-
Notifications
You must be signed in to change notification settings - Fork 8
No containers instructions ? #5
Comments
We are in the same situation. Documentation around scraping the native /vsanmetrics endpoint would be ideal. |
I was able to get this working without containers.
|
I tried with CentOS7 and RHEL7.9 and AlmaLinux8/RHEL8.6 and both are getting the same error
then I thought, i might as well try to get the docker version working but that's also failing
This is on brand new AlmaLinux/RHEL8.6 and docker ce installation. Obviously connectivity is there to packages.vmware site (see curl command) so i'm not sure what the next step I can make to get this running :-( |
The first error looks like you're missing pyvmomi. Also, you are missing the required parameters, "-s vcenter -u user -c cluster" |
Thanks for the response.
Originally i left out the IP/username because the error is the same whether i provide credentials or not.
|
Please try to copy the file |
Thank you - that has worked. I was able to get it working after that without containers or even exporters. 1, I wish there was a way to export the bearer token from within the vSphere GUI - or is there ? Then we could simply adjust Prometheus configs and scrape metrics away - for static environments. 2, Is it documented anywhere in VMware or perhaps vSAN documents what DOM OWNER LEAF or LASOM etc. acronyms mean ? We do use vSAN but we arent specialist by any stretch in vSAN. 3, I know this is a stretch, but is there anything like this but for ESXi host metrics ? Thank you for your help |
I found this link helpful in explaining the various vSAN components. https://vinception.fr/vmware-vsan-architecture-components/ I found the CompMgr role to match the "backend" statistics in the vCenter GUI. |
We use this exporter to scrape vCenter (non-vSAN) metrics, including ESXi hosts. https://github.com/pryorda/vmware_exporter |
I know about that exporter but its a bit buggy (if you clone vm for example) plus incomplete metrics (for example you cant get basic metrics such as network bandwidth utilization per interface - only network usage per host which is useless for hosts with several nics - ive got esxi server with 12 nic's for example) It'd be nice to be able to query esxi hosts directly for metrics without exporter in between like we can do here with vsan. |
Out of curiosity, are you using the method above by @JDDavila to scrape vSAN metrics? I'm currently looking for something similar. |
yes, im able to get vSAN metrics. |
Is your feature request related to a problem? Please describe.
Can you please described if it is possible to run this project with existing Prometheus/Grafana setup but without containers ?
We are not using docker (or any containers) in Production but we still would like to monitor our vSAN performance cluster.
We already have Prometheus and Grafana running - all we would like to is to scrape the metrics into existing Prometheus instance(s) ( for example from a single VM running red hat/cent os )
I tried to make some sense from the instruction and just simply extract needed binaries but documentation is confusing (I'm not sure which binaries exactly are needed even after re-reading "simple getting started" multiple times I'm totally lost)
Thank you
Describe the solution you'd like
I'd like instruction how to monitor VSAN using existing Prometheus/Grafana infrastructure but without the need to deploy containers.
Describe alternatives you've considered
We are using Telegraf at the moment but we've moved away from INFLUX and into native Prometheus/VictoriaMetrics database and VMware monitoring is the last solution where we cant find official vmware/vsan exporter binary.
(community supported vmware-exporter does not scrape vsan metrics)
Additional context
No response
The text was updated successfully, but these errors were encountered: