:24 +03:00
Roman Khavronenko
cfcb5ab15b
vmalert: set alert's source link to UI instead of JSON source (#2986)
We switch default alert's source link to redirect user
to vmalert's UI instead of previous JSON object. While it breaks
compatibility, it also supposed to improve user's experience.
The old behavior can be achieved by updating `-external.alert.source`
command-line flag.

Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-17 21:27:24 +03:00
Aliaksandr Valialkin
d62e9bc964
docs/sd_configs.md: clarify a bit honor_labels config option 2022-08-17 14:24:53 +03:00
Aliaksandr Valialkin
87e0d69bf4
app/vmselect/netstorage: fix a bug introduced in 1a254ea20c
The bug results in `duplicate output time series` error
because the same time series is added two times into the orderedMetricNames list
inside the tmpBlocksFileWrapper.Finalize().

While at it, properly release all the tmpBlocksFile structs on tbf.Finalize() error.
Previously only the remaining tbf entries were released. This could result in resource leak.
2022-08-17 14:07:51 +03:00
Aliaksandr Valialkin
1812d33a2d
lib/promscrape: automatically generate additional per-target labels for targets with non-zero series limit
The following metrics are generated:

- scrape_series_limit
- scrape_series_current
- scrape_series_limit_samples_dropped

These metrics simplify alerting on targets, which expose too many time series

See https://docs.victoriametrics.com/vmagent.html#automatically-generated-metrics
and https://docs.victoriametrics.com/vmagent.html#cardinality-limiter for more details
2022-08-17 13:22:02 +03:00
Roman Khavronenko
0261118765
docs: add new article link (#2989)
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-17 13:22:02 +03:00
Aliaksandr Valialkin
e5c61cac22
docs/CHANGELOG.md: group vmalert features closer to each other 2022-08-17 11:49:53 +03:00
Aliaksandr Valialkin
aa37e6b438
lib/promscrape: retry http requests if the server returns 429 status code
The 429 status code means that the server is overwhelmed with requests.
The client can retry the request after some wait time.
Implement this strategy for service discovery and scrape requests.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2940
2022-08-16 14:57:26 +03:00
Aliaksandr Valialkin
1a363192ff
lib/storage: typo fix in comments after f830edc0bc 2022-08-16 13:45:32 +03:00
Aliaksandr Valialkin
dc929e0d16
lib/storage: improve performance for /api/v1/labels and /api/v1/label/.../values endpoints when match[] filter matches small number of time series
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2978
2022-08-16 13:34:23 +03:00
Aliaksandr Valialkin
d39915ff8e
docs/MetricsQL.md: add the list of supported ... functions lines just before the corresponding lists
This improves the readability a bit
2022-08-16 12:10:09 +03:00
Roman Khavronenko
436b4d90af
docs: update vmalert docs (#2987)
* mention recently added `$alertID` and `$groupID` variables in the changelog
* properly escape template examples in the vmalert's README

Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-16 12:10:08 +03:00
Roman Khavronenko
0d5c403b6e
vmalert: support $alertID and $groupID in template variables (#2983)
Support of these two variables allows building custom URLs with
alert's ID and group ID params.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/517#issuecomment-1207141432

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-16 11:24:01 +03:00
Artem Navoiev
5ae251d074
fix typo in influx guide (#2984) 2022-08-16 11:23:59 +03:00
Aliaksandr Valialkin
7d7cf2b6fd
app/vmselect: follow-up after 63e0f16062
* Explicitly store a pointer to UserReadableError in the error interface.
  Previously Go automatically converted the value to a pointer before storing in the error interface.

* Add Unwrap() method to UserReadableError, so it can be used transparently with the other code,
  which calls errors.Is() and errors.As().

* Document the change in docs/CHANGELOG.md
2022-08-15 13:53:19 +03:00
Roman Khavronenko
8a26ec435d
vmselect: introduce UserReadableError type of error (#2894)
When read query fails, VM returns rich error message with
all the details. While these details might be useful
for debugging specific cases, they're usually too verbose
for users.
Introducing a new error type `UserReadableError` is supposed
to allow to return to user only the most important parts
of the error trace. This supposed to improve error readability
in web interfaces such as VMUI or Grafana.

The full error trace is still logged with the full context
and can be found in vmselect logs.

Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-15 13:53:18 +03:00
Roman Khavronenko
79e396dac9
docs: mention default vmalert's behavior change in Update notes (#2981)
Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-15 13:31:59 +03:00
Yury Molodov
dc52b283a3
vmui: shortcut keys legend (#2971)
* feat: add shortcut modal

* feat: add shortcut descriptions

* app/vmselect/vmui: `make vmui-update`

* docs/CHANGELOG.md: document the change

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2022-08-15 12:04:30 +03:00
Yurii Kravets
d7dc7c81e7
doc: add upgrade/downgrade without downtime to FAQ (#2973)
* doc: add upgrade/downgrade without downtime to FAQ

Added info on how to upgrade or downgrade VictoriaMetrics without downtime to FAQ
Based on reply https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2944#issuecomment-1207502038

* Udate pr

Fixing format
2022-08-15 11:31:00 +03:00
Aliaksandr Valialkin
c27bd63f6c
lib/promscrape: update links to sd_configs from Prometheus site to https://docs.victoriametrics.com/sd_configs.html 2022-08-15 01:40:48 +03:00
Aliaksandr Valialkin
1a00c9ef03
lib/promscrape/discovery/kubernetes: add __meta_kubernetes_pod_container_image label in the same way as Prometheus 2.38 does
See https://github.com/prometheus/prometheus/pull/11034
2022-08-15 01:18:57 +03:00
Aliaksandr Valialkin
2fb63dda83
lib/promscrape/discovery/kubernetes: add __meta_kubernetes_service_port_number label to role: service in the same way as Prometheus 2.38 does
See https://github.com/prometheus/prometheus/pull/11002
2022-08-15 01:07:19 +03:00
Aliaksandr Valialkin
40c2fae617
vendor: make vendor-update 2022-08-15 00:54:50 +03:00
Aliaksandr Valialkin
198d8eeeaa
app/vmalert/templates: add toTime() template function in the same way as Prometheus 2.38 does
See https://github.com/prometheus/prometheus/pull/10993
2022-08-15 00:49:52 +03:00
Aliaksandr Valialkin
2b58bd9876
lib/promscrape/discovery/dns: add support for resolving MX records
See https://github.com/prometheus/prometheus/pull/10099
2022-08-15 00:33:06 +03:00
Aliaksandr Valialkin
10402459d8
lib/vmselectapi: do not log connection accept/close from vmselect
These log messages became too spammy in production clusters
after the commit 190c8b463c ,
which closes idle connections from vmselect to vmstorage.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2508
2022-08-12 09:15:29 +03:00
Aliaksandr Valialkin
1b39be3305
lib/vmselectapi: add rpc call prefix to the trace of the rpc call in order to make it more clear 2022-08-12 00:20:49 +03:00
Aliaksandr Valialkin
78f584fb0b
docs/CHANGELOG.md: clarify the change at 28441711e6 2022-08-11 23:53:32 +03:00
Roman Khavronenko
36660bcfe2
vmalert: follow-up after 28441711e6 (#2972)
Signed-off-by: hagen1778 <roman@victoriametrics.com>

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-11 23:52:18 +03:00
Matthew Blewitt
b1b8c99b17
vmalert: mark some url flags as sensitive (#2965)
Other components, such as `vmagent`, mark these flags as sensitive and
hide them from the `/metrics` endpoint by default. This commit adds
similar handling to the `vmalert` component, hiding them by default, to
prevent logging of secrets inappropriately.

Showing of these values is controlled by an additional flag.

Follow up to https://github.com/VictoriaMetrics/VictoriaMetrics/pull/2947
2022-08-11 23:51:43 +03:00
Aliaksandr Valialkin
1a254ea20c
app/vmselect/netstorage: remove common contention points related to inter-CPU communcations
This should improve vmselect performance scalability on systems with many CPU cores.

The following tasks were done:

- Use separate temporary files for storing the data read from each vmstorage node.
  This may result in the following potential issues:
  - Up to N times higher memory usage for performing each query where N is the number
    of vmstorage nodes known to vmselect.
    This issue shouldn't increase chances of out of memory errors in most cases,
    since per-query memory overhead is quite low comparing to the overall vmselect memory usage.
  - Up to N times higher number of open temporary files where N is the number
    of vmstorage nodes known to vmselect.
    This issue should be fixed by increasing the limit on the number of open files.

- Use separate counters per each vmstorage node for various stats calculation
  when reading the data from vmstorage nodes.
2022-08-11 23:22:56 +03:00
Aliaksandr Valialkin
ec3df0b913
app/vmselect/netstorage: improve scalability of blocks processing on systems with multiple CPU cores
Previously a single syncwg.WaitGroup was used for tracking the lifetime of processBlock callbacks
across all the per-vmstorage goroutines. This could be slow on systems with many CPU cores
because of inter-CPU synchronization overhead.

