Looker Connect Training

Help Center

Documentation

Community

Cloud Certifications

The Kitchen Table

Explore the product knowledge we've built up together

2114 Topics

sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How to create a TopoJSON file for use in a Custom Map LayerKnowledge Drop

Last tested: June 2021This Help Center article explains how to make a topojson file to use in a map layer in Looker.Remember: The property_key field is CASE SENSITIVE! A custom map layer may be set up and referenced correctly, but the map may display blank if property_key is case mismatched.map_layer: my_neighborhood_layer { file: "neighborhoods.topojson" property_key: "neighborhood"} This content is subject to limited support.                

23765
Dawid
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How do I make my dashboard timezone listen to the timezone defined in my embed URL?Knowledge Drop

Last tested: Nov 2020   Enable User-Specific Timezones in the Admin panel. Add the user_timezone as one of the embed URL parameters when generating the embed URL. After the embed URL loads the dashboard, the embed user can switch to Viewer Timezone in the timezone dropdown menu at the top right. But what if i want to have the embed dashboard loaded with the user-specific timezone by default?Steps 1 and 2 stay the same. Instead of having the embed user switch the timezone in Step 3, use one of the following options.Option 1: Save the dashboard with the default timezone "Viewer Timezone". This will use the user's timezone as default for both embed users and non-embed users.Option 2: We can add &query_timezone=user_timezone as part of the dashboard URL. This will make the user's timezone selected by default even if it is not the saved default timezone for the dashboard. Note that the user_timezone is one of the embed URL parameters, but the query_timezone is not an embed URL param

4070
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How do I filter "on or before" a date?Knowledge Drop

Last tested: June 2021  Right now, filter syntax only allows for "before" a date. This means that in the example before 6/26/2017, the date 6/26/2017 will not be included.There is a feature request for this inclusive functionality. Right now, you either need to remember to add one day to the date you enter, or use two filters, before 6/26/2017 and on 6/26/2017. This content is subject to limited support.                

1910
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Why is there no Password Reset/Edit button for a user?Knowledge Drop

Last tested: April 2021No Password ResetMost likely, this user doesn't have a Looker email credential. Without an email credential, they can't have a password.This happens if the user was set up via Google Auth, SAML, LDAP, etc. This means that any management of passwords will have to be done in those external providers, not in Looker.If you need to add email credentials to users who have only external authentication credentials, you could follow the steps in Enabling the Alternate Login OptionNo Edit ButtonThe Edit button will be missing for users who have had their credentials deleted (happens with rogue API use, for example)In order to get that button to reappear, you will need to reassign credentials. This content is subject to limited support.                

2240
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Why can't I see my Slack channel in the Looker sending modal?Knowledge Drop

Last tested: April 2021  When you don't see the desired channel in the Looker sending modal:a) If you're using the Lookerbot, you have to invite it to your channel first (On your channel, "Add people to #channel_name", find the Lookerbot the same way you find individual users, and select), orb) if you're using a legacy token, you should see the Looker app under "Apps" on Slack and will have to invite the app to your channel ("Add Apps" on your channel and find the Looker app),then you should see that channel in the modal upon refresh.When Lookerbot isn't showing up in your Slack channel:Step a) from above.Further reading: Using Lookerbot for Slack This content is subject to limited support.                

500
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Does revoking a user on the idP side affect the user in Looker?Knowledge Drop

Last tested: June 2021No. Looker doesn't automatically keep track of whether the user has been revoked on the SAML/LDAP/OIDC/Google side. There is a request to Force all users to log out or Automatically recognize new SAML groups if mapping gets updated that you can add your vote through our feedback system.So the user will just stay logged in??Yes. Any users logged in will stay logged in for the remainder of their session (up to 30 days). Of course, once they log out of Looker or otherwise end their Looker session they won't be able to log back in. They will not be deleted or disabled by the Looker system.Can I force the user to log out?You can disable the user via the UI, which logs them out.A robust workaround would be an API script that: Watches the IdP for when a user is revoked. Finds the Looker user corresponding to the IdP user, and their current session ID(s). Kill sessions for that user. Disables that user. This content is subject to limited support.                

