Looker set-up and management
Open sourcing tools for Looker The link: https://github.com/caura/tools About these Tools Python/Shell scripts. Use them at your own risk. They are clearly not well-documented or productionized, but some are better than others. More, where that came from. If there is interest in us sharing such scripts, let us know.
When downloading queries, some queries could require large amounts of memory. Old Behavior When downloading all results for some types of queries, the amount of memory required could cause performance issues, even crashing an instance. The All Results option is disabled for users with download_with_limit permissions but other users could download all results even for queries that have memory-intensive characteristics. New Behavior Starting in Looker 4.14, you can prevent these queries by turning off the Allow unlimited downloads that might crash Looker legacy feature. When the legacy feature is off, the All Results option is disabled for queries returning more than 50,000 rows in situations where the rows need to be stored in memory. This includes queries returning more than 50,000 rows with one of these memory-intensive characteristics: involving row totals or table calculations using percent_of_total, percent_of_previous, or running_total measure types getting data from a database
Hi, I am using Looker version 4.22. Currently looker allows to have control (Role) by model set but whenever new model is added to a project, Admin manually have to add that model to model set. Is it possible to have role by project so whenever new model added to project, users/developers already have access to that model? The reason is we are creating seperate developers/users group based on data and want to restrict them by Projects so they cannot see/modify other explores/models/sapces. Also is it possible to have admin roles by project in Looker, by default Looker admin has full access to complete environment.
Hey there, gang. One of my TODOs this year was to get our on-premise looker machines a little more well-oiled. As part of that, our internal workflow infrastructure relies heavily on chef for devops automation. And, since there wasn’t an existing looker chef cookbook that I could find, I took a few hours and pounded one out. My question then, is: how can I get discreet URLs for downloading specific versions of looker? For example, our prod machine is still at 4.22, and we’d like to migrate to 5.x in the next month or two. But not 5.4 yet. So controlling which version of the jar we download is important. It looks like all the download links in emails, just go to “latest”. Help? (PS: more than happy to open source our looker cookbook as well.) Thanks!
I’m trying to configure our looker environments to have a staging environment and a live environment, with pull requests required before any changes can be deployed into production. We have this working in our live environment with a webhook created for the deployment, and have reviews required on all pull requests before they are deployed. However, when I merge a pull request, this will obviously deploy the change to both the live and development environment. Is there a way to have this changed so that a pull will deploy to the development environment, and then have further sign off before being deployed to live?
I migrated looker to different location due to space issue. After migrating looker, none of the users are able to login as password is not working. Unfortunately as we are doing in test environment, we do not have smtp set up. Looks like there is no other option apart from mail to reset password. can you please help. thanks.
we have a user running a query with no limit, it was my understanding this is not supposed to be possible. can you please let us know where users have this ability and how to prevent it? update: the query is coming from the “internal” source, looking for docs on this now… current version: 4.14.13 (though upgrading to: 5.0 this week)
hey there - does anyone know if there’s a way to white-list specific data action url end points starting with “http” (vs “https”)? would make hitting internal web service end points so much easier (vs setting up all the certs, etc). would love any advice here/thoughts on if we can add this functionality in the admin features in the near future.
My organisation has just tried to update to Looker version 5.0.14 from Looker version 4.12. Before we copied the new .jar file into place, we updated our Java JDK to version 1.8.0_151-b12. We then tried to start up Looker using the new startup script at; https://raw.githubusercontent.com/looker/customer-scripts/master/startup_scripts/looker The program does not load, but instead shows the following error message; ERROR! You are running Java Version (OpenJDK 64-Bit Server VM) which is unsupported. Please ensure that you are running Oracle Java 8 update 121+ or higher. Running “java -version” from the command-line shows the following; [looker@wf-looker-dev looker]$ java -version openjdk version "1.8.0_151" OpenJDK Runtime Environment (build 1.8.0_151-b12) OpenJDK 64-Bit Server VM (build 25.151-b12, mixed mode) OpenJDK 64-Bit Server VM (build 25.151-b12, mixed mode) The previous Looker version 4.12 runs perfectly, without any error messages. Please can someone help…?
