You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As a follow up to #4166, I wanted to use config cache svc in restore svc in order to improve querying node info.
The problem is that config cache svc identifies node by IP string. Apart from the problem that node IP might change over time and we should handle it gracefully, the IP string itself is not a unique representation for given IP, e.g. 2001:db9:200:0:0:0:0:32 equals 2001:0DB9:200::32.
In my case, some tests failed when SM wanted to get node info:
2024-12-17T14:50:42.0601679Z helper_integration_test.go:206: run backup: upload: 2001:db9:200:0:0:0:0:32: system_distributed_everywhere.cdc_generation_descriptions_v2: check Scylla backup API support: no config for node 2001:db9:200:0:0:0:0:32
This could also create other problems, as unfortunately SM code was written with the assumption that node can be identified by string IP.
The text was updated successfully, but these errors were encountered:
As a follow up to #4166, I wanted to use config cache svc in restore svc in order to improve querying node info.
The problem is that config cache svc identifies node by IP string. Apart from the problem that node IP might change over time and we should handle it gracefully, the IP string itself is not a unique representation for given IP, e.g.
2001:db9:200:0:0:0:0:32
equals2001:0DB9:200::32
.In my case, some tests failed when SM wanted to get node info:
This could also create other problems, as unfortunately SM code was written with the assumption that node can be identified by string IP.
The text was updated successfully, but these errors were encountered: