"Internal server error" error="[plugin.downstreamError]

[bibl2008]

Hello, I have: zabbix 6.4.1 grafana 9.4.7 (commit: 4add91f03d, branch: HEAD) I try to add the dashboard:

When I try it I have the next error logs in /var/log/grafana/grafana.log: logger=live t=2023-04-20T04:31:29.609969219Z level=info msg="Initialized channel handler" channel=grafana/dashboard/uid/dND7S1PVz address=grafana/dashboard/uid/dND7S1PVz logger=context userId=1 orgId=1 uname=admin t=2023-04-20T04:33:48.721072132Z level=error msg="Internal server error" error="[plugin.downstreamError] failed to query data: Failed to query data: rpc error: code = Canceled desc = context canceled" remote_addr=192.168.20.200 traceID= logger=context userId=1 orgId=1 uname=admin t=2023-04-20T04:33:48.721734454Z level=error msg="Request Completed" method=POST path=/api/ds/query status=500 remote_addr=192.168.20.200 time_ms=66 duration=66.41905ms size=116 referer="http://10.129.0.14:3000/d/dND7S1PVz/home?editPanel=2&orgId=1" handler=/api/ds/query

[chladic]

@alexanderzobnin Im facing same issue. When I have host with 300 items, im getting this error as well. I tried upgrading zabbix plugin, upgraded zabbix to 6.4.6, upgrading grafana to latest version. Also I tried to increase memory in PHP, I upgraded php to 8.2, nothing helped.

[jhaos]

Hello anyone could solve this? Im getting the same issue trying to get hosts with a big amount of items (around 500) and time ranges over 3h after updating grafana to 9.3.2, zabbix to 6.0.10 and the plugin version 4.2.10.

[julioreyesdatadope]

Hi Im using 9.3.2, zabbix to 6.0.10 and the plugin version 4.2.10 in Kubernetes.

After some troubleshooting I saw that the error is beceause the process of the plugin is getting killed and restarted. It is happening with hosts with 4k items which are split by 3 diferent tags.

The plugin holds correctly the 1 hour interval but more than that the plugin process starts to restart. We suspect it can be the amount of datapoints since when we tried in local environment for debugging and the datapoints for longer interval were about 200k.

I've tried the last version of the plugin as well but faced the same problem. There is in the roadmap something to fix this?

[zoltanbedi]

Hi @julioreyesdatadope,

Thanks for your investigation. Sadly it is not prioritized right now to do any work for zabbix but we want to commit to it in the next quarter.

[egorky]

I am having the same issue. Zabbix version 6.4.12, Grafana v10.4.1 and plugin 4.4.6. On panels with host groups with more than 300 hosts, I get 500 Internal Server Error. Could this be Zabbix's fault? It looks like the API is giving the error