2260
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Can LookML dashboards be used in SSO embed?Knowledge Drop

Last tested: June 2021Yes!Nothing really changes except that the dashboard ID is model_name::dashboard_name, and the embed parameters have to be defined via LookML. This doc explains the embedded parameters that you would usually define in "Embed settings" of a UDD.As always, make sure that the user has the see_lookml_dashboards permission.However, SSO embed users are not able to access the LookML dashboards "folder", even with the embed_browse_spaces permissions. This permission only grants access to the Shared and Personal folders, as well as an external group folder (if using external_group_id). Here's a visualization:If we want LookML dashboards to be accessed this way, we'll have to convert these to UDD, so they can be accessed from the personal or Shared folder. This content is subject to limited support.                

1480
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How long does a deleted PDT live on the scratch schema for?Knowledge Drop

Last tested: June 2021The PDT reaper process runs at most every hour (while respecting the PDT maintenance cron string).It will reap all persist_for PDTs whose time has expired, and any sql_trigger_value PDTs for which a new table has already been built.The persist_for clock begins when the PDT is first used in a query. This content is subject to limited support.                

620
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

What is the difference between the query id and history id in system activity?Knowledge Drop

Last tested: June 2021The query ID is tied to a particular query. It's uniquely identified by a combination of SQL text, connection, and possibly other fields. It is created the first time a SQL query is run on a connection. Query IDs are not changeable or destructible.The history ID is tied to a running of a query. So, every time a query is run, a history ID is created. A given query ID could be associated with several history IDs: one for each time it's run. History IDs are also not changeable or destructible.This content is subject to limited support.                

7330
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Can I extend Looker's default proxy timeout of 60 mins?Knowledge Drop

Last tested: June 2021No. Any query run in the UI (or API) will be subject to the proxy timeout rules, and it is not possible to increase that timeout past 60 minutes.If you're trying to build PDTs that take longer than an hour then you have a few workaround options.Workaround 1Use the --if dev-- parameter or some other where clause to make the PDT take less than an hour.Workaround 2Push the code to production and let the regenerator handle the build. (Make sure it has a sql_trigger_value so that the regenerator does attempt to build it.) That will circumvent the proxy. That way, you can have the PDT take as long as you need.Workaround 3If using the API, you can leverage async query tasks to get around this! You can run create_query() to make a query, pass that returned query id into create_query_task(query_id), then check on the query using query_task(task_id) to see when it is finished, then finally use query_task_results(task_id) to get the results of the query!⚠️ Only scheduled/tri

1700
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How is the "Estimated Time" for a query calculated?Knowledge Drop

Last tested: Oct 2020  Looker uses the time it took to run that query last time to predict estimated time. However, it uses the query parameters to determine this, not the SQL, so if you drastically change a LookML definition the estimate may be wrong.If the query hasn't been run before, Looker does not show an estimate. This content is subject to limited support.                

150
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Why doesn't my dashboard have a "Run" button?Knowledge Drop

Last tested: Sep 2020If Run on Load is “On” for a dashboard, and there are no filters on the dashboard, we do not show the “Run” button or the "Stop" button. This content is subject to limited support.                 

810
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Can I use a VPN to connect a Looker-hosted instance to a my database?Knowledge Drop

Last tested: June 2021No, but depending on your needs, you may be able to setup a host (tunnel server) that has VPN access to the desired database, and Looker can then set up an SSH tunnel from your instance to this tunnel server .If you host your own instance: You can use a VPN, if both the instance and the database are under the same VPN, then it would be fine. Otherwise, you would need to use an SSH tunnel. "under the same VPN" just means that they (instance and database) are both accessible in that private network. If the db and the instance are in the same network (via VPN), you should be able to talk to the db directly from the Looker instance. This content is subject to limited support.                

4260
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Is prefetching allowed in Looker?Knowledge Drop

Last tested: June 2021Prefetching is strongly discouraged. Use scheduling and datagroups instead. To pre-load dashboards when an ETL job finishes:Create a datagroup that triggers when the ETL job finishes. (Perhaps using a trigger that checks for new rows such as  “SELECT MAX(id) FROM table”) Schedule the dashboard to yourself or a placeholder email, based on the datagroup trigger.This content is subject to limited support.                 

