I'm having issues with self-hosted version, either on v0.75 and on lastest v0.76. I raun the install script and I get nothing on the dashboards. I see the containers logs on click house and I see a lot of errors:

hey - are you migrating from an earlier version of SigNoz or installing a fresh version? We are making some changes in v0.76. You might want to check this migration guide if you are migrating from an older version - https://signoz.io/docs/operate/migration/upgrade-0.76

I did a fresh install

hey @Nuno Aguiar , If you're trying with a fresh installation then could you please share the errors you're seeing. it would help us to understand more about the problems you're facing

Untitled

Adding @Vibhu Pandey for any suggestions

That's no problem @Nuno Aguiar, you can ignore broken pipe errors in clickhouse

ok, but something is not right. i've setup a simple collector for host and docker_stats and I can validate the data is beeing saved in clickhouse, but I can't see any data in any dashboard. after digging some stuff i've managed to check that i have some memory errors !? in clickkhouse. For instance:

Received exception from server (version 24.1.2):
Code: 173. DB::Exception: Received from 127.0.0.1:9000. DB::Exception: Couldn't allocate 638 bytes when parsing JSON: while executing 'FUNCTION JSONExtractString(labels :: 0, 'host_name' :: 1) -> JSONExtractString(labels, 'host_name') String : 2'. (CANNOT_ALLOCATE_MEMORY)

Right, can you increase the size of the clickhouse disk?

I'm running a self hosted signoz in docker inside a linux vm with 64Gb Ram and 128Gb disk

and how does your disk utilization look?

20%

Interesting, I don't have something concrete for you right now, can you create an issue https://github.com/SigNoz/signoz? Probably just copy paste this conversation

for instance, in infrascture-montoring, as long I setup a hostmetrics receiver and export to signozclickhousemetrics, I should have data there right?

For infra monitoring, we use a couple of receivers to populate that page. Our k8s-infra chart has all the defaults required for the infra-monitoring page to work perfectly.

but that's only for running on k8s right? I'm using simple docker containers for testing on a linux vm. and since all dashboards use the same JsonExtractString function, then all dashboards fail for me