Use a separate per-vmstorage sync.WaitGroup instead in order to reduce inter-CPU synchronization overhead.
This should imrpove performance for heavy queries over big number of blocks on multi-CPU systems.
2022-08-11 21:37:24 +03:00
Aliaksandr Valialkin
690b505975
deployment/docker: specify docker image tags for all the docker images for reproducible docker-compose up runs 2022-08-09 12:30:22 +03:00
Roman Khavronenko
f42853275f
lib/storage: prevent excessive loops when storage is in RO (#2962)
* lib/storage: prevent excessive loops when storage is in RO

Returning nil error when storage is in RO mode results
into excessive loops and function calls which could
result into CPU exhaustion. Returning an err instead
will trigger delays in the for loop and save some resources.

Signed-off-by: hagen1778 <roman@victoriametrics.com>

* document the change

Co-authored-by: Aliaksandr Valialkin <valyala@victoriametrics.com>
2022-08-09 12:17:47 +03:00
Roman Khavronenko
f90f654cf2
vmalert: sort groups at /alerts page (#2968)
Sorting will produce deterministic output
of grops on the page.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-09 12:08:54 +03:00
Aliaksandr Valialkin
91d0b45537
docs/CHANGELOG.md: cut v1.80.0 2022-08-08 19:59:02 +03:00
Aliaksandr Valialkin
7e892d3322
docs/CHANGELOG.md: add changes for v1.79.2 2022-08-08 18:00:38 +03:00
Aliaksandr Valialkin
e0865f4174
vendor: update github.com/VictoriaMetrics/metrics from v1.21.0 to v1.22.1 2022-08-08 17:19:16 +03:00
Aliaksandr Valialkin
b8877da35b
docs/CHANGELOG.md: link to the issue related to vmselect panic in multi-level cluster setup
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2961
2022-08-08 15:09:39 +03:00
Aliaksandr Valialkin
ceb158c5bb
docs/vmagent.md: typo fix in __meta_kubernetes_annotation_prometheus_io_tenant label name
It must be __meta_kubernetes_pod_annotation_prometheus_io_tenant
2022-08-08 14:53:28 +03:00
Aliaksandr Valialkin
310779d8b5
lib/promscrape: follow-up after 2c553d5a2f
- fix broken tests
- cosmetic code cleanup
- document the change at https://docs.victoriametrics.com/vmagent.html#multitenancy
- document the change at https://docs.victoriametrics.com/CHANGELOG.html
2022-08-08 14:49:16 +03:00
Fury
59fdb4cb72
add support to scrape multi tenant metrics (#2950)
* add support to scrape multi tenant metrics

* add support to scrape multi tenant metrics

Co-authored-by: 赵福玉 <zhaofuyu@zhaofuyudeMac-mini.local>
2022-08-08 14:49:15 +03:00
Aliaksandr Valialkin
69d62d5736
docs: sync -help output with the latest changes 2022-08-08 14:05:36 +03:00
Roman Khavronenko
f31132b70b
lib/promrelabel: fix expected test result (#2957)
follow-up after 68c4ec9472

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-08 13:48:14 +03:00
Aliaksandr Valialkin
221dd3a224
all: bump the minimum supported version of Go from 1.17 to 1.18
This is needed because some dependencies uses generics, which have been appeared in Go1.18

This is a follow-up for caf3dd4fa2
2022-08-08 13:45:39 +03:00
Roman Khavronenko
6af40f6275
vmalert: remove notions of vmalert being compatible with VM only (#2954)
vmalert can be successfully used with datasources
compatible with Prometheus HTTP API. So we remove comments or
notes in Readme which are saying opposite.

Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-08 13:44:15 +03:00
Aliaksandr Valialkin
1996e36cf0
app/vmselect/netstorage: prevent from calling processBlocks callback after the exit from ProcessBlocks function
This should prevent from panic at multi-level vmselect
when the top-level vmselect is configured with -replicationFactor > 1
2022-08-08 13:32:44 +03:00
Roman Khavronenko
3c583c16a1
dashboards: add Cache usage % panel to Caches row (#2960)
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2941
Signed-off-by: hagen1778 <roman@victoriametrics.com>
2022-08-08 11:45:17 +02:00
Aliaksandr Valialkin
12f4c89ff7
docs/CHANGELOG.md: link to the issue with improper handling of enpoint option at ec2_sd_configs
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/2917
2022-08-08 03:33:17 +03:00