See https://github.com/prometheus/prometheus/issues/16911
This will create a denser layout by default, enabling people to see more
information on the page without having to discover the global settings menu.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Added tests for durationWithUnitRegexp to validate matching of complete durations with units and ensure non-matching cases are correctly identified.
Signed-off-by: ADITYA TIWARI <142050150+ADITYATIWARI342005@users.noreply.github.com>
- No duration unit suggestions shown if a valid unit follows the digit (e.g. , )
- Adds related test cases
Signed-off-by: ADITYA TIWARI <adityatiwari342005@gmail.com>
To reduce main UI clutter, I added a new settings submenu above the chart
itself for the new setting. So far it only has the one new axis setting, but it
could accommodate further settings in the future.
For now I'm only adding a boolean on/off setting to the UI to set the Y axis to
0 or not. However, the underlying stored URL field is already named
y_axis_min={number} and would support other Y axis minima, in case we want to
support custom values in the UI in the future - but then we'd probably also
want to add an axis maximum and possibly other settings.
Fixes https://github.com/prometheus/prometheus/issues/520
Signed-off-by: Julius Volz <julius.volz@gmail.com>
We already have infinite scrolling for all the rules within a group, but this
adds a second (nested) level of infinite scrolling for the firing alerts within
a rule.
This should further improve the rendering speed for the /alerts page for
Prometheus setups with many firing alerts and when you expand the individual
rule panels that contain those firing alerts.
Fixes https://github.com/prometheus/prometheus/issues/17254
Signed-off-by: Julius Volz <julius.volz@gmail.com>
In contrast to Bootstrap, Mantine's Accordion component didn't remove its
panel contents from the DOM when collapsed, so rendering pages with lots of
collapsed Accordion items was way slower and more resource-intensive in the
new Mantine UI. While I talked to Vitaly from Mantine and he managed to add
unmounting of collapsed panel contents in Mantine 9, this will only be
available next year. So for now, I'm forking over the Accordion component
from Mantine and adding a hacky modification to it that removes contents
for collapsed panels. This fork can be removed after upgrading to Mantine 9
sometime in 2026. I removed all the unnecessary test files and so on and
just kept the core Accordion code files.
This should really help with the following issues:
https://github.com/prometheus/prometheus/issues/17254https://github.com/prometheus/prometheus/issues/16830
The /alerts and /rules pages should be the most affected since the panels
on those are collapsed by default. The /targets and /service-discovery
pages have expanded panels by default, but I still swapped out the
Accordion implementation for consistency and in case someone collapses a
bunch of panels.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
I followed the migration guide at https://mantine.dev/guides/7x-to-8x/ to
add back both manual timezone handling in the time input, as well as syntax
highlighting in the config code box.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Ensure params are joined with & and remove trailing dot to restore working links from Alerts/Rules to Graph.
Signed-off-by: Akshat Sinha <akshatsinhasramhardy@gmail.com>
Without `staleTime: Infinity`, the query data would be immediately marked
stale, and in combination with the `gcTime: 0` setting and a brief unmount
during data loading (via Suspense), the data would have to be reloaded a second
time during/after the first real render.
`gcTime: 0` + `staleTime: Infinity` should give us the desired behavior where
data loaded for a page should be valid forever as long as that page is being
displayed, but always thrown away (and later reloaded) whenever we navigate
away and again to that page.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
This adds:
* A `ScrapePoolConfig()` method to the scrape manager that allows getting
the scrape config for a given pool.
* An API endpoint at `/api/v1/targets/relabel_steps` that takes a pool name
and a label set of a target and returns a detailed list of applied
relabeling rules and their output for each step.
* A "show relabeling" link/button for each target on the discovery page
that shows the detailed flow of all relabeling rules (based on the API
response) for that target.
Note that this changes the JSON encoding of the relabeling rule config
struct to output the original snake_case (instead of camelCase) field names,
and before merging, we need to be sure that's ok :) See my comment about
that at https://github.com/prometheus/prometheus/pull/15383#issuecomment-3405591487
Fixes https://github.com/prometheus/prometheus/issues/17283
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Bump versions and changelog.
