Secoda Community

image.png

We’ve also added this to bugs to fix this week, thanks for reporting Ryan!

You’re correct <@U031Z4XUNMD>, thanks for reporting. I’ve also added this to the other items you brought up yesterday

Untitled.jpg

I believe this is because your session ID has expired <@U02FSG2JZ39>. You’ll have to refresh in *Option B* in the documentation here. <https://help.secoda.co/en/articles/5461987-metabase>

Unfortunately, Metabase’s API doesn’t allow for the refresh of Google Oauth session IDs automatically

I suppose that we use username/password access at the moment. Previously, dedicated metabase user with admin rights for secoda has been created.


Can we still use pure username/password access to metabase?

I’m going to move this to a DM <@U02FSG2JZ39>

Hey Rodrigo, just going to take this conversation to a DM

Hi Luciana, we’re looking into this right now. Thanks for reporting

Hey <@U0345H7N094>, we can try and troubleshoot this in a private DM since you’re using on-premise

We uploaded a lot of blank terms from de API and we need to delete it!!!!
Thanks!

We’re working with <@U035ZUNQN9G> so that he can delete them via the API. Should be available shortly

It's inverted. Like scrolling my mouse wheel up zooms in for one but zooms out for the other

Hi <@U031Z4XUNMD>, thanks for catching that! I'll let you know when we push that fix out :slightly_smiling_face:

Hey <@U0383EVU5C0>, good question. <@U03113A05T6> on our team is actually pushing an update that will have the correct nomenclature in Secoda. This should be available in the version of Secoda you see next week.

Hey <@U0383EVU5C0> *we can pull the manifest.json daily from an S3 bucket that you specify*.

We’ve been flexible in the past about setting up additional options if the current options don’t satisfy how you store your manifest.json files, so let us know!

This sounds like a perfect use case for "Collections". In Secoda, you can "tag" Q&amp;A, tables, Looker Explores ect. into a collection so if someone wants to go deep on "Conversion Rate", they have a place to go. Here's a video of me setting up a collection for your specific use case to help you see how this works: <https://www.loom.com/share/d90f8ce153fe416a891d24535ad1718d>

Amazing answer. Thanks a bunch, will look into this

Thanks for pointing this out <@U031Z4XUNMD>, we’ll push a fix for it asap!

<@U031Z4XUNMD> check out the new documents too. You can now add them to collections CC. <@U02LVDYD1AS>

This is awesome, question... is it possible to make the side panel expandable by the user? Most of the table names are cut off due to sizing of the panel

Also, can the tables/views be sorted alphabetically?

Good points on both <@U02739P8464>, the alphabetical is no problem. And we do want to make the sidebar expandable it’s a little more tricky. I’ll give an estimate about when that would be available

Screen Shot 2022-03-24 at 2.59.42 PM.png

Hey <@U02QJG595KM> apologies about the delay, can you refresh the page and see if you get the popup still?

<@U01MHQR7DE2> It's gone now. I appreciate the fast response :smiley:

Screen Shot 2022-03-25 at 2.31.30 pm.png

We’ll fix this issue today, thanks for reporting <@U028E7B1U31>  <https://linear.app/secoda/issue/SEC-498/blank-documents-showing-up-in-search>

Looking into it right now <@U02QJG595KM>!

Do you have an email of someone that cannot view those resources <@U02QJG595KM>?

It looks like a lot of your resources are in “Draft” mode which means they won’t be viewable for viewers. If you change all the items that you want viewers to see to be “Published” they should be able to see them

They're both editors. They don't even see the published docs

Can you ask them to sign out and sign back in and see it if resolves the issue

<@U01MHQR7DE2> Can you please let us know if there is any information available for the above question?

Hey <@U02H1V13V99> we don't currently, but we can extend the other sharing capabilities to the questions quite easily. 

I've added the issue to our backlog <https://linear.app/secoda/issue/SEC-499/question-share-permissions|https://linear.app/secoda/issue/SEC-499/question-share-permissions>

I was asking about the share option below in the above question. My manager figured that out and thank you for answering before...

Oh sorry I misunderstood. Glad you found it!

Screenshot 2022-03-30 at 2.13.47 PM.png

Hi Jonathan! Looking into this now for you.

Hi Jonathan, your search results will appear if you go here: <https://app.secoda.co/search>. It seems your url contains a broken filter. I will push a hotfix for that specific issue now.

Hi Carter, I confirm now I can see the resources! Thanks for the quick fix!

Hey Richard, just so I’m understanding correctly, do you mean that you’d want to push all the markdown content of the the documents associated to a collection in GitHub? If that’s correct, it would be good to understand what structure you’d like the data to have?

yea im not sure what is possible as far as structure goes... I have a collection on github of individual markdown files with images and content populated and I would like the ability to "publish" these markdown files to secoda without having to manually copy over would be the dream.

looking deeper i see there is a way to link a github repo to secoda for "backup" currently enabling that now to see how I can use it. Being able to edit the markdown on my local machine and then update the repo and it finds its way into secoda may be to much to ask for from this but going to look into what I can do with this connection.

yea may not be what im looking for as it ouputs a bunch of json meta data....

