Configuration

AIS configuration comprises:

Cluster-wide (global) configuration is protected, namely: checksummed, versioned, and safely replicated. In effect, global config defines cluster-wide defaults inherited by each node joining the cluster.

Local config includes:

1. node’s own hostnames (or IP addresses) and mountpaths (data drives);
2. optionally, names-and-values that were changed for this specific node. For each node in the cluster, the corresponding capability (dubbed config-override) boils down to:
   • inheriting cluster configuration, and optionally
   • optionally, locally overriding assorted inherited defaults (see usage examples below).

Majority of the configuration knobs can be changed at runtime (and at any time). A few read-only variables are explicitly marked in the source; any attempt to modify those at runtime will return “read-only” error message.

CLI

For the most part, commands to view and update (CLI, cluster, node) configuration can be found here.

The same document also contains a brief theory of operation, command descriptions, numerous usage examples and more.

Important: as an input, CLI accepts both plain text and JSON-formatted values. For the latter, make sure to embed the (JSON value) argument into single quotes, e.g.:

1
$ ais config cluster checksum.type='{"type":"md5"}'

However, plain-text updating is more common, e.g.:

1
$ ais config cluster log.level 4
2
3
$ ais config cluster log.modules <TAB-TAB>
4
transport    memsys       fs           ec           ios          backend      mirror       downloader   s3
5
ais          cluster      reb          stats        xs           space        dsort        etl          none
6
7
$ ais config cluster log.modules space,s3

To show the current cluster config in plain text and JSON:

1
$ ais config cluster log
2
PROPERTY         VALUE
3
log.level        4 (modules: space,s3)
4
log.max_size     4MiB
5
log.max_total    128MiB
6
log.flush_time   40s
7
log.stats_time   1m
8
log.to_stderr    false

And the same in JSON:

1
$ ais config cluster log --json
2
3
    "log": {
4
        "level": "540676",
5
        "max_size": "4MiB",
6
        "max_total": "128MiB",
7
        "flush_time": "40s",
8
        "stats_time": "1m",
9
        "to_stderr": false
10
    }

Note: some config values are read-only or otherwise protected and can be only listed, e.g.:

1
$ ais config cluster backend --json
2
    "backend": {"aws":{},"azure":{},"gcp":{},"oci":{}}

See also:

• Backend providers and supported backends
• Disable/Enable cloud backend at runtime
• Networking Model: three logical networks and additional intra-cluster data plane

Configuring for production

Configuring AIS cluster for production requires a careful consideration. First and foremost, there are assorted performance related recommendations.

Optimal performance settings will always depend on your (hardware, network) environment. Speaking of networking, AIS supports 3 (three) logical networks and will, therefore, benefit, performance-wise, if provisioned with up to 3 isolated physical networks or VLANs. The logical networks are:

• user (aka public)
• intra-cluster control
• intra-cluster data

with the corresponding JSON names, respectively:

• hostname
• hostname_intra_control
• hostname_intra_data

Example

1
$ ais config node <TAB-TAB>
2
3
p[ctfooJtb]   p[qGfooQSf]   p[KffoosQR]   p[ckfooUEX]   p[DlPmfooU]   t[MgHfooNG]   t[ufooIDPc]   t[tFUfooCO]   t[wSJfoonU]   t[WofooQEW]
4
p[pbarqYtn]   p[JedbargG]   p[WMbargGF]   p[barwMoEU]   p[OUgbarGf]   t[tfNbarFk]   t[fbarswQP]   t[vAWbarPv]   t[Kopbarra]   t[fXbarenn]
5
6
## in aistore, each node has "inherited" and "local" configuration
7
## choose "local" to show the (selected) target's disks and network
8
9
$ ais config node t[fbarswQP] local --json
10
{
11
    "confdir": "/etc/ais",
12
    "log_dir": "/var/log/ais",
13
    "host_net": {
14
        "hostname": "10.51.156.130",
15
        "hostname_intra_control": "ais-target-5.nvmetal.net",
16
        "hostname_intra_data": "ais-target-5.nvmetal.net",
17
        "port": "51081",
18
        "port_intra_control": "51082",
19
        "port_intra_data": "51083"
20
    },
21
    "fspaths": {"/ais/nvme0n1": "","/ais/nvme1n1": "","/ais/nvme2n1": ""},
22
    "test_fspaths": {
23
        "root": "",
24
        "count": 0,
25
        "instance": 0
26
    }
27
}

