...
Name (SNMP) | Default | Description & Examples |
|---|---|---|
cache.expirationTime | 600 | In seconds; defaults to 10 minutes. Set 0 to disable. How long to hold an object after its last access. |
cache.maxCacheableSize | 1048576 | In bytes, defaulting to 1 MB. The largest object that can be stored in the content cache. If increased to greater than 5 MB, then scsp.readBufferAllowance must be increased to the same value. |
cache.percentage | 10 | Percentage, 0-100; set 0 to disable. How much I/O buffer memory to reserve for the content cache, which improves access to active content by storing it in geographically proximate locations. The reserve is reported when the node starts up: 'MAIN ANNOUNCE: Memory allocation at startup.' For best performance, especially with writing named objects, do not disable the content cache unless directed by Support. |
cache.realmStaleTimeout | 600 | In seconds, 60 or higher. How long before the security user list cache for domains is cleared. Lower this value if user lists update frequently. |
| The user-defined chassis name. | |
cip.histogramInterval | 0.01 | In seconds. The histogram bucket bin size. |
cip.processes | 32 | The number of CIP processes to run. Ignored if cip.useCFilter is False. |
cip.queryMinimumTimeout | 0.0 | In seconds. The minimum CIP query session time. |
cip.queryTimeout | 0.03 | In seconds. How long after booting that the cluster will initially wait for node replication bids. Once the cluster is running, bid wait times are calculated dynamically based on response times. For clusters with network latency, the initial wait time may need to be increased until the cluster can correctly calibrate. |
cip.readBufferSize | 1048576 | In bytes. The size of the multicast UDP socket read buffer. This value is capped the Linux /proc/sys/net/core/rmem_max value, which may be set via sysctl.coreRMemMax. |
disk.defragBufferBytes | 0 | Size in bytes of the per-disk buffer allocated for bulk defragmentation operations. Disable bulk mode by setting to 0. |
disk.defragUntilPercentage | 0.8 | Ratio, 0.0-1.0. The portion of known unused space that, when untrapped, will stop the disk defrag process. |
disk.enableMultipath | false | [deprecated] Whether to enable support for Device Mapper Multipathing (DM-Multipath). Multipath support was dropped in Swarm 10.0. |
disk.encryptNewVolumes | false | Whether to encrypt new Swarm volumes. Enabling encryptNewVolumes means that any newly-formatted Swarm volume will be encrypted |
disk.encryptionCipher | aes-xts-plain64 | The encryption cipher to be used when setting encryption for new Swarm volumes. Supported values are aes-xts-plain64 and aes-cbc-essiv |
disk.encryptionHash | sha512 | The encryption hash algorithm to be used when setting encryption for new Swarm volumes. Supported values are sha256 and sha512. |
disk.encryptionIterationTime | 5000 | In seconds. The maximum amount of time to be spent while iterating to generate an internal LUKS key from a Swarm encryption key, which will be used when setting encryption for new Swarm volumes. |
disk.encryptionKeyPrimary |
| The mnemonic name of the encryption key to use for encrypting new Swarm volumes. Do not use quotes. For this key to be used, disk.encryptNewVolumes must be set to True. |
disk.encryptionKeySize | 512 | The size of the internal LUKS key to be used when setting encryption for new Swarm volumes. Supported values are 128, 256, and 512. |
disk.encryptionKeys | {} | A comma-separated list of mnemonic name and encryption key pairs, used for accessing encrypted Swarm volumes. |
disk.encryptionType | luks | The encrypted volume format type used when formatting new volumes. Supported values are: 'luks', 'luks1', 'luks2'. |
disk.ioErrorToRetire | 2 | Count. How many consecutive I/O errors (no more than disk.ioErrorWindow seconds between each error) that will force a volume to retire. |
disk.ioErrorTolerance | 200 | Count. How many I/O errors are tolerated, past which the volume is taken offline immediately. Swarm then marks the volume as Unavailable and initiates both the volume recovery process (FVR) and the erasure coding recovery process (ECR) to relocate all the volume's objects. |
