chilly kelp
It would be great to have more dashboards for StorageGrid, including hardware metrics and NIC port utilization. This would help identify which ports are heavily used and how their usage changes over time.
This information will assist in planning load balancer extensions and determining whether we need to switch to 100Gb ports instead of 40Gb.
sterile stratus
hi @chilly kelp those sound like great additions. Have you started working on these and if so, would you like to contribute them?
chilly kelp
Unfortunately, no. At the moment, we are only using Nabox. Would it be difficult for a Nabox expert to implement or add such metrics?
sterile stratus
Do you know if Harvest is already collecting the StorageGRID metrics you want? If so, then adding a dashboard to visualize those metrics is straightforward for someone familiar with creating Grafana dashboards.
If Harvest isn't collecting the metrics you want already, then adding those missing ones would be the first step.
I updated the existing Harvest issue for StorageGRID dashboards with your comments https://github.com/NetApp/harvest/issues/1420#issuecomment-2358730677
chilly kelp
Awesome, many thanks. I checked, but Harvest isn’t collecting those metrics yet.
sterile stratus
if you have the names of the metrics you're interested in collecting, please drop them here when you get a chance
delicate cape
Would also be interested in the stuff @chilly kelp is proposing. +1
sterile stratus
thanks @delicate cape are you the original author of issue #1420?
delicate cape
Yess @sterile stratus , was nearly 2 years ago now😊
sterile stratus
Thanks @delicate cape We've managed to add a couple since then 🙂 If you or @chilly kelp share which metrics from StorageGRID you would like visualized, we can add more dashboard/panels with those
chilly kelp
Sorry, I had to get some data out of Harvest because Discord is blocked in our company. I’m checking the Grid metrics to see if I can find something similar, but what helped us with the Nabox metrics was monitoring the load balancing of our A900 FC and Ethernet ports for our VMware infrastructure. At the moment, our biggest Grid customer is Splunk, and they are ingesting and retrieving a large amount of data. We can see the overall GB/s in and out, but I can’t currently determine if a port is 100% overloaded or not.
Such a feature for StorageGrid would be great. Being able to select a storage node or load balancer and see the utilization of individual ports would be really useful.
sterile stratus
ok that helps. You've shared a screenshot of Harvest's ONTAP: Network dashboard and your request is to provide a StorageGRID: Network dashboard with similar panels
chilly kelp
would be ideal yes. 🙂 i will search as well for more storageGRID metrics if i can find some i will share it here. for your notes we are running the Grid with SG6060 and SG1000 and we would be very intrested in first draft for the sg1000 ports usage.
sterile stratus
thanks for the details @chilly kelp
If it helps, I dumped the StorageGRID metrics here https://github.com/NetApp/harvest/issues/1420#issuecomment-2358810719
I see storagegrid_network_received_bytes and storagegrid_network_transmitted_bytes
It's not clear yet if StorageGRID provides the necessary metrics to build a dashboard like ONTAP: Network
sacred cape