scylladb/scylladb

Scylla causes 'memory per shard too low' on <2GB RAM EC2 instance-type

Open

#2630 opened on Jul 25, 2017

View on GitHub
 (10 comments) (0 reactions) (0 assignees)C++ (11,972 stars) (1,169 forks)batch import
good_first_issue

Description

It seems we are not able to run our AMI on t2.micro(1GB RAM) / t2.small(2GB RAM).

Installation details Scylla version (or git commit hash): 1.7.3 Cluster size: OS (RHEL/CentOS/Ubuntu/AWS AMI): AWS AMI

Jul 25 09:20:19 ip-172-16-99-91 scylla[2192]: Scylla version 1.7.3-0.20170719.41b4055 starting ...
Jul 25 09:20:19 ip-172-16-99-91 scylla[2192]:  [shard 0] init - Only 476 MiB per shard; this is below the recommended minimum of 1 GiB/shard; terminating.Configure more memory (--memory option) or decrease shard count (--smp option).
Jul 25 09:20:19 ip-172-16-99-91 scylla[2192]:  [shard 0] seastar - Exiting on unhandled exception: std::runtime_error (configuration (memory per shard too low))
Jul 25 09:20:20 ip-172-16-99-91 systemd[1]: scylla-server.service: main process exited, code=exited, status=1/FAILURE

Contributor guide

Tech stack
cpp
Domain
databaseinfrastructure
Issue type
bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
C++Seastarmemory configuration
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
25
Research direction
Investigate the memory initialization code in ScyllaDB, particularly the check for memory per shard. Look into the memory and smp command line options to see if they can be adjusted to allow lower memory configurations. Review the relevant error message and the logic that terminates if memory per shard is below 1 GiB. Consider modifying the minimum threshold or allowing a flag to override the check for low memory instances.