Multi-homing

All aistore nodes - both ais targets and ais gateways - can be deployed as multi-homed servers. But of course, the capability is mostly important and relevant for the targets that may be required (and expected) to move a lot of traffic, as fast as possible.

Building up on the previous section’s example, here’s how it may look:

1
$ ais config node t[fbarswQP] local host_net --json
2
{
3
    "host_net": {
4
        "hostname": "10.51.156.130, 10.51.156.131, 10.51.156.132",
5
        "hostname_intra_control": "ais-target-5.nvmetal.net",
6
        "hostname_intra_data": "ais-target-5.nvmetal.net",
7
        "port": "51081",
8
        "port_intra_control": "51082",
9
        "port_intra_data": "51083"
10
    },
11
}

Note: additional NICs can be added (or removed) transparently for users, i.e. without requiring (or causing) any other changes.

The example above may serve as a simple illustration whereby t[fbarswQP] becomes a multi-homed device equally utilizing all 3 (three) IPv4 interfaces

References

1. Networking Model: three logical networks and additional intra-cluster data plane
2. For Kubernetes deployment, please refer to a separate ais-k8s repository that also contains:
   • AIS/K8s Operator and its configuration-defining
   • resources.

Cluster and Node Configuration

The first thing to keep in mind is that there are 3 (three) separate, and separately maintained, pieces:

1. Cluster configuration that comprises global defaults
2. Node (local) configuration
3. Node’s local overrides of global defaults

Specifically:

Cluster Config

To show and/or change global config, simply type one of:

1
# 1. show cluster config
2
$ ais show cluster config
3
4
# 2. show cluster config in JSON format
5
$ ais show cluster config --json
6
7
# 3. show cluster-wide defaults for all variables prefixed with "time"
8
$ ais show config cluster time
9
# or, same:
10
$ ais show cluster config time
11
PROPERTY                         VALUE
12
timeout.cplane_operation         2s
13
timeout.max_keepalive            4s
14
timeout.max_host_busy            20s
15
timeout.startup_time             1m
16
timeout.send_file_time           5m
17
timeout.transport_idle_term      4s
18
19
# 4. for all nodes in the cluster set startup timeout to 2 minutes
20
$ ais config cluster timeout.startup_time=2m
21
config successfully updated

Typically, when we deploy a new AIS cluster, we use configuration template that contains all the defaults - see, for example, JSON template. Configuration sections in this template, and the knobs within those sections, must be self-explanatory, and the majority of those, except maybe just a few, have pre-assigned default values.

Node configuration

As stated above, each node in the cluster inherits global configuration with the capability to override the latter locally.

There are also node-specific settings, such as:

• log directories
• network configuration, including node’s hostname(s) or IP addresses
• node’s mountpaths

Since AIS supports n-way mirroring and erasure coding, we typically recommend not using LVMs and hardware RAIDs.

Example: show node’s configuration

1
# ais show config t[CCDpt8088]
2
PROPERTY                                 VALUE                                                           DEFAULT
3
auth.enabled                             false                                                           -
4
auth.secret                              **********                                                     -
5
backend.conf                             map[aws:map[] gcp:map[]]                                        -
6
checksum.enable_read_range               false                                                           -
7
checksum.type                            xxhash                                                          -
8
checksum.validate_cold_get               true                                                            -
9
checksum.validate_obj_move               false                                                           -
10
checksum.validate_warm_get               false                                                           -
11
...
12
...
13
(Hint: use `--type` to select the node config's type to show: 'cluster', 'local', 'all'.)
14
...
15
...

