Nexus System Settings
Jump to Section
Notes:
-
Config TAGNames are case sensitive
-
Also, see sample Nexfs cloud and object configurations
-
Nexfscli or the Admin Web Console for Configuration Management is recommended
Service Managment
Config TagName
Live Update
or Restart
Allowed
Values
Description
NEXFSPAUSED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If set to 1, Nexfs will stop accepting file opens/creates/reads and writes, which does not affect the job scheduler or bg migrations.
It may not be possible to unpause nexfs using the standard liveconfig update methods; this occurs when the request queue becomes full with blocked with write requests. If this occurs, Nexfs can be unblocked by creating the special file /tmp/unpausenexfs on the nexfs server, for example, 'touch /tmp/unpausenexfs' Note nexfs will only look for this file during a standard job schedular cycle; as such, it will take upto JOBSCHEDULARSLEEP after creating the special file and Nexfs unpausing. Also, note that nexfs will not remove the special file from /tmp; that must be done manually.
BGMIGRATIONSPAUSED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If set to 1 background nexfs data migrations between storage tiers will be paused
DELETEPAUSED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If background deletions can run or should be paused
See “File Deletion Management” for information on actions taken when a file is deleted.
SMARTPROTECTENABLED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
Set to 1 to enabled Smart Protect. Affects SMARTPROTECT and SMARTPROTECTOPENFILECYCLE
See SmartProtectAndTier for details of Nexfs Smart Protection
SMARTTIERENABLED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
Set to 1 to enabled Smart Tiering. Affects SMARTTIER and SMARTTIEROPENFILECYCLE
See SmartProtectAndTier for details of Nexfs Smart Tiering
JOBSCHEDULARPAUSED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If set to 1 the internal nexfs job scheduler will be paused
Tier 3 Cloud and Object Storage Connection Settings
Config TagName
Live Update
or Restart
Allowed
Values
Description
T3S3URL
Live
Default: amazonaws.com
A valid URL to be used to connect to the object or cloud storage
URL for accessing T3 S3 compatible storage
For AWS this does not include the region.
T3S3REGION
Live
Default: us-west-1
Valid values: Any valid region.
Note this setting is not validated
The AWS S3 Region to connect to, or equivalent S3 compatible storage region.
T3S3ENABLEREGION
Live
Default:1
Valid values: 0 (no) or (1) Yes
If to include the region configured by T3S3REGION in the final URL used to connect to the object or cloud storage
T3S3BUCKET
Live
Default: nexfs
Valid values: Any existing Cloud or object storage bucket
The S3 or compatible storage bucket to use.
Note: The bucket configured here must already exist before it can be used by Nexfs. Nexus will NOT create the bucket if it does not exist.
T3S3PORT
Live
Default: 9000
Valid values: a TCP port between 0 and 16384
The TCP port us when connecting to the cloud or object storage.
Port 9000 is the default port used by minio.
T3ALLOWINVALIDSSL
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
Validate the T3 destination servers SSL certificate. (0) Reject the server if invalid, (1) Allow the connection even if the certificate is not valid
T3USEHTTPS
Live
Default: 1
Valid Values: 0 (no) or 1 (yes)
If set to 1 then nexfs connects to the S3 storage system over encrypted HTTPS
If set to 0 then nexus connect using basic HTTP
T3S3MAXREQTIME
Live
Default: 120
Valid values: 5 to 3600
The maximum number of seconds a S3 request can take before it is aborted
Note: Slow (progressing) requests will also be aborted after this set number of seconds
T3DDIRENABLED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If to enabled T3 Storage
Currently, Nexfs supports Amazon AWS S3 and S3 API compatible Object and Cloud Storage.
Nexus releases are tested against AWS S3 cloud storage and Minio Object storage
T3S3MAXCONNECTTIMEOUT
Live
Default: 15 (seconds)
Valid values: 15 to 180
How long to before timing out a new connection to the cconfigured Tier3 S3 Storage
T3AWSSecretAccessKey
Live
Default: minioadmin
Valid value: the correct Secret Key to use to connect to the configured bucket
The Secret key that has been configured against the cloud or object storage system.
