PhanLe1010

> @PhanLe1010 afaik, influxdb v2 is using libflux written in rust, while influx itself in go. why not simply only go-lang wrapper? Thanks @cz172638 for the wrapper idea. We will...

Thanks a lot @ming1 ! We are very appreciate your support! > > Hi @ming1 > > Thank you very much for the explanation! Yeah, it totally makes sense as...

> I think actual size is very misleading and there should be displayed also the real size of the pvc. How should I monitor pv size otherwise? Inorder to monitor...

> What's the point of having that "wrong" displayed size? If I have to show the dashboard to a human technician that isn't confident with the kubernetes stack? He'll just...

> > As you can see I don't have any snapshot, the only snapshot i had was "removed" and I pressed 3 or 4 times on the delete button. Then...

Hi @simonebenati > i just don’t still understand why prometheus shows me the correct pv metric, and longhorn can’t. Longhorn currently doesn't show the size of filesystem (this is showed...

Update: * @shuo-wu and I were able to reproduce the fstrim corner case in our lab which prevents the actual size of Longhorn volume from reducing after fstrim. @shuo-wu will...

Surely @simonebenati Take a rest! We agree that the conner case should also exist in the latest Longhorn version. Please refer to the above comment from @shuo-wu to quickly verify...

Maybe we should introduce a backoff mechanism to slow down the attach/detach retry instead of stopping it indefinitely? Agree that we should record the failed to attach reason inside volume...

TODO: need to come out the LEP with the points mentioned in https://suse.slack.com/archives/C02EDCA93RA/p1728437994567579