Example: same as above in JSON format:

1
$ ais show config CCDpt8088 --json | tail -20
2
    "lastupdate_time": "2021-03-20 18:00:20.393881867 -0700 PDT m=+2907.143584987",
3
    "uuid": "ZzCknLkMi",
4
    "config_version": "3",
5
    "confdir": "/ais",
6
    "log_dir": "/tmp/ais/log",
7
    "host_net": {
8
        "hostname": "",
9
        "hostname_intra_control": "",
10
        "hostname_intra_data": "",
11
        "port": "51081",
12
        "port_intra_control": "51082",
13
        "port_intra_data": "51083"
14
    },
15
    "fspaths": {"/ais/mp1": "","/ais/mp2": "","/ais/mp3": "","/ais/mp4": ""},
16
    "test_fspaths": {
17
        "root": "/tmp/ais",
18
        "count": 0,
19
        "instance": 0
20
    }

See also:

• local playground with two data drives

###Example: use --type option to show only local config

1
# ais show config koLAt8081 --type local
2
PROPERTY                         VALUE
3
confdir                          /ais
4
log_dir                          /tmp/ais/log
5
host_net.hostname
6
host_net.hostname_intra_control
7
host_net.hostname_intra_data
8
host_net.port                    51081
9
host_net.port_intra_control      51082
10
host_net.port_intra_data         51083
11
fspaths.paths                    /ais/mp1,/ais/mp2,/ais/mp3,/ais/mp4
12
test_fspaths.root                /tmp/ais
13
test_fspaths.count               0
14
test_fspaths.instance            0

Local override (of global defaults)

Example:

1
$ ais show config t[CCDpt8088] timeout
2
# or, same:
3
$ ais config node t[CCDpt8088] timeout
4
5
PROPERTY                         VALUE   DEFAULT
6
timeout.cplane_operation         2s      -
7
timeout.join_startup_time        3m      -
8
timeout.max_host_busy            20s     -
9
timeout.max_keepalive            4s      -
10
timeout.send_file_time           5m      -
11
timeout.startup_time             1m      -
12
13
$ ais config node t[CCDpt8088] timeout.startup_time=90s
14
config for node "CCDpt8088" successfully updated
15
16
$ ais config node t[CCDpt8088] timeout
17
18
PROPERTY                         VALUE   DEFAULT
19
timeout.cplane_operation         2s      -
20
timeout.join_startup_time        3m      -
21
timeout.max_host_busy            20s     -
22
timeout.max_keepalive            4s      -
23
timeout.send_file_time           5m      -
24
timeout.startup_time             1m30s   1m

In the DEFAULT column above hyphen (-) indicates that the corresponding value is inherited and, as far as the node CCDpt8088, remains unchanged.

Rest of this document is structured as follows

• Basics
• Startup override
• Managing mountpaths
• Disabling extended attributes
• Enabling HTTPS
• Filesystem Health Checker
• Networking
• Curl examples
• CLI examples

The picture illustrates one section of the configuration template that, in part, includes listening port:

Further, test_fspaths section (see below) corresponds to a single local filesystem being partitioned between both local and Cloud buckets. In other words, the test_fspaths configuration option is intended strictly for development.

In production, we use an alternative configuration called fspaths: the section of the config that includes a number of local directories, whereby each directory is based on a different local filesystem solely utilizing one or more non shared disks.

For fspath and mountpath terminology and details, please see section Managing Mountpaths in this document.