T3AWSAccessKeyId
Live
Default: minioadmin
Valid value: the correct Access Key to use to connect to the configured bucket
The Access key which has been configured against the cloud or object storage system.
T3S3USEVIRTUALHOST
Live
Default: 1
Valid values:
0 (no) or 1 (yes)
If set to 1 then AWS Virtual Host style URLs are used to connect to the storage system.
If set to 0 then the older AWS Path Host style is used.
T3S3COMPATIBLE
(Read description before changing, it likely this should still be set to 0 when connecting to s3 compatible storage systems)
Live
Default: 0
Valid value: 0 or 1
This setting should be set to 0 when connecting to most s3 compatible storage systems.
When set to 0 the word “s3” is included in the final URL used to connect t the cloud or object storage, for example when set to 0 the URL format will be:
https://bucketname.s3.Region.amazonaws.com
When set to one it will be
https://bucketname.Region.amazonaws.com
This setting may need to be set to 1 when connecting to some S3 compatible storage systems.
T3S3RETRIES
Live
Default: 3
Valid values: 0 to 12
How many times to retry a request that fails to the S3 cloud or object storage.
T3S3SIGNATUREV4
Live
Default: 1
Valid values: 0 (no) or 1 (yes)
Use T3 AWS Signature Version 4 (otherwise V2) to connect to the S3 storage endpoint
T3S3RETRYSLEEP
Live
Default: 1
Valid values: 0 to 360
The number of seconds to wait for retiring a failed request to S3 cloud or object storage.
T3S3RETRY404
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If to retry an S3 request when nexus received an http 404 from the storage system.
An http 404 means that the storage system is stating the requested object or cloud file does not exist. Some storage systems may return a 404 if an object is created soon after it has been created, or if that storage system is too busy to serve the request. If this happens nexfs could believe that the object does not exist and create an empty one instead of returning the expected data. This behaviour does not happen with AWS S3, but can occur with Swarm object storage, you should check with your storage object or cloud storage vendor whether this should setting should be enabled or not.
T3S3RETRY404SLEEP
Live
Default: 1 (ms)
Valid values: 0 to 999999
The time in ms to wait before retring 404 T3 S3 request, a value > 0 will impact performance.
T3S3ENABLEMD5DIGEST
Live
Default: 0 Valid values: 0 (off) or 1 (on)
Enable MD5 Digest for Tier-3 S3 data transfers
Tier 1 and Tier 2 Storage Configuration
Config TagName
Live Update
or Restart
Allowed
Values
Description
T1SDIR
Live
Default: /nexfs/s1
A existing path up to a max 2048 characters
The path to the mountpoint/folder that nexus will use to storage tier-1 file structure information, must be a full path starting with /, the path can contain a max of 2048 characters.
T1DDIR
Live
Default: /nexfs/t1
A path up to a max 2048 characters
The path to the mountpoint/folder that nexus will use to storage tier-1 data, must be a full path starting with /, the path can contain a max of 2048 characters.
T1DDIRENABLED
Live
Default: 1
Valid values: 0 (no) or 1 (yes)
If to enabled T1 Storage
Nexfs requires at least either T1 or T2 data storage to be enabled.
T2DDIR
Live
Default: /nexfs/t2
A path up to a max 2048 characters
The path to the mountpoint/folder that nexus will use to storage tier-2 data, must be a full path starting with /, the path can contain a max of 2048 characters.
While available in public preview releases, Enabling T2 storage otherwise requires a Nexfs enterprise license, subscription or a non expired time-limited trial license to be enabled in general releases.
T2DDIRENABLED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If to enabled T2 Storage
Nexfs requires at least either T1 or T2 data storage to be enabled.
While available in public preview releases, Enabling T2 storage otherwise requires a Nexfs enterprise license, subscription or a non expired time-limited trial license to be enabled in general releases.
