Runtime Tuning
Overview
RabbitMQ runs on the Erlang virtual machine and runtime. A compatible version of Erlang must be installed in order to run RabbitMQ.
The Erlang runtime includes a number of components used by RabbitMQ. The most important ones as far as this guide is concerned are
- The Erlang virtual machine executes the code
epmd
resolves node names on a host to an inter-node communication port
This guide will focus on the virtual machine. For an overview of epmd
, please refer to the
Networking guide.
Topics covered include:
- How to configure Erlang VM settings for RabbitMQ nodes
- Runtime schedulers, what they are, how they relate to CPU cores, and so on
- Runtime thread activity metrics: where is scheduler and CPU time spent
- Runtime features that affect CPU utilisation
- How to reduce CPU utilisation on moderately or lightly loaded nodes
- Memory allocator settings
- Open file handle limit
- Inter-node communication buffer size
- Erlang process limit
- Erlang crash dumps
- Atom usage
VM Settings
The Erlang VM has a broad range of options that can be configured that cover process scheduler settings, memory allocation, garbage collection, I/O, and more. Tuning of those flags can significantly change runtime behavior of a node.
Configuring Flags
Most of the settings can be configured using environment variables. A few settings have dedicated variables, others can only be changed using the following generic variables that control what flags are passed by RabbitMQ startup scripts to the Erlang virtual machine.
The generic variables are
RABBITMQ_SERVER_ERL_ARGS
allows all VM flags to be overridden, including the defaults set by RabbitMQ scriptsRABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
allows a set of flags to be appended to the defaults set by RabbitMQ scriptsRABBITMQ_CTL_ERL_ARGS
controls CLI tool VM flags
In most cases RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
is the recommended option. It can be used to override defaults
in a safe manner. For example, if an important flag is omitted from RABBITMQ_SERVER_ERL_ARGS
, runtime performance
characteristics or system limits can be unintentionally affected.
As with other environment variables used by RabbitMQ, RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
and friends can be set using a separate environment variable file.
CPU Utilisation
CPU utilisation is a workload-specific topic. Generally speaking, when a workload involves more queues, connections and channels than CPU cores, all cores will be used without any configuration necessary.
The runtime provides several features that control how the cores are used.
Runtime Schedulers
Schedulers in the runtime assign work to kernel threads that perform it. They execute code, perform I/O, execute timers and so on. Schedulers have a number of settings that can affect overall system performance, CPU utilisation, latency and other runtime characteristics of a node.
By default the runtime will start one scheduler for one CPU core it detects. Starting with Erlang 23, this takes CPU quotas into account in containerized environments such as Docker and Kubernetes.
The number of schedulers can be explicitly set using the +S
flag. The following example configures
the node to start 4 schedulers even if it detects more cores to be available to it:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+S 4:4"
Most of the time the default behaviour works well. In shared or CPU constrained environments (including containerised ones), explicitly configuring scheduler count may be necessary.
CPU Resource Contention
The runtime assumes that it does not share CPU resources with other tools or tenants. When that's the case, the scheduling mechanism used can become very inefficient and result in significant (up to several orders of magnitude) latency increase for certain operations.
This means that in most cases colocating RabbitMQ nodes with other tools or applying CPU time slicing is highly discouraged and will result in suboptimal performance.
Scheduler Busy Waiting
The runtime can put schedulers to sleep when they run out of work to execute. There's a certain cost to bringing them back online, so with some workloads it may be beneficial to not do that.
This can be compared to a factory with multiple conveyor belts. When one belt runs out of items, it can be stopped. However, once more work is there for it to do, restarting it will take time. Alternatively the conveyor can be speculatively kept running for a period of time.
By default, RabbitMQ nodes configure runtime schedulers to speculatively wait for a short period
of time before going to sleep. Workloads where there can be prolonged periods of inactivity
can choose to turn off this speculative busy waiting using the +sbwt
and related runtime flags:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+sbwt none +sbwtdcpu none +sbwtdio none"
This can also reduce CPU usage on systems with limited or burstable CPU resources.