An example of 12 fspaths (and 12 local filesystems) follows below:

Example: 3 NVMe drives

1
$ ais config node <TAB-TAB>
2
3
p[ctfooJtb]   p[qGfooQSf]   p[KffoosQR]   p[ckfooUEX]   p[DlPmfooU]   t[MgHfooNG]   t[ufooIDPc]   t[tFUfooCO]   t[wSJfoonU]   t[WofooQEW]
4
p[pbarqYtn]   p[JedbargG]   p[WMbargGF]   p[barwMoEU]   p[OUgbarGf]   t[tfNbarFk]   t[fbarswQP]   t[vAWbarPv]   t[Kopbarra]   t[fXbarenn]
5
6
## in aistore, each node has "inherited" and "local" configuration
7
## choose "local" to show the target's own disks and network
8
9
$ ais config node t[fbarswQP] local --json
10
{
11
    "confdir": "/etc/ais",
12
    "log_dir": "/var/log/ais",
13
    "host_net": {
14
        "hostname": "10.51.156.130",
15
        "hostname_intra_control": "ais-target-5.nvmetal.net",
16
        "hostname_intra_data": "ais-target-5.nvmetal.net",
17
        "port": "51081",
18
        "port_intra_control": "51082",
19
        "port_intra_data": "51083"
20
    },
21
    "fspaths": {"/ais/nvme0n1": "","/ais/nvme1n1": "","/ais/nvme2n1": ""},
22
    "test_fspaths": {
23
        "root": "",
24
        "count": 0,
25
        "instance": 0
26
    }
27
}

See also:

• local playground with two data drives

Basics

First, some basic facts:

• AIS cluster is a collection of nodes - members of the cluster.
• A node can be an AIS proxy (aka gateway) or an AIS target.
• In either case, HTTP request to read (get) or write (set) specific node’s configuration will have /v1/daemon in its URL path.
• The capability to carry out cluster-wide configuration updates is also supported. The corresponding HTTP URL will have /v1/cluster in its path.

Both daemon and cluster are the two RESTful resource abstractions supported by the API. Please see AIS API for naming conventions, RESTful resources, as well as API reference and details.

• To get the node’s up-to-date configuration, execute:

1
$ ais show config <daemon-ID>

This will display all configuration sections and all the named knobs - i.e., configuration variables and their current values.

Most configuration options can be updated either on an individual (target or proxy) daemon, or the entire cluster. Some configurations are “overridable” and can be configured on a per-daemon basis. Some of these are shown in the table below.

For examples and alternative ways to format configuration-updating requests, please see the examples below.

Following is a table-summary that contains a subset of all settable knobs:

NOTE (May 2022): this table is somewhat outdated and must be revisited.

Startup override

AIS command-line allows to override configuration at AIS node’s startup. For example:

1
$ aisnode -config=/etc/ais.json -local_config=/etc/ais_local.json -role=target -config_custom="client.timeout=13s,log.level=4"

As shown above, the CLI option in-question is: confjson. Its value is a JSON-formatted map of string names and string values. By default, the config provided in config_custom will be persisted on the disk. To make it transient either add -transient=true flag or add additional JSON entry:

1
$ aisnode -config=/etc/ais.json -local_config=/etc/ais_local.json -role=target -transient=true -config_custom="client.timeout=13s, transient=true"

Another example. To override locally-configured address of the primary proxy, run:

1
$ aisnode -config=/etc/ais.json -local_config=/etc/ais_local.json -role=target -config_custom="proxy.primary_url=http://G"
2
# where G denotes the designated primary's hostname and port.

To achieve the same on temporary basis, add -transient=true as follows:

1
$ aisnode -config=/etc/ais.json -local_config=/etc/ais_local.json -role=target -config_custom="proxy.primary_url=http://G"

Please see AIS command-line for other command-line options and details.

Managing mountpaths

• Mountpath - is a single disk or a volume (a RAID) formatted with a local filesystem of choice, and a local directory that AIS can fully own and utilize (to store user data and system metadata). Note that any given disk (or RAID) can have (at most) one mountpath (meaning no disk sharing) and mountpath directories cannot be nested. Further:
  • a mountpath can be temporarily disabled and (re)enabled;
  • a mountpath can also be detached and (re)attached, thus effectively supporting growth and “shrinkage” of local capacity;
  • it is safe to execute the 4 listed operations (enable, disable, attach, detach) at any point during runtime;
  • in a typical deployment, the total number of mountpaths would compute as a direct product of (number of storage targets) x (number of disks in each target).

