The primary form of configuration is via the use of key/value properties, stored in etc/grid/config.xml and read on server startup. Backing up and copying these properties is as easy as copying this file to a new server version.
The etc/omero.properties file of your distribution defines all the default configuration properties used by the server. Changes made to the file are not recognized by the server. Instead, the omero config command is used to change those properties that you would like to customize.
Examples of doing this are on the main Unix and Windows pages, as well as the LDAP installation page.
Sets the number of connections to PostgreSQL which will be used by OMERO. Your database installation will need to be configured to accept at least as many, preferably more, connections as this value.
Default: 10
Default values for the current profile will be hard-coded into the hibernate.properties file in the model-*.jar. By using a different jar, you can modify the defaults.
Note: some other properties are defined in the file etc/profiles/${omero.db.profile} Especially of importance is omero.db.port
Default: psql
Whether JMX statistics are collected for DB usage (by Hibernate, etc)
Default: true
Checksum algorithms supported by the server for new file uploads, being any comma-separated non-empty subset of:
In negotiation with clients, this list is interpreted as being in descending order of preference.
Default: SHA1-160, MD5-128, Murmur3-128, Murmur3-32, CRC-32, Adler-32, File-Size-64
Template for FS managed repository paths. Allowable elements are:
%user% bob
%userId% 4
%group% bobLab
%groupId% 3
%year% 2011
%month% 01
%monthname% January
%day% 01
%time% 15-13-54.014
%institution% University of Dundee
%hash% 0D2D8DB7
%increment% 14
%subdirs% 023/613
%session% c3fdd5d8-831a-40ff-80f2-0ba5baef448a
%sessionId% 592
%perms% rw----
/ path separator
// end of root-owned directories
These are described further at FS configuration options
The path must be unique per fileset to prevent upload conflicts, which is why %time% includes milliseconds.
A // may be used as a path separator: the directories preceding it are created with root ownership, the remainder are the user’s. At least one user-owned directory must be included in the path.
The template path is created below omero.managed.dir, e.g. /OMERO/ManagedRepository/<omero.fs.repo.path>/
Default: %user%_%userId%//%year%-%month%/%day%/%time%
Rules to apply to judge the acceptability of FS paths for writing into omero.managed.dir, being any comma-separated non-empty subset of:
Minimally, the “required” appropriate for the server is recommended. Also applying optional rules may make sysadmin tasks easier, but may be more burdensome for users who name their files oddly.
Default: Windows required, UNIX required
registry_timeout is the milliseconds which the registry and other services will wait on remote services to respond.
Default: 5000
Disable IPv6 by setting to 0. Only needed in certain situations.
Default: 1
Contains other parameters which should be passed to the JVM. The value of “append” is treated as if it were on the command-line and so will be separated on whitespace. For example, ‘-XX:-PrintGC -XX:+UseCompressedOops’ would results in two new arguments.
Default: [empty]
Toggles on or off heap dumps on OOMs. Default is “off”. The special value “tmp” will create the heap dumps in your temp directory.
Default: [empty]
Explicit value for the -Xmx argument, e.g. “1g”
Default: [empty]
Suggestion for strategies as to the maximum memory that they will use for calculating JVM settings (MB).
Default: 48000
Suggestion for strategies as to the minimum memory that they will use for calculating JVM settings (MB).
Default: 3414
Used only by the percent strategy. An integer between 0 and 100 which is the percent of active memory that will be used by the service.
Default: [empty]
Explicit value for the MaxPermSize argument to the JVM, e.g. “500M”
Default: [empty]
Memory strategy which will be used by default. Options include: percent, manual
Default: percent
Manual override of the total system memory that OMERO will think is present on the local OS (MB). If unset, an attempt will be made to detect the actual amount: first by using the Python library psutil and if that is not installed, by running a Java tool. If neither works, 4.0GB is assumed.
Default: [empty]
LDAP server base search DN, i.e. the filter that is applied to all users. (can be empty in which case any LDAP user is valid)
Default: ou=example,o=com
Enable or disable LDAP (true or false).
Default: false
LDAP server binard password (if required; can be empty)
Default: [empty]
Available referral options are: “ignore”, “follow”, or “throw” as per the JNDI referral documentation.
Default: ignore
Whether or not values from LDAP will be sychronized to OMERO on each login. This includes not just the user name, email, etc, but also the groups that the user is a member of.
