#LogQL
LightNews LightNews
itisamystery.com
love love love never having to deal with grafana charts and dashboards ever again. promql and logql is the worst query language I've ever used.
September 19, 2026 at 6:24 PM
zuzur.eurosky.social
c'est fait avec l'api openrouter ? c'est sympa, j'ai fait un dashboard un peu comme ça en extrayant les infos de access logs du logql
September 22, 2026 at 9:23 AM
hikikomorphism.bsky.social
An implication: if your problem is 'shaped like' alerting off of logs it actually makes sense to provide a logQL interface even if it's, like, a user's habitica habit checkins. LLMs know logQL already. They don't know your API. Give them what they know!
January 11, 2026 at 10:13 PM
victoriametrics.bsky.social
Working with #logs in #Loki or #VictoriaLogs? 🧵
#LogQL is not the same as #LogsQL — both are optimized for querying, but with key differences.
Here’s how to convert LogQL queries to LogsQL
docs.victoriametrics.com/victorialogs...

#DevOps #Observability #LogManagement #Logs
May 8, 2025 at 1:56 PM
jaz.sh
Okay this is kinda neat, I can filter to see the post rates of anyone by handle now. I wanted to get a list of the top N (50?) posters over the time range but Grafana said No! (maximum of series (500) reached for a single query). LogQL is cool but has its limits for high cardinality queries...
May 18, 2023 at 1:44 AM
ineumann.fr
New version of qwctl 1.2.0 : comply with LogQL syntax (Loki syntax, inspired by PromQL syntax for logs) www.cwcloud.tech/docs/tutoria...
June 12, 2026 at 1:29 PM
rawkode.academy
🎬 Resharing this deep dive into Loki! 🚀 Hosted by @rawkode.dev, this video is perfect for anyone looking to get hands-on with Grafana's log aggregation system. 📊 Don't miss out on learning about LogQL, dashboards, and more! #Loki #Grafana
Hands-on Introduction to Loki
Watch on the Rawkode Academy
rawkode.academy
October 6, 2025 at 2:00 AM
cwcloud.tech
📣 Quickwit CLI "qwctl" is now supporting SQL syntax (alongside with PromQL, LogQL and Lucene) and it improve way better our agentic AI experience for observability!

➡️ Documentation: www.cwcloud.tech/docs/tutoria...
July 15, 2026 at 10:49 AM
akhen.bsky.social
SELECT count(*) FROM read_json_auto('log.jsonseq', format = 'newline_delimited') where duration > 10000;

This simple example could be done in logql but often going to DuckDB is easier.

What's your workflow?
December 3, 2024 at 5:33 PM
qxip.bsky.social
#qryn supports HEC ingestion (through OTEL) and offers LogQL over it at this time- SPL remained a prototype plugin we never completed or released as there was little interest.
December 16, 2024 at 9:29 PM
opsmatters.com
The latest update for #Last9 includes "Trace Go Apps Using Runtime Tracing and #OpenTelemetry" and "Query and Analyze Logs Visually, Without Writing LogQL".

#Monitoring #Observability #TimeSeries #SRE #DevOps https://opsmtrs.com/41EhXHh
Last9
Last9 provides tools to improve Reliability in large-scale cloud-native environments.
opsmtrs.com
July 17, 2025 at 10:32 PM
larvitz.burningboard.net.ap.brid.gy
Posted some new blog-articles during the weekend .. Now I saw a quite substancial spike in traffic, that doesn't really look like normal human interaction ...

Grafana/Loki with some LogQL did quickly reveal it. It's scrapers of the AI-Slop generators […]

[Original post on burningboard.net]
January 19, 2026 at 6:54 PM
qxip.bsky.social
We don't need it - qryn is a designed as a transpiler from/to LogQL, PromQL, TraceQL and Pyroscope to SQL (ClickHouse, DuckDB, Influx/Datafusion) so it doesn't need any features from the backend which makes it easy to port to "the next" cool thing, always
January 28, 2025 at 3:37 PM
grafana.bsky.social
ICYMI: We open sourced o11y-bench: 63 tasks for evaluating AI agents on real observability workflows — PromQL, LogQL, TraceQL, incident investigation, dashboard editing.
Introducing o11y-bench: an open benchmark for AI agents running observability workflows | Grafana Labs
To help the Grafana community navigate this new world of AI-assisted observability, we’re open sourcing grafana/o11y-bench, a benchmark for evaluating AI agents on observability workflows.
grafana.com
May 12, 2026 at 3:00 PM
deployharmlessly.bsky.social
Ever notice how servers are a bit like dice in a cosmic casino? 🎲 Grafana Loki rolls the logs into searchable stories. 🌌 Master log querying and explore data wonders! #Grafana #Loki #Logging #DevOps ✨🔍
Grafana Loki Fundamentals and Architecture
Grafana Loki is a scalable, cost-efficient log system for cloud-native environments. It uses minimal indexing, compressed storage, and LogQL for fast queries.
dzone.com
March 9, 2025 at 5:52 PM
devtocash.bsky.social
Letting an AI agent search Loki? Never return raw log lines.
500 lines → ~12 collapsed patterns, ~2% of the tokens. Reject bare {} selectors. Keep the delete API unroutable.
https://devtocash.com/blog/2026-07-29-loki-mcp-server-safe-log-search-agent
#DevOps #SRE
Build a Loki MCP Server: Safe Log Search for AI Agents
Build a Loki MCP server for safe read-only log search by AI agents — LogQL guardrails, line caps, pattern collapsing, and prompt-injection defenses that hold.
devtocash.com
July 29, 2026 at 1:14 AM
8-p.info
Kubernetes 関係の長い YAML とか Claude に書かせているのは前からなのですが、最近 Grafana 関係の PromQL と LogQL も Claude に書かせるようになりました。文法とどういうメトリクスがあるのかの両方を知らないといけないので、AI むき (というか人間にむかない) と思う。
github.com/grafana/mcp-...
GitHub - grafana/mcp-grafana: MCP server for Grafana
MCP server for Grafana. Contribute to grafana/mcp-grafana development by creating an account on GitHub.
github.com
May 9, 2026 at 5:30 AM
palark.com
Promruval validates #Prometheus, Thanos, Mimir, and Loki rules against your actual setup and requirements (allowed values, limits, etc.). It comes with a huge list of available validators, which cover rule groups, labels, PromQL & LogQL expressions, and more. GitHub repo: github.com/FUSAKLA/prom...
GitHub - FUSAKLA/promruval: Validate Prometheus/Thanos/Mimir/Loki rules metadata and expression properties to match requirements and constrains of your setup.
Validate Prometheus/Thanos/Mimir/Loki rules metadata and expression properties to match requirements and constrains of your setup. - FUSAKLA/promruval
github.com
February 14, 2025 at 9:49 AM
jeremymorrell.dev
This post touches on a lot of the things that I suspected work well with agentic workflows, but haven't had a chance to try out.

I keep thinking back to the infra that Facebook had a decade+ ago that let 100s of people work in the same giant codebase. We all need that now openai.com/index/harnes...
We did the same for observability tooling. Logs, metrics, and traces are exposed to Codex via a local observability stack that’s ephemeral for any given worktree. Codex works on a fully isolated version of that app—including its logs and metrics, which get torn down once that task is complete. Agents can query logs with LogQL and metrics with PromQL. With this context available, prompts like “ensure service startup completes in under 800ms” or “no span in these four critical user journeys exceeds two seconds” become tractable. Increasing levels of autonomy As more of the development loop was encoded directly into the system—testing, validation, review, feedback handling, and recovery—the repository recently crossed a meaningful threshold where Codex can end-to-end drive a new feature. Given a single prompt, the agent can now: Validate the current state of the codebase Reproduce a reported bug Record a video demonstrating the failure Implement a fix Validate the fix by driving the application Record a second video demonstrating the resolution Open a pull request Respond to agent and human feedback Detect and remediate build failures Escalate to a human only when judgment is required Merge the change This behavior depends heavily on the specific structure and tooling of this repository and should not be assumed to generalize without similar investment—at least, not yet. We learned that we needed to push more and more context into the repo over time. That Slack discussion that aligned the team on an architectural pattern? If it isn’t discoverable to the agent, it’s illegible in the same way it would be unknown to a new hire joining three months later. The diagram below shows the rule: within each business domain (e.g. App Settings), code can only depend “forward” through a fixed set of layers (Types → Config → Repo → Service → Runtime → UI). Cross-cutting concerns (auth, connectors, telemetry, feature flags) enter through a single explicit interface: Providers. Anything else is disallowed and enforced mechanically. This is the kind of architecture you usually postpone until you have hundreds of engineers. With coding agents, it’s an early prerequisite: the constraints are what allows speed without decay or architectural drift.
February 13, 2026 at 9:38 PM
hugovalters.bsky.social
Slow LogQL range queries? Tune Loki's query-frontend to split requests across time boundaries and cache results in Memcached. Cuts load on queriers, speeds dashboards. Config for K8s/Loki 2.9+: https://www.valtersit.com/vault/optimize-loki-query-performance-using-queryfrontend-ca
September 9, 2026 at 6:00 AM
opscanopy.bsky.social
LogQL vs PromQL: the same query in both languages

LogQL borrows PromQL’s shape but starts from log lines, not metrics. Here is how the two query languages line up, where they translate cleanly, and where they simply don’t.
LogQL vs PromQL: the same query in both languages
LogQL borrows PromQL’s shape but starts from log lines, not metrics. Here is how the two query languages line up, where they translate cleanly, and where they simply don’t.
opscanopy.com
August 25, 2026 at 4:06 PM
hugovalters.bsky.social
Turn Loki log lines into numeric fields in Grafana 10.x. Use a LogQL parser to extract `duration`, then visualize it directly in Explore. No PromQL needed—just derived fields for instant metrics.

https://www.valtersit.com/vault/use-grafana-to-visualize-loki-logs-with-derived-fields-9fbd49/
August 5, 2026 at 7:21 PM
devtocash.bsky.social
Half of an agent's failed LogQL is guessing label names - querying {app="checkout"} when your convention is {service="checkout"}.

Expose a cheap label-discovery call and it stops hallucinating selectors. Grounding beats prompting.
August 2, 2026 at 2:56 PM
devtocash.bsky.social
So the SERVER owns the limits, never the agent.

The agent passes a LogQL query and a lookback in seconds. The server validates the shape, computes the time range, and caps the line count.

Same rule as the Prometheus MCP: expensive knobs are computed, not caller-controlled.
August 2, 2026 at 2:56 PM
devtocash.bsky.social
Three tools: discover labels, guarded search, error overview. Nothing else.

Read-only, bounded, and it can't open every stream in your tenant.

Full build:

https://devtocash.com/blog/2026-07-29-loki-mcp-server-safe-log-search-agent
Build a Loki MCP Server: Safe Log Search for AI Agents
Build a Loki MCP server for safe read-only log search by AI agents — LogQL guardrails, line caps, pattern collapsing, and prompt-injection defenses that hold.
devtocash.com
August 2, 2026 at 2:56 PM