T2SREPLICATIONMODE
Live
Default: 0 Valid values: 0 (disabled), 1=async, 2=sync
Replication of T1 structure index information mode. 0=disabled, 1=asynchronous 2=synchronous
T2SDIR
Live
Default: /nexfs/s2
A existing path up to a max 2048 characters
The path to the mountpoint/folder that nexus will use to replicate tier-1 file structure information, must be a full path starting with /, the path can contain a max of 2048 characters.
Tier Capacity Management
Config TagName
Live Update
or Restart
Allowed
Values
Description
NFS Management Configuration
Config TagName
Live Update
or Restart
Allowed
Values
Description
iSCSI Management Configuration
Config TagName
Live Update
or Restart
Allowed
Values
Description
General Configuration
Config TagName
Live Update
or Restart
Allowed
Values
Description
Background Migration Settings
Config TagName
Live Update
or Restart
Allowed
Values
Description
Deletion and Storage Cleanup Settings
Config TagName
Live Update
or Restart
Allowed
Values
Description
Smart Protect and Tier
Config TagName
Live Update
or Restart
Allowed
Values
Description
Job Management
Config TagName
Live Update
or Restart
Allowed
Values
Description
JOBSCHEDULARPAUSED
Live
Default: 0
Valid values: 0 (no) or 1 (yes)
If set to 1 the internal nexfs job scheduler will be paused
JSMAXQUEUESIZE
Restart
Default: 10000
Valid values:
Min: 1000
Max: 100000
The size of the Jobqueue, when full Nexfs will start to recycle to the beginning of the queue by removing completed tasks freeing queue slots. If the job queue is full with incomplete tasks, then no new jobs can be submitted; some actions will wait for the queue to become available. Others may run the job in real-time, delaying operation completion and increasing IO latency.
JSMAXTHREADS
Restart
Default: 4
Valid values: 1 to 16
The maximum number of Job Scheduler Threads that can run concurrently
JOBSCHEDULARSLEEP
Live
Default 60
Valid values: 0 to 3600
Seconds the internal jobs and bg migration schedulers will sleep between determining if any jobs should be started
JOBSCHEDULARSALLOWMANUALWHENPAUSED
Live
Default: 0
Valid values: 0 or 1
If set to 1 then manual “admin” jobs can still be run when the job scheduler is paused
Security
Config TagName
Live Update
or Restart
Allowed
Values
Description
ENFORCEPOSIXACLS
Restart
Default: 1
Valid values: 0 (no) or 1 (yes)
Enforce POSIX ACL (file permission checking), restricting access based on a files POSIX access control
ROOTONLYACCESS
Restart
Default: 0
Valid values: 0 (no) or 1 (yes)
If enabled only the root user can access the nexfs filesystem, or if access is allowed for all users
(Requires nexfs to be started using nexfscli)"
SECURITYCMDGROUP
Live
Default: 0 (root)
Valid values: 0 to 65536
User Group whose members can run nexfs admin commands
SECURITYCMDGROUPACCESS
Live
Default: 2
Valid values: 0=no access, 1 read-only, 2 read/write
The access level nexfs cmd group members have for running admin commands
SECURITYCONFGROUP
Live
Default: 0 (root)
Valid values: 0 to 65536
User Group whose members can view and make nexfs configuration changes
SECURITYCONFGROUPACCESS
Live
Default: 2
Valid values: 0=no access, 1 read-only, 2 read/write
The access level nexfs configuration group members have for viewing and making configuration changes
Advanced Settings
Config TagName
Live Update
or Restart
Allowed
Values
Description
NEXFSID
Live
Default: 0
Valid values:
Min: 0
Max: 32768
Unique ID of the Nexfs Server in the Region Cluster
NEXFSREGION
Live
Default: nexfs
Valid values:
between 0 and 64 characters
Nexfs server cluster region (also the services S3 API region)
TERMSANDCONDITIONSACCEPTED
Restart
Default: 0
Valid values: 0 (no) or 1 (yes)
Must be set to 1 confirm acceptance of Nexustorage T&Cs available at https://www.nexustorage.com/nexustorage-terms-and-conditions
PARTSIZE
Live
Default: 1024000
Valid values:
Min: 4096
Max: 8192000
Size of nexfs file dataparts in bytes.