Configuration option fspaths specifies the list of local mountpath directories. Each configured fspath is, simply, a local directory that provides the basis for AIS mountpath.

In regards non-sharing of disks between mountpaths: for development we make an exception, such that multiple mountpaths are actually allowed to share a disk and coexist within a single filesystem. This is done strictly for development convenience, though.

AIStore REST API makes it possible to list, add, remove, enable, and disable a fspath (and, therefore, the corresponding local filesystem) at runtime. Filesystem’s health checker (FSHC) monitors the health of all local filesystems: a filesystem that “accumulates” I/O errors will be disabled and taken out, as far as the AIStore built-in mechanism of object distribution. For further details about FSHC, please refer to FSHC readme.

Disabling extended attributes

To make sure that AIStore does not utilize xattrs, configure:

• checksum.type=none
• versioning.enabled=true, and
• write_policy.md=never

for all targets in AIStore cluster.

Or, simply update global configuration (to have those cluster-wide defaults later inherited by all newly created buckets).

This can be done via the common configuration “part” that’d be further used to deploy the cluster.

Extended attributes can be disabled on per bucket basis. To do this, turn off saving metadata to disks (CLI):

1
$ ais bucket props ais://mybucket write_policy.md=never
2
Bucket props successfully updated
3
"write_policy.md" set to: "never" (was: "")

Disable extended attributes only if you need fast and temporary storage. Without xattrs, a node loses its objects after the node reboots. If extended attributes are disabled globally when deploying a cluster, node IDs are not permanent and a node can change its ID after it restarts.

Enabling HTTPS

To switch from HTTP protocol to an encrypted HTTPS, configure net.http.use_https=true and modify net.http.server_crt and net.http.server_key values so they point to your TLS certificate and key files respectively (see AIStore configuration).

The following HTTPS topics are also covered elsewhere:

• Generating self-signed certificates
• Deploying: 4 targets, 1 gateway, 6 mountpaths, AWS backend
• Accessing the cluster
• Testing with self-signed certificates
• Updating and reloading X.509 certificates
• Switching cluster between HTTP and HTTPS

Filesystem Health Checker

Default installation enables filesystem health checker component called FSHC. FSHC can be also disabled via section “fshc” of the configuration.

When enabled, FSHC gets notified on every I/O error upon which it performs extensive checks on the corresponding local filesystem. One possible outcome of this health-checking process is that FSHC disables the faulty filesystems leaving the target with one filesystem less to distribute incoming data.

Please see FSHC readme for further details.

Networking

In addition to user-accessible public network, AIStore will optionally make use of the two other networks:

• intra-cluster control
• intra-cluster data

The way the corresponding config may look in production (e.g.) follows:

1
$ ais config node t[nKfooBE] local h... <TAB-TAB>
2
host_net.hostname                 host_net.port_intra_control       host_net.hostname_intra_control
3
host_net.port                     host_net.port_intra_data          host_net.hostname_intra_data
4
5
$ ais config node t[nKfooBE] local host_net --json
6
7
    "host_net": {
8
        "hostname": "10.50.56.205",
9
        "hostname_intra_control": "ais-target-27.ais.svc.cluster.local",
10
        "hostname_intra_data": "ais-target-27.ais.svc.cluster.local",
11
        "port": "51081",
12
        "port_intra_control": "51082",
13
        "port_intra_data": "51083"
14
    }

The fact that there are 3 logical networks is not a “limitation” - not a requirement to specifically have 3. Using the example above, here’s a small deployment-time change to run a single one:

1
    "host_net": {
2
        "hostname": "10.50.56.205",
3
        "hostname_intra_control": "ais-target-27.ais.svc.cluster.local",
4
        "hostname_intra_data": "ais-target-27.ais.svc.cluster.local",
5
        "port": "51081",
6
        "port_intra_control": "51081,   # <<<<<< notice the same port
7
        "port_intra_data": "51081"      # <<<<<< ditto
8
    }

Ideally though, production clusters are deployed over 3 physically different and isolated networks, whereby intense data traffic, for instance, does not introduce additional latency for the control one, etc.

