Was this page helpful?
System configuration steps are performed automatically by the ScyllaDB RPM and deb packages. For information on getting started with ScyllaDB, see Getting Started.
All ScyllaDB AMIs and Docker images are pre-configured by a script with the following steps. This document is provided as a reference.
Several system configuration settings should be applied. For ease of use, the necessary scripts and configuration files are provided. Files are under dist/common and seastar/scripts in the ScyllaDB source code and installed in the appropriate system locations. (For information on ScyllaDB’s own configuration file, see ScyllaDB Configuration.)
File Name |
Description |
|---|---|
scylla.conf |
Remove system resource limits |
scylla-server |
Server startup options |
(written by |
Configure core dumps to use the |
The following scripts are available for you to run for configuring ScyllaDB. Some of these scripts are included in the scylla_setup script. This script is used for configuring ScyllaDB the first time, or when the system hardware changes.
perftune.py |
Configures various system parameters in order to improve the Seastar application performance |
|---|---|
scylla_bootparam_setup |
Sets the kernel options in the bootloader. In addition, it tunes Linux boot-time parameters for the node that ScyllaDB is running on (e.g. huge page setup). |
scylla_coredump_setup |
Sets up coredump facilities for ScyllaDB. This may include uninstalling existing crash reporting software for compatibility reasons. |
scylla_io_setup |
Benchmarks the disks and generates the io.conf and io_properties.yaml files. |
scylla_ntp_setup |
Configures Network Time Protocol |
scylla_prepare |
This script is run automatically every time ScyllaDB starts and the machine needs to be tuned. |
scylla_raid_setup |
Configures RAID and makes an XFS filesystem. |
scylla_save_coredump |
Compresses a core dump file (Ubuntu only) |
scylla_setup |
Sets up the ScyllaDB configuration. Many of these scripts are included in the setup script. |
scylla_stop |
Resets network mode if running in virtio or DPDK mode. |
scylla_swap_setup |
Configures a swap space on the host. |
scylla_sysconfig_setup |
Rewrites the /etc/sysconfig/scylla file. |
Script Name |
Description |
|---|---|
node_health_check |
Gathers metrics and information on the node, checking that the node is configured correctly. |
scylla-blocktune |
Tunes the filesystem and block layer (e.g. block size I/O scheduler configuration) for ScyllaDB. |
scylla_cpuscaling_setup |
Configures the CPU frequency scaling (IOW, puts the CPU in “performance” mode, instead of the slower “powersave” mode). |
scylla_cpuset_setup |
Configures which CPUs the ScyllaDB server threads run on. |
scylla_fstrim |
Runs |
scylla_fstrim_setup |
Configures a job so that |
scylla-housekeeping |
Checks if there are new versions of ScyllaDB available, and also shares some telemetry information for us to keep track of what versions are installed on the field. |
scylla_rsyslog_setup |
Configures the “rsyslog” service, which is used to send logs to a remote server. |
scylla_selinux_setup |
Disables SELinux for ScyllaDB. |
Note
It’s important to keep I/O scheduler configuration in sync on nodes with the same hardware. That’s why we recommend skipping running scylla_io_setup when provisioning a new node with exactly the same hardware setup as existing nodes in the cluster.
/etc/scylla.d/io.conf
/etc/scylla.d/io_properties.yaml
Using different I/O scheduler configuration may result in unnecessary bottlenecks.
If ScyllaDB is installed on an Amazon AMI, the bootloader should provide the clocksource=tsc and tsc=reliable options. This enables an accurate, high-resolution Time Stamp Counter (TSC) for setting the system time.
This configuration is provided in the file /usr/lib/scylla/scylla_bootparam_setup.
Remove the apport-noui or abrt packages if present, and set up a location and file name pattern for core dumps.
This configuration is provided in the file /usr/lib/scylla/scylla_bootparam_setup.
It is highly recommended to enforce time synchronization between ScyllaDB servers.
Run ntpstat on all nodes to check that system time is synchronized. If you are running in a virtualized environment and your system time is set on the host, you may not need to run NTP on the guest. Check the documentation for your platform.
If you have your own time servers shared with an application using ScyllaDB, use the same NTP configuration as for your application servers. The script /usr/lib/scylla/scylla_ntp_setup provides sensible defaults, using Amazon NTP servers if installed on the Amazon cloud, and other pool NTP servers otherwise.
Setting the file system to XFS is the most important and mandatory for production. ScyllaDB will significantly slow down without it.
The script /usr/lib/scylla/scylla_raid_setup performs the necessary RAID configuration and XFS filesystem creation for ScyllaDB.
Arguments to the script are
-d specify disks for RAID
-r MD device name for RAID
-u update /etc/fstab for RAID
On the ScyllaDB AMI, the RAID configuration is handled automatically in the /usr/lib/scylla/scylla_prepare script.
When installing ScyllaDB, it is highly recommended to use the scylla_setup script. ScyllaDB should not share CPUs with any CPU consuming process. In addition, when running ScyllaDB on AWS, we recommend pinning all NIC IRQs to CPU0 (due to the same reason). As a result, ScyllaDB should be prevented from running on CPU0 and its hyper-threading siblings. To verify that ScyllaDB is pinning CPU0, use the command below: If the node has four or fewer CPUs, don’t use this option.
To verify:
cat /etc/scylla.d/cpuset.conf
Example output:
--cpuset `1-15,17-31`
To configure networking, we recommend running the
perftune.pyscript with the following parameters:
--tune net
--nic <your network interface name>
See Seastar Perftune for details.
Configuration for ScyllaDB itself is in the ScyllaDB Configuration section of the administration guide.
The following item is not required in production.
When working on DPDK support for ScyllaDB, enable hugepages.
NR_HUGEPAGES=128
mount -t hugetlbfs -o pagesize=2097152 none /mnt/huge
mount -t hugetlbfs -o pagesize=2097152 none /dev/hugepages/
for n in /sys/devices/system/node/node?; do
echo $NR_HUGEPAGES > $n/hugepages/hugepages-2048kB/nr_hugepages;
done
Huge page configuration is written to /etc/sysconfig/scylla-server by the script /usr/lib/scylla/sysconfig_setup
System Limits - outlines the system limits which should be set or removed.
Was this page helpful?