WRITEBUFSIZE
Live
Default: 4096000
Valid values:
Min: 4096
Max: 4096000
Write buffer size in bytes. The size of incoming write buffers (for receiving data from the kernel)
ENABLEWRITEBUFFERING
Restart
Default: 1
Valid values: 0 (no) or 1 (yes)
Enable Write Buffering, group small block writes together to increase performance, a client may receive write ok before data is written to permanent storage.
When enabled, if there is no space available to store written data for an already open file, Nexfs will not return from that write request until space becomes available to secure that data.
Additionally, if nexfs is paused and a write request for an already open file is received, nexfs will not complete that request until nexfs is unpaused. When not enabled, write requests that cannot be written to permanent due to no available space, the request will return instantly, informing the caller that the storage is full and the request was not successful. When nexfs is paused, write requests will return EAGAIN: resource busy, try again
CTIMEISREQTIME
Live
Default: 0 (Off) Valid values: 0 (Off), 1 (On)
When enabled, the ctime of a file stat result is always set to the current time of the stat operation, NFSv4 clients can rely heavily on ctime changing to manage their internal cache. This setting is recommended for Nfsv4 use cases
BUFFERCHUNKONREAD
Live
Default: 0 Valid values: 0 (No) or 1 (Yes)
Load data chunk into a a lower buffer on read from Tier3+
MAXFILENAMELENGTH
Restart
Defaut: 2048
Valid values:
Min: 1024
Max: 2048
The maximum length in characters of a file name including its full path that Nexfs can process.
The upper limit will soon be increased to 65536
MAXHANDLEPAGES
Restart
Defaut: 4
Valid values:
Min: 1
Max: 128
The maximum number of in memory dataparts that nexfs can have open for each open file handle.
More handles allows more concurrent client requests to be serviced for each open filehandle, and allows more dataparts to be cached in memory. To many parts can lower IO as nexfs needs to scan each cached datapart for a match for every read/wrtie request.
CACHELOCKWAIT
Live
Defaut: 4000
Valid values:
Min: 5
Max: 99999999
The number of nano seconds to wait between attempts to take a buffer lock.
Having this set to large will reduce performance with nexfs is sleeping, having it to low will increase CPU usage as nexfs attempts to take unavailable buffer locks more often.
NEXFSMAXOPENFILES
Restart
Default: 65536
Valid values:
Min: 1024
Max: 1048576
Max number of open file handles Nexfs can support (includes all internal file handles)
STRUCTERRTHRESHOLD
Live
Defaut: 10
Valid values:
Min: 1
Max: 65535
The number of errors received when writing to the active structure data volume that triggers a failover to the replication volume
DROPCACHEONOPEN
Restart
Default: 0
Valid values: 0 (no) or 1 (yes)
If enabled, any data previously cached for a file will be dropped when the file is re-opened.
Enabling this can negatively impact performance
MGMTWEBSERVERHTTPSENABLED
Live
"Default: 0
Valid values: 0 (no) or 1 (yes)"
Enable the Nexfs integrated SSL Management web server
MGMTWEBSERVERMAXCONNECTIONS
Live
Default: 4
Valid values:
Min: 1
Max: 64
Maximum number of concurrent managment webserver connection requests, management webserver must be stopped for liveupdate
MGMTWEBSERVERPORT
Live
Default: 9200
Valid values: Min: 32 to Max: 65535
The tcp port the managment web server should listen for requests on
MGMTWEBTOKENEXPIRY
Live
Default: 300 seconds
Valid values: Min: 60 to Max: 315600000
Number of seconds after which a idle management api (and webconsole) token should expire
MGMTWEBSERVERENABLED
Live
Default: 1
Valid values: 0 (no) or 1 (yes)
Enable the Nexfs integrated management web server
LOOKUPCACHEEXPIRY
Restart
Default: 30 (Seconds)
Valid values: 0 (Don't cache) to 86400
Time in seconds that directory entries should be cached by the Kernel