Cr_default_p_neos_asset_usage

Hello

I am using Neos 9 beta 13 and I have today with

./flow content:createVariantsRecursively

created my nodes in three other dimensions. Now unfortunately I have a cr_default_p_neos_asset_usage table with around 10 gigabytes of data. I’ve tried a few things: contentstream:prune and then a cr:projectionreplayall, but the table blows up again. The system now reacts extremely slowly when publishing the workspace. Does anyone have an idea what I can do?

Greetings,

Jan

Hi Jan,

thanks for reporting this. This is very helpful for us.

We are already in the process of rewriting the asset usage storage, which should™ also address your issue.

This is currently in the review process and will make it hopefully into the next beta release.

May I ask you for patience until the next beta release?

If there is there any chance to get a dump of your event stream, I could check in depth if this issue will still persist after the overhaul.

Best
Denny

Thank you for your work. I sent you a link to a SQL dump. I help myself right now by exporting and importing again :wink:

Great, I will have a look into this.

Even if I import from an export, there are a lot of entries created. But much less:

And that happens after the first login:

And when I run

./flow contentgraphintegrity:runviolationdetection

I get:

Hierarchy relation {“position”:128,“contentstreamid”:“9ac41adf-835f-4a93-9774-403cebba594b”,“dimensionspacepointhash”:“5797e6cfaca8c2b6d12c0fbb5c753642”,“parentnodeanchor”:32,“childnodeanchor”:33,“subtreetags”:“{}”} is disconnected.

Subgraph defined by content stream a0682a3e-7a8f-4b70-bdb9-9aeef55b84e5 and dimension space point {“language”:“fr”} is cyclic for node aggregates 9530b0a6-391a-4f69-b59f-be3e6a63c0d6,4656bc1a-cbff-4742-a37d-5e7b0727bb3d,019bceb1-6cc6-43a1-aeae-2f3bcd19a40d,d5ee101f-acdc-421d-9b56-824e2ea0af48,afd2322e-60d6-450a-ad83-e234e804fafc,8f84f268-0c59-410d-8afe-2c6c5f77545c,db697162-b712-41d2-82f9-562e7d56d7ed

and

Destination node aggregate 11e33631-2145-4874-a0d9-a6c024058dfb does not cover any dimension space points the source ea20c788-84b2-4bea-a94c-679c1c2daa01 does in content stream 9ac41adf-835f-4a93-9774-403cebba594b

These are only examples…

Hi @jan82,

I had a look into your data and as I said this issue with the asset usage table will be solved with the next beta release. We completly changed the way how the asset usages are stored, which is more efficent.

Regarding your disconnected nodes, I would suggest to create a new topic, as this seems to be unrelated to the asset usage.

Best
Denny

1 Like

Thank you!