disk.ioErrorWindow | 172800 | In seconds; defaults to 2 days. The length of time after which an I/O error is forgotten, if no other errors followed and the volume's state is OK. Works with disk.ioErrorToRetire to control when volumes are retired. The default values means that if more than one error occurs within 2 days, the volume is retired. |
disk.minGB | 64 | How many GB a device must have to be eligible for automatic storage volume assignment with volumes = all. Set to 0 to include all disk devices. |
disk.smudgesToRetire | 4 | How many soft errors (smudges) over the life of a volume will trigger Swarm to retire the volume. A soft error occurs when the health processor does not get the expected data when validating the object but the disk gave no explicit I/O (hard) error. Set to 0 to disable the automatic retire. |
disk.standbyTimeout | 360 | In seconds. How long until an idle disk spins down automatically. |
disk.trappedToTotalPercentage | 0.0001 | Ratio, 0.0-0.01. The portion, of trapped space to total space, below which, will stop the defrag process (0.0 for no limit). |
disk.volumes |
| Required. Specifies the volume storage devices for Swarm to use. Valid entries: all, or a space-separated list of Linux volume identifiers, such as /dev/sda, /dev/sdb. all (recommended) is required for hot plugging and lets Swarm to use all volumes larger than disk.minGB. If a node is shut down longer than disk.obsoleteTimeout, all of its volumes are stale and cannot be used unless you force a volume remount by adding the :k (keep) policy option modifier. To specify the size, add a modifier with units: vols1:100m vols2:250g. |
ec.inProgressConsolidationTimeout | 86400 | Time in seconds, 0 to disable. An 'in progress' multipart PATCH complete cannot be consolidated before this timeout. |
feeds.maxMem | 100000 | In bytes. The maximum memory allowed per feed, for queue management. |
health.startDelay | 900 | In seconds; defaults to 900 seconds (15 minutes). How long after a node starts up to begin Health Processor checking and recovery processes. This option creates a grace period for the remaining nodes to stabilize in the cluster, which is useful in situations in which an entire cluster must be shut down and restarted. |
license.url | <Swarm default 2T license> | The location and name of the Swarm license file, caringo/license.txt. Can be a pathname or a URL. To use the default 2 TB license, you must keep the default location. |
mdns.readBufferSize | 1048576 | In bytes. The size of the read buffer for the multicast UDP socket. |
network.gateway |
| Optional. The default gateway IP address in the subnet. Ignored unless network.ipAddress is set. |
network.ipAddress |
| The single static IP address for a node to use, or blank to use DHCP. |
network.iptablesFileUrl |
| Optional. Location (URL) of Linux firewall rules to apply. When specified, Swarm transmits the rules without validation to the 'iptables-restore' command before starting the storage node processes. |
network.netmask |
| Optional. Sets the IP network mask for a node. Ignored unless network.ipAddress is set. |
network.timeSource |
| Recommended. List of one or more NTP servers by IP address or by name if network.dnsServers is set. You must have at least one usable NTP server in order for the storage node to start. If you don't assign a value here, a list from *.pool.ntp.org will be generated. |
node.subcluster | default | Specifies the name of the subcluster to which the chassis belongs. Names can have no more than 16 characters and no special characters, such as quotes and hyphens. |
shutdown.gracePeriod | 120 | In seconds; defaults to 2 minutes. How long to allow ongoing SCSP requests to complete during shutdown. |
snmp.enabled | true | Master switch to enable or disable the SNMP daemon |
snmp.sysContact | Unspecified | The value for the SNMP system contact, SNMPv2-MIB::sysContact. Must be a valid email address in 7-bit USASCII in one of these forms: Name <email@domain> First Last <email@domain> |
snmp.sysLocation | Unspecified | The value for the SNMP system location, SNMPv2-MIB::sysLocation. |
snmp.sysName | Unspecified | The value for the SNMP system name, SNMPv2-MIB::sysName. |
...