In order to determine how much time schedulers spend in busy wait, consult thread activity metrics.
Busy waiting will usually be accounted as system time in the output of tools such as top
and pidstat
.
Scheduler-to-CPU Core Binding
The number of schedulers won't always match the number of CPU cores available and the number of CPU cores does not necessarily correlate to the number of hardware threads (due to hyperthreading, for example). As such the runtime has to decide how to bind scheduler binding to hardware threads, CPU cores and NUMA nodes.
There are several binding strategies available. Desired strategy can be specified using the
RABBITMQ_SCHEDULER_BIND_TYPE
environment variable or using the +stbt
runtime flag
value.
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+stbt nnts"
RABBITMQ_SCHEDULER_BIND_TYPE="nnts"
Note that the strategy will only be effective if the runtime can detect CPU topology in the given environment.
Valid values are:
db
(used by default, alias fortnnps
in current Erlang release series)tnnps
nnts
nnps
ts
ps
s
ns
See VM flag documentation for more detailed descriptions.
Reducing CPU Usage
CPU usage is by definition very workload-dependent metric. Some workloads naturally use more CPU resources. Others use disk-heavy features such as quorum queues, and if disk I/O throughput is insufficient, CPU resources will be wasted while nodes are busy waiting for I/O operations to complete.
A couple of general recommendations can be applied to "moderately loaded" systems where a large percentage or most connections and queues can go idle from time to time. Put differently, in this section we consider any system that's not hovering around its peak capacity to be "moderately loaded".
Such system often can reduce their CPU footprint with a few straightforward steps. These recommendations can significantly decrease CPU footprint with some workloads: consider this community case for example.
Collect Runtime Thread Statistics
Collect runtime thread activity stats data to understand how scheduler and CPU time is spent. This is a critically important step for making informed decisions.
Turn off Speculative Scheduler Busy Waiting
Turn off speculative scheduler busy waiting using the +sbwt
and related runtime flags:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+sbwt none +sbwtdcpu none +sbwtdio none"
Speculative busy waiting usually not productive on moderately loaded systems.
Reduce Statistics Emission Frequency (Increase the Statistics Emission Interval)
Increase statistics emission interval from default 5 seconds to 15 or 30 seconds. This will reduce periodic activity that all connections, channels and queues carry out, even if they would otherwise be idle as far as client operations go. With most monitoring tools such monitoring frequency would be sufficient or even optimal.
Thread Statistics
RabbitMQ CLI tools provide a number of metrics that make it easier to reason about runtime thread activity.
rabbitmq-diagnostics runtime_thread_stats
is the command that produces a breakdown of how various threads spend their time.
The command's output will produce a table with percentages by thread activity:
emulator
: general code executionport
: external I/O activity (socket I/O, file I/O, subprocesses)gc
: performing garbage collectioncheck_io
: checking for I/O eventsother
,aux
: busy waiting, managing timers, all other taskssleep
: sleeping (idle state)
Significant percentage of activity in the external I/O state may indicate that the node and/or clients have maxed out network link capacity. This can be confirmed by infrastructure metrics.
Significant percentage of activity in the sleeping state might indicate a lightly loaded node or suboptimal runtime scheduler configuration for the available hardware and workload.
Memory Allocator Settings
The runtime manages (allocates and releases) memory. Runtime memory management is a complex topic with many tunable parameters. This section only covers the basics.
Memory is allocated in blocks from areas larger pre-allocated areas called carriers. Settings that control carrier size, block size, memory allocation strategy and so on are commonly referred to as allocator settings.
Depending on the allocator settings used and the workload, RabbitMQ can experience memory fragmentation of various degrees. Finding the best fit for your workload is a matter of trial, measurement (metric collection) and error. Note that some degree of fragmentation is inevitable.