I was going to mention the git integration <@U033ER26CPN>. It is a bunch of json meta data but included in that metadata is the markdown documentation content. It’s currently only one way from Secoda -&gt; git, but I imagine even if it was Secoda -&gt; git and git -&gt; Secoda it wouldn’t solve your use case here.

One thing you could immediately do is use the API, but that doesn’t solve the version control issue. Let me add this item to our backlog and I’ll let you know if the rest of the team has any ideas about the best way we can solve it.

yea the API was our first thought too. Thank you for keeping me posted.

Hey Ryan, we’re pushing a fix for this apologies about that

Can you refresh and try again <@U031Z4XUNMD> it should be resolved now

Sorry about this <@U030MM6J3T7>, it’s been fixed now. If you hard refresh you should be good

<@U02LVDYD1AS>, we have column level lineage for Snowflake at the moment, we are working out edge cases in parsing column lineage from sql queries. I have moved Redshift Lineage to this week's set of tasks and will keep you in the loop on this thread :slightly_smiling_face:

<@U02LVDYD1AS>, do your tables have primary and foreign keys set?

Oh, that's great to hear! And...likely they do not. I could set them via the yml file in dbt though, I believe.

Capture d’écran 2022-04-14 à 12.25.33.png

Hey Marin, you’re correct that’s a bug with the export feature. We’ve added it to our bug list to resolve. We’re working on improving the export/import functionality so we can add this to the scope.

For importing, you’re also correct that you can import only a few descriptions, you don’t have to import everything.

Hi <@U02NN9PTP71> apologies about the delayed follow up the team was OOO on Friday.

The platform should work on Linux/Firefox. When the user is saying that Secoda didn’t work for them did they mean the application didn’t load or something more specific? It would be great if they joined the community so we could ask them some questions to understand the issue.

ok, thanks for the feedback. i didn't get more detail but follow up if i do.

Hey Sherri, here’s some documentation about your questions:
• <https://docs.secoda.co/user-management/roles|User Roles> on Secoda
• <https://docs.secoda.co/integrations/postgres|Connecting to Postgres>
I’ll also send you a DM with the Postgres queries we run if you’re interested?

yes please. any docs are good docs. :slightly_smiling_face:

Hey <@U03C85WHYSX> the removal of the badges and owners is a bit confusing and needs to be re-worked but you can remove them by clicking the + button and then clicking “Remove PII” it’s similar for owners

Hello Alex, I think this is something failing on our end when trying to update the search indexes after an extraction completes. I am going to work on fixing it today

<https://linear.app/secoda/issue/SEC-638/redshift-extraction-fails-on-last-run-es-jobs>

Hi <@U03DA7NBWEB> unfortunately we cannot integrate with Airflow MWAA because AWS decided that the Airflow API is a security risk and disabled it. I did some research on this about a month ago, so if you know of anything that’s changed let me know and we can look into it!

yeah :confused: it kind of sucks! i was mostly asking because there are certain workarounds like using the REST CLI client (which is what we use) so i was just wondering if someone had created an integration using that

We have had one other company request the same thing. It’s definitely possible, but for these types of integrations that a limited number of companies are asking for, we want to make sure that the work we put in won’t be wasted.

So usually before we start to build out the integration it requires a conversation with <@U03A4PFACG1> and <@U01PM6182AX>. Let me know if you’d like me to arrange that conversation

completely understand :slightly_smiling_face: i don't want to distract from all other high priorities, so just if you get more requests on MWAA i will tag along with a +1 from my side

Hi <@U02HEA5UD7F> I’m looking into this issue now. I’ll let you know once I’ve found the cause of the issue.

Hello Andrew. I can access the tables now. I just deleted and recreated both DBT and BQ integrations. However, I have noticed weird things and behaviours. I’ll message you directly

Thanks Benoit, apologies about the delay on the fix to the original issue. Since things started working properly after resetting the integration I have a good idea of what the problem is. I think that there’s a number of unique queries that are associated with the tables and so that’s making the API call slow. What we can do is separate the query GET method from the actual table GET method since right now they are in the same API call

Hey <@U03DK5XE1FZ> are you documenting through the yml files in dbt?

The other thing you can do is check if the documentation exists in the `manifest.json` that is outputted after the dbt job finishes

Hey <@U031Z4XUNMD> I’m taking a look at this now

<@U031Z4XUNMD> this should be working now

Screen Shot 2022-05-09 at 2.34.00 pm.png

same for me. I can't access any of the tables (Snowflake integration)

<@U028E7B1U31> <@U03EG2VM88G> Apologies about this. We are pushing a fix that will be available in 5 minutes. I’ll let you know when it’s live

<@U03EG2VM88G> could you try to see if things are loading correctly when you get a chance pls