We have 3 Looker environments setup for dev, QA and production to allow gating of our deployments. Due to the nature of our business we need to be able to develop both LookML and Looks and then push them between the environments. We need to source control all elements (including Looks) and be able to pack them into nuget packages to allow our automated deployment tool to push them between environments. We also need to perform code and implementation reviews and log them (for audit purposes). I have been investigating the Looker API and I know I can copy Looks, dashboard etc. and I have those functions as a proof of concept. Our current setup for other development is to work locally (on our own PC). Request a code review. Commit our changes. TFS build (which includes packing a nuget package with the artifacts and pushing it to a nuget server). Run Powershell to push the changes to the relevant server within the correct environment. We are trying to work out the best way of using Looker.
Hi there, We are running into an issue that I’d love some feedback and suggestions on. We have a nice Looker model running for which my team is responsible on the data side of things. We make sure the data ends up in the correct database, but also maintain the Looker model. We have prepared a couple of dashboards for the end users, so that even people with no Looker experience can see the data. This setup worked fine when we were slowly releasing Looker into the organisation with a couple of end-users and the communication lines were short and easy. Due to the success of the tool, more and more people take an interest in the data. This on itself is awesome of course! However, this means we have to find another way to communicate some of the properties and quirks of the data and model to the end user. I am looking at a central place to communicate this to all users, a bit similar to a .README for the model. Any suggestions for this? Interested to hear how other people deal with this as
Hi all. When we do a Looker upgrade, we basically zip up the entire looker folder and if we needed to roll back, we would just unzip our backup. My question is whether or not we need to backup the .tmp and .cache folders. These can get pretty large, and adds a non-trivial amount of time to generating the backup. For example, right now, .tmp and .cache are around 6GB in total. Is it safe to exclude these from the backups?
Anyone figure out how to use the Looker scratch temp database feature on a read-only instance of MySQL? Cannot use persistent derived tables in “looker_scratch”[created cached table] Java::JavaSql::SQLException: The MySQL server is running with the --read-only option so it cannot execute this statement Thanks
I am trying to do some maintenance on our Looker install and have found a whole bunch of views that someone created but I am pretty sure are not being used anywhere. But thus far I have not found a reliable way of telling if specific views and dimensions are in fact being used in Looks anywhere. The closest that I can come up with is going to dev mode, removing something and then using the content validator to see if it comes up somewhere. It would be great of there was a “Find Uses Of…” function for Views/Dimensions/Measures.
Is there a report or way to get a list of all the Looks that have a Public URL turned on? We are considering turning this on for some integrations, but in the interest of data sensitivity we want to be able to review a list (and drill into the details) of all Looks that have this feature turned on to avoid sharing things that aren’t allowed. Thanks! ~Forrest
Since we started using Looker I’ve been creating self-signed certificates that expire every three months, so we went ahead and bought one from our usual provider, Thawte. We are hosting Looker on an Ubuntu Linux server, and while we have Apache on it, my understanding is that Looker has it’s own server built in. In working through the certificate installation instructions here, I’m not sure what files I should be using. Thawte gives me options when downloading the cert, such as choosing Apache, Windows, or Other. If Apache, it gives a choice of HTTP Server or Tomcat server, which would yield certificates in either X509 or PKCS#7 formats, respectively. In either case, the files I download don’t match up with the instructions linked above, so I’m not sure what I need to do. I get a file called IntermediateCA.crt, ssl_certificate.crt, and I have the key and csr files I originally used to create the cert request. So my question is are either of the .crt files I got equivalent to the .
Our on-premise looker instance has started getting more use, so we recently increased the memory available to the JVM by setting JAVAMEM=6100m in the looker startup file. While looking thru the startup options I noticed there are options for --in-memory-cache-size and --on-disk-cache-size that default to 50mb and 2000mb. What are the recommended settings for these values on a system with 8GB of RAM and only running looker?
Already have an account? Login
Login to the community
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.