Start listings consistently with the current PG state, add wait_up_timeout

This still doesn't make listings 100% consistent yet; for 100% consistent
listings we have to receive listings only from the primary OSD, not from all
peer OSDs, but this issue will be fixed separately.
This commit is contained in:
Vitaliy Filippov
2025-01-01 10:58:22 +03:00
parent d0396267d0
commit d9f9b0bca5
8 changed files with 416 additions and 252 deletions
+5 -4
View File
@@ -98,10 +98,11 @@ static const char* help_text =
"\n"
"vitastor-cli rm-data --pool <pool> --inode <inode> [--wait-list] [--min-offset <offset>]\n"
" Remove inode data without changing metadata.\n"
" --wait-list Retrieve full objects listings before starting to remove objects.\n"
" Requires more memory, but allows to show correct removal progress.\n"
" --min-offset Purge only data starting with specified offset.\n"
" --max-offset Purge only data before specified offset.\n"
" --wait-list Retrieve full objects listings before starting to remove objects.\n"
" Requires more memory, but allows to show correct removal progress.\n"
" --min-offset Purge only data starting with specified offset.\n"
" --max-offset Purge only data before specified offset.\n"
" --wait_up_timeout 16 Timeout for waiting until PGs are up in seconds.\n"
"\n"
"vitastor-cli merge-data <from> <to> [--target <target>]\n"
" Merge layer data without changing metadata. Merge <from>..<to> to <target>.\n"