Tomaz Muraus
Tomaz Muraus
I had a look and by default we only cache data in memory for drivers for which pricing data is requested by the using so this will only affect people...
I added a note on pricing file size + potential small memory usage increase (ce5b74479f4d21f2928d2d2b56d1cf734c4ee38a) and also updated the script to include progress bar when parsing sku pricing data (7bffd095b151d451bf6f723e89a88745c52399a8).
Merged, thanks!
@pgandhi-delphix Sorry for the delay, I will look into it this weekend.
As far as your two questions go: > I have also merged the disk API versions because the zones parameter was only supported in newer API versions. I do not...
@lowercase00 That approach reasonable to me. It's also worth noting that some APIs (e.g. storage) support both approaches - driver methods call object methods or vice versa. So the object...
Thanks for the contribution. Overall the change looks good to me, but I would be good to see the following: 1. I would like to understanding the reasoning / context...
Thanks for reporting this. I just fixed this in #1423 recently (that PR still needs some tests, etc. though).
Correct, GKE doesn't expose those operations through their API. One option is to connect directly to the API service running on the GKE Kubernetes cluster using regular Kubernetes driver.
Thanks for reporting this issue. I'm looking at the code and I wonder if / how this ever worked since that code seems incorrect and would throw on every instantiation.