# Last week in scylladb.git master (issue #208; 2023-12-10)

**URL:** <https://forum.scylladb.com/t/last-week-in-scylladb-git-master-issue-208-2023-12-10/1100>\
**Category:** ScyllaDB\
**Tags:** git-news\
**Created:** [December 10, 2023, 11:39am UTC](https://forum.scylladb.com/t/last-week-in-scylladb-git-master-issue-208-2023-12-10/1100 "2023-12-10T11:39:00Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![avikivity](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.scylladb.com/avikivity/32/38_2.png) [@avikivity](https://forum.scylladb.com/u/avikivity)\
**Post date:** [December 10, 2023, 11:39am UTC](https://forum.scylladb.com/t/last-week-in-scylladb-git-master-issue-208-2023-12-10/1100/1 "2023-12-10T11:39:00Z")

</div>

This short report brings to light some interesting commits to [scylladb.git master](https://github.com/scylladb/scylladb/) from the last week. Commits in the 01e54f5b12…d62a5fc60b range are covered.

There were 146 non-merge commits from 18 authors in that period. Some notable commits:

  

Commitlog [format](https://github.com/scylladb/scylladb/commit/60af2f3cb21914f855fd5cc1929629c113b7999b) has changed to individually checksum each disk sector. This allows discrimination between three types of sectors: those containing valid data, those corrupted, and those that contain data from a previous use of the commitlog segment file. This should reduce false-positive warnings about commitlog corruptions during replay. Note the format is not compatible with previous versions, so upgrades should flush memtables first.

Tablets are a new, experimental way of distributing data across the cluster. A table that uses tablets and has a materialized view will now [correctly](https://github.com/scylladb/scylladb/commit/4505a86f46dfafa039246369c46140a714efadbd) and [consistently](https://github.com/scylladb/scylladb/commit/300e549267b2f51d0afc25b11707ca1d23b441a9) pair base-table tablets with corresponding view-table tablets, reducing chances for losing consistency compared to vnode-based tables.

When tablets are migrated to a different node, they will [not prevent](https://github.com/scylladb/scylladb/commit/9c0f05efa1f43834a7a150e5e9a0e9f468edf8e5) other tablets from initiating migration, increasing concurrency.

It’s now possible to use [cmake](https://github.com/scylladb/scylladb/commit/c1d0baf11a01a4968ab598635840f699e6f366ef) to build ScyllaDB from the configure.py entry point.

The scylladb-kernel-conf package tunes the Linux kernel scheduler via sysfs to improve latency. These tunings were lost in Linux 5.13+ due to kernel changes. They are now [restored](https://github.com/scylladb/scylladb/commit/6eb9344cb3e37ce6809dde8a628419faac3763e6).

ScyllaDB can store changes to its own metadata using a separate commitlog, to prevent it waiting for user data to commit. This separate commitlog is now [mandatory](https://github.com/scylladb/scylladb/commit/c8ee7d4499cb5a15d3e66b51f2e8b6ee24e03791).

There is now a script to assist in [decoding](https://github.com/scylladb/scylladb/commit/50332f796ecd3b77b0b332397ae04f9951912f45) UUID-based sstable generation numbers.

ScyllaDB detects internal stalls using a stall detector. It can do so using a timer, or using a more hardware performance counter, which can also track kernel stalls. It now [sets up permissions](https://github.com/scylladb/scylladb/commit/f90c10260fbcf0d902480176555971780cbfbdff) for itself to use the performance counter.

The native nodetool implementation, `scylla nodetool`, now supports [more commands](https://github.com/scylladb/scylladb/commit/d62a5fc60beb9d0cec9c59c5a0027641199f2ea8): decommission, rebuild. removenode, logging level commands, move, and refresh. It is not yet used by default.

  

See you in the next issue of last week in scylladb.git master!