Here are the allocator arguments used by default:
RABBITMQ_DEFAULT_ALLOC_ARGS="+MBas ageffcbf +MHas ageffcbf +MBlmbcs 512 +MHlmbcs 512 +MMmcs 30"
Instead of overriding RABBITMQ_DEFAULT_ALLOC_ARGS
, add flags that should be overridden to RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
. They
will take precedence over the default ones. So a node started with the following RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
value
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+MHlmbcs 8192"
will use in the following effective allocator settings:
"+MBas ageffcbf +MHas ageffcbf +MBlmbcs 512 +MHlmbcs 8192 +MMmcs 30"
For some workloads a larger preallocated area reduce allocation rate and memory fragmentation.
To configure the node to use a preallocated area of 1 GB, add +MMscs 1024
to VM startup arguments
using RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+MMscs 1024"
The value is in MB. The following example will preallocate a larger, 4 GB area:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+MMscs 4096"
To learn about other available settings, see runtime documentation on allocators.
Open File Handle Limit
Most operating systems limit the number of file handles that can be opened at the same time. When an OS process (such as RabbitMQ's Erlang VM) reaches the limit, it won't be able to open any new files or accept any more TCP connections.
This limit is covered in detail in the Networking guide. Note that it cannot be configured using Erlang VM flags.
Inter-node Communication Buffer Size
Inter-node traffic between a pair of nodes uses a TCP connection with a buffer known as the inter-node communication buffer. Its size is 128 MB by default. This is a reasonable default for most workloads. In some environments inter-node traffic can be very heavy and run into the buffer's capacity. Other workloads where the default is not a good fit involve transferring very large (say, in hundreds of megabytes) messages that do not fit into the buffer.
In this case the value can be increased using the RABBITMQ_DISTRIBUTION_BUFFER_SIZE
environment variable
or the +zdbbl
VM flag.
The value is in kilobytes:
RABBITMQ_DISTRIBUTION_BUFFER_SIZE=192000
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+zdbbl 192000"
When the buffer is hovering around full capacity, nodes will log a warning
mentioning a busy distribution port (busy_dist_port
):
2019-04-06 22:48:19.031 [warning] <0.242.0> rabbit_sysmon_handler busy_dist_port <0.1401.0>
Increasing buffer size may help increase throughput and/or reduce latency.
Erlang Process Limit
The runtime has a limit on the number of Erlang processes ("lightweight threads") that can exist on a node. The default is about 1 million. In most environments this is sufficient with a wide safety margin.
Environments that have a particularly high number of concurrent connections or a very large number
of queues (say, hundreds of thousands) this limit might need adjusting. This is done using the
RABBITMQ_MAX_NUMBER_OF_PROCESSES
environment variable, which is a convenient way of
setting the +P
Erlang VM flag:
RABBITMQ_MAX_NUMBER_OF_PROCESSES=2000000
To set the flag directly, use the RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS
environment
variable:
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+P 2000000"
Atom Usage
Similarly to the Erlang Process Limit covered above, the runtime has a limit to the number of atoms that can exist on a node. RabbitMQ nodes use the default of 5 million. This limit should suffice for the majority of use cases. However, in environments with high quorum queues churn (such workloads are recommended against), the limit may need a bump.
In order to increase the limit, use the +t
runtime argument:
# sets the limit to 9M
RABBITMQ_SERVER_ADDITIONAL_ERL_ARGS="+t 9000000"
Or the RABBITMQ_MAX_NUMBER_OF_ATOMS
environment variable:
# sets the limit to 9M
RABBITMQ_MAX_NUMBER_OF_ATOMS=9000000
The value must be a power of 10.
Erlang Crash Dumps
When the runtime terminates abnormally, or is sent a SIGUSER1
signal, it will produce a local file
known as the crash dump.
The file contains certain runtime information at the time of termination. This file can be useful for debugging of certain types of problems. It can also get really large on nodes with a large memory footprint.
To disable crash dump files, set the ERL_CRASH_DUMP_BYTES
environment variable to 0.