1040
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How can I stop or change the order of my PDT builds?Knowledge Drop

Last tested: Feb 2021ProblemI have several PDTs building, and I want to either: Pause a PDT from building Control the order of my PDT builds (maybe I want a certain one to always build first) ExplanationThe order of PDT builds is arbitrary except when dependencies are involved. So there isn't any UI for pausing a PDT or changing the order of PDT builds (here's a feature request to +1 if you'd like to see this feature).WorkaroundsTo pause trigger checks for an entire connection:You can use the PDT Maintenance Schedule to control when Looker sends trigger queries to that database.It's a cron string, which means you can get a full pause by setting the date to something in the future. For example, if today is the 2nd and you want to pause all trigger checks you could set the cron task to 0 * 31 12 * for "every hour on the 31st of December".To pause trigger checks for one PDT:You can comment out sql_trigger_value or persist_for for a given PDT and push it to production to stop it from bui

4601
S
4 months ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

CORS support for LookerKnowledge Drop

Last tested:  June 2020What is it?The Cross-Origin Resource Sharing (CORS) is a mechanism that allows browser-based applications to directly call the Looker API without having to use an intermediary proxy server for the API calls! This will make it so you can more easily use the Looker API in you apps without implementing additional infrastructure. CORS greatly simplifies PBL application setup. It saves developer time, infrastructure costs, and operational overhead.How to install it? Go to {{ host_url }}/admin/labs page Enable the Experimental feature “Looker API CORS” Open the Web API Explorer (most often {{ host_url }}:19999/api-docs/index.html) Select the version API 4.0 from the dropdown (3.0 and 3.1 do not have this call) Login with your API client_id & client_secret (or go create them if you haven’t) Use register_oauth_client_app() to define your web app for Looker In your web app, when you want to engage the Oauth login with the interactive user, you send a web r

6440
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

If a user schedules Looks and then is disabled will the scheduled Looks still go out?Knowledge Drop

Last tested: Dec 2020 If the schedule was created by a user that is now disabled, the schedule will fail.The error in the schedule history will tell you that the User is not enabled. Contact your Looker administrator to reassign this scheduled report to an enabled user.This content is subject to limited support.                

2871
mitchellposluns
2 months ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

Error: Your account has been temporarily locked due to too many bad login attemptsKnowledge Drop

Last tested: April 2021 This error comes up if you try and log in with a bad password too many times.You only have to wait 5 minutes for it to unlock. This is non-negotiable, however you can instead use SAML/LDAP which may have providers that allow you to change these settings.Docs: https://docs.looker.com/admin-options/settings/login-lockouts This content is subject to limited support.                

3450
sam
Looker Staff
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

How do I filter out null in a filtered measure?Knowledge Drop

Last tested: May 2020Most of these examples can be found in the filter expressions doc.Filter out NULL values for non-number dimensionsThe most common filter syntax to filter out nulls is simply "-NULL" . This works for most data types that aren't numbers, such as strings, dates, etc.measure: count { type: count filters: { field: user.status value: "-NULL" # assuming that user.status is a string }}Filter out NULL values for number dimensionsFor type number, use NOT NULL instead.measure: count { type: count filters: { field: user.age value: "NOT NULL" # assuming that user.age is a number }}Filter out the string "NULL" from a string dimensionWhat if your string dimension literally has the string "NULL" in it? (As opposed to a real null (Ø)) Then you'll use this special syntax:measure: count { type: count filters: { field: user.status value: "-'NULL'" # assuming that user.status is a string that actually has 'NULL" as a value }}This content is subject to limited support.                