Separately, there’s a multi-homing capability motivated by the fact that today’s server systems may often have, say, two 50Gbps network adapters. To deliver the entire 100Gbps without LACP trunking and (static) teaming, we could simply have something like:

1
    "host_net": {
2
        "hostname": "10.50.56.205, 10.50.56.206",
3
        "hostname_intra_control": "ais-target-27.ais.svc.cluster.local",
4
        "hostname_intra_data": "ais-target-27.ais.svc.cluster.local",
5
        "port": "51081",
6
        "port_intra_control": "51082",
7
        "port_intra_data": "51083"
8
    }

No other changes. Just add the second NIC - second IPv4 addr 10.50.56.206 above, and that’s all.

Curl examples

The following assumes that G and T are the (hostname:port) of one of the deployed gateways (in a given AIS cluster) and one of the targets, respectively.

Cluster-wide operation (all nodes)

• Set the stats logging interval to 1 second

1
$ curl -i -X PUT -H 'Content-Type: application/json' -d '{"action": "set-config","name": "periodic.stats_time", "value": "1s"}' 'http://G/v1/cluster'

or, same:

1
$ curl -i -X PUT 'http://G/v1/cluster/set-config?periodic.stats_time=1s'

Notice the two alternative ways to form the requests.

Cluster-wide operation (all nodes)

• Set the stats logging interval to 2 minutes

1
$ curl -i -X PUT -H 'Content-Type: application/json' -d '{"action": "set-config","name": "periodic.stats_time", "value": "2m"}' 'http://G/v1/cluster'

Cluster-wide operation (all nodes)

• Set the default number of n-way copies to 4 (can still be redefined on a per-bucket basis)

1
$ curl -i -X PUT -H 'Content-Type: application/json' -d '{"action": "set-config","name": "mirror.copies", "value": "4"}' 'http://G/v1/cluster'
2
3
# or, same using CLI:
4
$ ais config cluster mirror.copies 4

Single-node operation (single node)

• Set log level = 1

1
$ curl -i -X PUT -H 'Content-Type: application/json' -d '{"action": "set-config","name": "log.level", "value": "1"}' 'http://T/v1/daemon'
2
# or, same:
3
$ curl -i -X PUT 'http://T/v1/daemon/set-config?log.level=1'
4
5
# or, same using CLI (assuming the node in question is t[tZktGpbM]):
6
$ ais config node t[tZktGpbM] log.level 1

CLI examples

AIS CLI is an integrated management-and-monitoring command line tool. The following CLI command sequence, first - finds out all AIS knobs that contain substring “time” in their names, second - modifies list_timeout from 2 minutes to 5 minutes, and finally, displays the modified value:

1
$ ais show config p[rZTp8080] --type all --json | jq '.timeout.list_timeout'
2
"2m"
3
4
$ ais config cluster timeout.list_timeout=5m
5
Config has been updated successfully.
6
7
$ ais show config p[rZTp8080] --type all --json | jq '.timeout.list_timeout'
8
"5m"

The example above demonstrates cluster-wide configuration update but note: single-node updates are also supported.

Cluster-wide operation (all nodes)

• Set periodic.stats_time = 1 minute, periodic.iostat_time_long = 4 seconds

1
$ ais config cluster periodic.stats_time=1m disk.iostat_time_long=4s

Single-node operation (single node)

AIS configuration includes a section called disk. The disk in turn contains several knobs - one of those knobs is disk.iostat_time_long, another - disk.disk_util_low_wm. To update one or both of those named variables on all or one of the clustered nodes, you could:

• Set disk.iostat_time_long = 3 seconds, disk.disk_util_low_wm = 40 percent on daemon with ID target1

1
$ ais config node target1 periodic.stats_time=1m disk.iostat_time_long=4s