Co-authored-by: Arthur Silva Sens <arthursens2005@gmail.com>
Co-authored-by: Ayoub Mrini <ayoubmrini424@gmail.com>
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Implements expand/collapse functionality for displaying final scrape
configuration (interval + timeout) in the targets page timing column.
- Add ScrapeDetails component with expand/collapse chevron
- Keep existing "Last Scrape" and "Scrape Duration" badges always visible
- Display "Scrape interval: every \<interval\>" and "Scrape timeout: after \<timeout\>" when expanded
- Use IconRepeat for interval and IconPlugConnectedX for timeout
- Follow TargetLabels.tsx pattern for consistency
- Implement performance optimization with conditional DOM rendering
- Maintain existing hover tooltip functionality
Signed-off-by: ADITYATIWARI342005 <142050150+ADITYATIWARI342005@users.noreply.github.com>
Add a first_over_time function, and corresponding ts_of_first_over_time
function. Both are behind the experimental functions feature flag.
Signed-off-by: Craig Ringer <craig.ringer@enterprisedb.com>
Although these suffixes always need to be removed before querying metadata for
metrics that follow the Prometheus naming best practices, there can also be
metrics that don't follow these naming practices and have these suffixes
without being part of either a histogram or a summary metric.
Fixes https://github.com/prometheus/prometheus/issues/16907
Signed-off-by: Julius Volz <julius.volz@gmail.com>
This RC reverts the feature "OTLP: Support promoting OTel scope attributes".
Add the line back into the CHANGELOG for 3.5.0-rc.0, since we are not changing that version.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* Preserve source files in codemirror-promql package
This allows for sourcemaps to work when the package is imported via ESM-native CDNs such as esm.sh
Signed-off-by: wmTJc9IK0Q <171362836+wmTJc9IK0Q@users.noreply.github.com>
* Preserve source files in lezer-promql package
Signed-off-by: wmTJc9IK0Q <171362836+wmTJc9IK0Q@users.noreply.github.com>
---------
Signed-off-by: wmTJc9IK0Q <171362836+wmTJc9IK0Q@users.noreply.github.com>
This commit adds the ts_of_(max,min,last)_over_time functions behind the experimental feature flag.
Signed-off-by: Michael Hoffmann <mhoffmann@cloudflare.com>
Addresses part of https://github.com/prometheus/prometheus/issues/16515
For now, I'm adding very similar filtering to the /rules page as we have on
the /alerts page, with the difference being:
* The state filter filters by rule health (ok/warn/unknown) instead of
alert state (firing/pending/inactive)
* We don't collect & show detailed stats on the different state counts as
we do on the /alerts page
There is a lot of copied / very similar code between those two pages (and
also some others) around filtering and pagination, so maybe there is an
opportunity for more code sharing in the future here.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
When opening the status pages menu while already viewing one of the
status pages, the whole page would be re-rendered because the menu target's
default action of following the current page's URL was not prevented. Also,
we don't need to use a NavLink component for the menu target when we are
not viewing a status page, because then the component won't need to be
highlighted anyways.
Discovered + fixed with the help of react-scan.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Moving the debouncing of the search field to the parent component and then
memoizing the ScrapePoolsList component prevents a lot of superfluous
re-renders of the entire scrape pools list that previously got triggered
immediately when you typed in the search box or even just collapsed a pool.
(While the computation of what data to show was already memoized in the
ScrapePoolList component, the component itself still had to re-render a lot
with the same data.)
Discovered this problem + verified fix using react-scan.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Updated/fixed/changed a lot of content, changed wordings and style, etc.
Things should now be in an order that most users / developers expect, with the
most relevant sections (how to build + run) at the top.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
In stacked / flex layouts like on the service discovery page, elements get a
100% width unless you tell them to only be as wide as their contents.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
Technically, we are not filtering by the state of a single rule, but by the
whole rule group state (if even a single alert is firing within a group,
filtering for "Pending" won't show the group, even if it has pending
alerts).
Signed-off-by: Julius Volz <julius.volz@gmail.com>
As discussed in https://github.com/prometheus/prometheus/issues/15717, this
should help show more information on one screen, to be more similar to the
old UI in this regard.
Signed-off-by: Julius Volz <julius.volz@gmail.com>
