Thank you @Renee Cooper, @Jimmy PANG and everyone for your kind words and empathy. It's a really difficult time here, but we are doing everything we can to keep ourselves sane, and that includes staying available for our community. Please continue to reach out in #C02CTC89LAX or #C03LPJ5EYU9 if you need anything and we will do our best to assist 🙏

Hi Experts, I have a models have multiple product_names, we want to apply anomaly detection separately on each product name. Is this possible in elementary anomaly detection Instead of adding for each product , is there why which can do group by product

models:
  - name: metric_abc
    config:
      elementary:
        timestamp_column: 'date'
    columns:
      - name: product_value
        tests:
          - elementary.column_anomalies:
              column_anomalies:
                - sum
              time_bucket:
                period: month
                count: 1
              days_back: 365
              config:
                severity: ERROR
              anomaly_sensitivity: 1
              where_expression: "product_name = 'ABC_1'"

          - elementary.column_anomalies:
              column_anomalies:
                - sum
              time_bucket:
                period: month
                count: 1
              days_back: 365
              config:
                # optional - change severity
                severity: ERROR
              anomaly_sensitivity: 1
              where_expression: "product_name = 'ABC_2'"

This morning a model has failed and we received the same alert for more than -literally- 1000 times. Does anyone know what is going on?

Hello Experts, New user here. Is there a config to turn off capturing the detailed failure rows? We are fine with just capturing the metrics of how many rows failed a given test and don't need row level failures to be stored.

Hi! I'm wondering whether to use all column anomalies vs. column anomalies for my tables. What is an example use-case for all column anomalies? It seems like a catch-all but I'm wondering in what cases people have found it useful

Don't know if this was just a stealth update (don't see anything in #C02BPB78J4T), but appreciate this change in how push notifications are appearing from Elementary. The Header is much more descriptive, so it's a lot easier to see at a glance what the alert is about. Attaching an alert from yesterday vs 1 I got today for context. (had I given a name to this test over the default, I presume this would have appeared in the quotes as well which is super neat)

Hello folks, Is there a config to not include sources in the monitored tables Visual on the elementary report? We want to see how many dbt models are truly unmonitored

Hello folks, One more question, is there a config to disable or enable elementary executions based the dbt target. We were thinking to run elementary models in our production target but not in dev. Is there any config to do so? Thanks, Kiran

To the first question, the new --filter tag may or may not help - see the announcement in #C02BPB78J4T To your second question, sure:

Hello, newbie here. It's looking good after first try out, nice job 🙂 Question: I have in Databricks already quite some runs from which I can download artefacts with following structure (attached). I guess it's not possible to use this as an input to retroactively build history right?

Hi folks, hope you're all well! I've been trying to get up to speed with everything Elementary, and took a first pass at going through the quickstart docs. Just wanted to check if there's any other material / loom / course that can sort of help me with understanding, setting up, and using elementary in my prod workflows? TIA!

Hi Team, Our team wanted to do a run in order of elementary on-run-end and than custom macro run Is there a way to do this? Also can we assign a separate warehouse for elementary runs in dbt cloud? Thank you

Hi Team, I wanted to put the latest version of elementary_data in my requirements.txt file but it has conflicts with others version of librairies langchain, pandas ... When I want to solve the problem only the first version 0.4 appears

Hello @Elon Gliksberg. Is there a way to query data from the incident feature (beta) in elemetnary-cloud?

Hi, we are getting following error when elementary makes query like `%insert into `dbt_tests__tmp_20240823213130188816``

(metadata_hash) values'

Resources exceeded during query execution: Not enough resources for query planning - too many subqueries or query is too complex.

I am trying to understand what does the query really do? Why is it adding only single column values to the tmp table

Minor suggestion, but when a model run fails, we probably don’t want to highlight the 100% reduction in execution time, in green 😂

Quick question: Is column-level lineage only available in elementary cloud?

Hi, Could someone help me with the below error:

2024-09-12 06:52:50 — INFO — Running with edr=0.8.2
[31;20m2024-09-12 06:52:55 — ERROR — Unable to get the latest invocation: Failed to run dbt command.[0m
[31;20m2024-09-12 06:52:57 — ERROR — Failed to parse Elementary's database and schema.[0m
2024-09-12 06:52:57 — INFO — Running dbt --log-format json run-operation elementary.log_macro_results --args {"macro_name": "get_test_results", "macro_args": {"days_back": 7, "invocations_per_test": 720, "disable_passed_test_metrics": false}} --project-dir /usr/local/lib/python3.9/site-packages/elementary/monitor/dbt_project --profiles-dir /home/dbt/elementary_docs/profile/
[31;20m2024-09-12 06:53:00 — ERROR — Could not generate the report - Error: Failed to run dbt command.
Please reach out to our community for help with this issue.[0m
NoneType: None

I've all the necessary access to db and schema access to the elementary profile. This was running fine for a long. Suddenly past 1 week I'm facing this error

Did you notice your "Community" link is broken on https://www.elementary-data.com/ ?

Hi, made PR with initial support of MS Fabric (dbt-fabric)! Would love some feedback and hopefully approval😁 😅 For us, its already running over month in production logging run results😉 https://github.com/elementary-data/dbt-data-reliability/pull/777

Is there possibility to identify all the upstream models used by the model , i know i can check in dbt docs but something i can query

I am using Airflow to run DBT which has elementary package , I use elementary table to list out all dbt model and their execution time , but if the airflow dag is running for long tim e , is it possible find out which dbt model is running currently in Elementary tables ?

Hi folks, Has anyone been able to successfully load dbt source freshness results into the warehouse? https://docs.elementary-data.com/cloud/guides/collect-source-freshness#dbt-core-users It's parsing everything from run_results.json, but not the sources.json. Seems like there's not enough documentation and I can't find anything in other communities

hello! do we have a read on when the Hex integration might be released?

Morning everyone. I wanted to ask if anyone has an idea where I could get live data sources that are updated daily. Perhaps an API or two? I'm working on a project that has to do with modeling with incremental models in SQLMesh. Working with live data would give me a better sense of how SQLMesh incremental models work, which I've read is really good & robust. BQ live datasets are out of the question for me

Hello! Random question. Has anyone here used the column anomalies feature to track changes in trends? For example, something that points out inflection points within a certain timeframe in the past (between 7-15 days as a concrete example). It seems like the current anomaly detection is good for unexpected spikes or drops, but not as much for a gradual decline where historically it has been flat or up (or vice versa!).

Hi I was wondering if the zoom call for larger teams would also be a guided walkthrough on how to set up elementary for our systems?

Hey everyone, I recently built dbt Flow Lineage, a VS Code extension for visualizing column-level lineage in dbt core projects. The idea came from a simple problem: dbt core shows model lineage well, but column-level debugging still means manually reading SQL across files. This extension lets you: • click a column and trace where it came from • see renames, transforms, and aggregates visually • isolate lineage for one model • export diagrams as PNG It supports: •

SELECT *

• CTEs • Jinja Works in: • Visual Studio Code • Cursor • Windsurf Try it here: https://marketplace.visualstudio.com/items?itemName=tripleaceme.dbt-flow-lineage Would love feedback from anyone working with messy real-world dbt SQL, especially edge cases that should be handled better

I am trying to use volume_threshold with the latest elementary version (0.23.0), but getting the error "'dict object' has no attribute 'test_volume_threshold'". Is volume_threshold released and available in OSS?

Does volume anomaly check in OSS always read the historical volume from the source table or is there a way to read the history from data_monitoring_metrics?