Stop K/V listings at the end of their range #111

Closed

koolkhel wants to merge 2 commits from koolkhel/vitastor:kv-list-end into master

Author	SHA1	Message	Date
Yury Luneff	c3303a4122	Stop NFS prefix scans at the end of the prefix The emptiness checks of RMDIR and of RENAME over a directory, READDIR and the defrag scans list a key prefix. They now stop at the end of the prefix instead of running on to the next existing key through blocks emptied by deletes, which made RMDIR slower with every deleted file. This also fixes RENAME of a directory over an empty directory failing with ENOTEMPTY: its check took the first key after the prefix, which belongs to another directory or inode, for an entry.	2026-10-27 10:15:27 +03:00
Yury Luneff	bf89e7e0d2	Add an end key to K/V listings A listing goes on from block to block until it finds a key, and blocks emptied by deletes are never freed. So a listing that runs past the end of its range reads every empty block up to the next existing key, each with a version check on an OSD. After many creates and deletes, this made a single RMDIR in the NFS proxy take hundreds of milliseconds. list_start(start, end) stops at the first key >= end and does not read blocks beyond it. vitastor-kv list/dump/dumpjson now pass their end argument.	2026-10-27 10:15:27 +03:00

Author

SHA1

Message

Date

Yury Luneff

c3303a4122

Stop NFS prefix scans at the end of the prefix

The emptiness checks of RMDIR and of RENAME over a directory, READDIR
and the defrag scans list a key prefix. They now stop at the end of the
prefix instead of running on to the next existing key through blocks
emptied by deletes, which made RMDIR slower with every deleted file.

This also fixes RENAME of a directory over an empty directory failing
with ENOTEMPTY: its check took the first key after the prefix, which
belongs to another directory or inode, for an entry.

2026-10-27 10:15:27 +03:00

Yury Luneff

bf89e7e0d2

Add an end key to K/V listings

A listing goes on from block to block until it finds a key, and blocks
emptied by deletes are never freed. So a listing that runs past the end
of its range reads every empty block up to the next existing key, each
with a version check on an OSD. After many creates and deletes, this
made a single RMDIR in the NFS proxy take hundreds of milliseconds.

list_start(start, end) stops at the first key >= end and does not read
blocks beyond it. vitastor-kv list/dump/dumpjson now pass their end
argument.

2026-10-27 10:15:27 +03:00

Stop K/V listings at the end of their range #111

2 Commits