109652
J
1 month ago
Lesley_Rahbany
Community Admin
Lesley_RahbanyCommunity Admin
published in Common Account Questions

How can I investigate why my instance is so slow?Knowledge Drop

The first thing you want to do is figure out which dashboards are the slow performers. Our System Activity Pages will give you performance statistics and help with your investigation.  Below are some impactful changes you can make for optimizing Looker query performance: Build explores using many_to_one joins. Take advantage of datagroups. Applying datagroups within Explores and using the persist_with parameter, will allow Looker to integrate closely with your backend data pipeline. This will allow cache usage to be maximized without the risk of analyzing stale data.  Use persistent derived tables (PDTs). Convert Explores with many complex or non performant joins, or dimensions with subqueries or subselects, into PDTs. Learn about Aggregate Awareness logic to find the smallest, most efficient table available in your database to run a query while still maintaining accuracy. Leverage the Explain in SQL Runner tool for benchmarking. EXPLAIN produces an overview of your database's

24322
Lesley_Rahbany
Community Admin
1 year ago
molly.lippsett
Looker Staff
molly.lippsettLooker Staff
published in Technical Tips & Tricks

What does history count by source in System Activity mean?If you are looking at an explore like this that count represents the number of queries run from each source - the count matches the query run count measure in this situationKnowledge Drop

Last tested: Jul 1, 2020 If you are looking at an explore like thisthat count represents the number of queries run from each source - the count matches the query run count measure in this situation  This content is subject to limited support.                  

980
molly.lippsett
Looker Staff
1 year ago
molly.lippsett
Looker Staff
molly.lippsettLooker Staff
published in Technical Tips & Tricks

Best practice for handling group ids over multiple instancesKnowledge Drop

Last tested: Aug 13, 2020 Say you have staging/dev/production instances and your group ids are different per instance.The recommendation is to keep the names the same, then1. run api call on group name to get group id, save as variable2. add the group id variable to the sso embed urlNo magic there. best you can do is query for all_groups then pick them out by name. or the app code could keep a mapping somewhere of all the names => id for every environment, so you don't have to keep looking it up This content is subject to limited support.                  

460
molly.lippsett
Looker Staff
1 year ago
molly.lippsett
Looker Staff
molly.lippsettLooker Staff
published in Technical Tips & Tricks

Bigquery Error while reading table: Found corrupted data while opening fileKnowledge Drop

Last tested: Jun 30, 2020 This error indicates that one of the tables being queried has 'corrupted data' - more info hereRunning the query directly on the database will yield the same results in this case. A possible next step from that point to consider might be reverting the corrupted table to a previous state - some instructions on that here This content is subject to limited support.                   

550
molly.lippsett
Looker Staff
1 year ago
molly.lippsett
Looker Staff
molly.lippsettLooker Staff
published in Technical Tips & Tricks

List of workarounds for sorting pivots by measure / sorting by column totals / dynamic top n rankingKnowledge Drop

Last tested: Jul 13, 2018 The problemSay you want to see some metrics for your top 10 customers by number of users, where the customer dimension is pivoted. Currently there is no easy way from an explore to sort a pivoted dimension based on measure values—you can only sort the pivoted dimension alphanumerically. Some features requests have popped up related to this issue like:Sort by Row / Sort pivots by value Rank top N / Pivots sortable by measures Sort or limit pivot columns by top valuesThere are a few workarounds you can use.Workaround #1: Derived table with rank window functionThis is the approach used in a couple of help centre articles:Sort by Pivots, Rank with Other Bucket, and Percentile Tail Analysis Dynamic Rankings (Filtering Dimensions by Custom Rank)This is the most robust solution, but it requires some work to set up the query for the derived table. Workaround #2: Sort row totals and transposeThis workaround can only be used with a table visualisation and can only suppo

39181
chriswragge
1 year ago
sam
Looker Staff
samLooker Staff
published in Technical Tips & Tricks

If the cache is full, then what gets dropped first?Knowledge Drop

Last tested: May 2021 Looker's default max cache size is 2GB. This can be changed with the --on-disk-cache-size=<i> startup option, where <i> is a number of megabytes (so, default 2000).If your Looker instance detects that the on-disk cache exceeds the max, then it starts purging old cache entries until the limit is no longer exceeded. The algorithm to determine which entries to purge considers the age of a cache entry as well as the last time that cache entry was accessed.  This content is subject to limited support.                 

2171
Anton
1 month ago

Badges

  • Friendly Face
    sbrylowskihas earned the badge Friendly Face
  • Helpful
    carriehas earned the badge Helpful
  • Community Star
    carriehas earned the badge Community Star
  • Friendly Face
    carriehas earned the badge Friendly Face
  • Community Star
    Dawidhas earned the badge Community Star
Show all badges
Terms of Service