Warning
Currently setting this to true the user will be removed from any groups to which they have been added outside of LDAP! Please use carefully.
Default: false
Set the URL of the LDAP server. A SSL URL for this property would be of the form: ldaps://ldap.example.com:636
Default: ldap://localhost:389
LDAP server bind DN (if required; can be empty)
Default: [empty]
Which bean to use: nullMetrics does nothing defaultMetrics uses the properties defined below
Default: defaultMetrics
Address for Metrics to send server data
Default: [empty]
Number of minutes to periodically print to slf4j 0 or lower disables the printout.
Default: 60
Sets the duration of inactivity in milliseconds after which a login is required.
Default: 600000
Value for the indexer is extended to 1 day
Default: 86400000
Value for the indexer is extended to 1 hour
Default: 3600000
Name of the spring bean which will be used to calculate the backoff (in ms) that users should wait for an image to be ready to view.
Default: ome.io.nio.SimpleBackOff
Number of instances indexed per indexing.
Default: 50
Polling frequency of the pixeldata processing. Set empty to disable pixelsdata processing.
Cron Format: seconds minutes hours day-of-month month day-of-week year (optional). For example, “0,30 * * * * ?” is equivalent to running every 30 seconds. For more informatin, see the CronTrigger Tutorial.
Default: */4 * * * * ?
Whether the PixelData.dispose() method should try to clean up ByteBuffer instances which may lead to memory exceptions. See ticket #11675 for more information. Note: the property is set globally for the JVM.
Default: false
Maximum time in milliseconds that file parsing can take without the parsed metadata being cached to BioFormatsCache.
Default: 0
Instead, it is possible to tell the server to run more pixeldata repetitions, each of which gets completely committed before the next. This will only occur when there is a substantial backlog of pixels to process.
Default: 1
How many pixel pyramids will be generated at a single time. The value should typically not be set to higher than the number of cores on the server machine.
Default: 2
Default sizes for tiles are provided by a ome.io.nio.TileSizes implementation. By default the bean (“configuredTileSizes”) uses the properties provided here.
Default: configuredTileSizes
No value implies use sys.executable
Default: [empty]
Analyzer used both index and to parse queries
Default: ome.services.fulltext.FullTextAnalyzer
Size of the batches to process events per indexing. Larger batches can speed up indexing, but at the cost of memory.
Default: 5000
Extra bridge classes, comma-separated, to be invoked on each indexing. Bridges are used to parse more information out of the data.
Default: [empty]
Polling frequency of the indexing. Set empty to disable search indexing.
Cron Format: seconds minutes hours day-of-month month day-of-week year (optional). For example, “0,30 * * * * ?” is equivalent to running every 30 seconds. For more informatin, see the CronTrigger Tutorial.
Default: */2 * * * * ?
Indexing takes place on all EventLogs as they occur in the database. The types listed here will be skipped if they appear in the “entityType” field of the EventLog table.
Default: ome.model.annotations.ChannelAnnotationLink,ome.model.core.Channel,ome.model.core.PlaneInfo,ome.model.core.PixelsOriginalFileMap,ome.model.containers.DatasetImageLink,ome.model.containers.ProjectDatasetLink,ome.model.containers.CategoryGroupCategoryLink,ome.model.containers.CategoryImageLink,ome.model.display.ChannelBinding,ome.model.display.QuantumDef,ome.model.display.Thumbnail,ome.model.meta.Share,ome.model.meta.Event,ome.model.meta.EventLog,ome.model.meta.GroupExperimenterMap,ome.model.meta.Node,ome.model.meta.Session,ome.model.annotations.RoiAnnotationLink,ome.model.roi.Roi,ome.model.roi.Shape,ome.model.roi.Text,ome.model.roi.Rect,ome.model.roi.Mask,ome.model.roi.Ellipse,ome.model.roi.Point,ome.model.roi.Path,ome.model.roi.Polygon,ome.model.roi.Polyline,ome.model.roi.Line,ome.model.screen.ScreenAcquisitionWellSampleLink,ome.model.screen.ScreenPlateLink,ome.model.screen.WellReagentLink,ome.model.stats.StatsInfo
EventLog.action values which will be indexed. Unless custom code is generating other action types, this property should not need to be modified.
Default: INSERT,UPDATE,REINDEX,DELETE
Whitelist of object types which will be indexed. All other types will be ignored. This matches the currently available UI options but may need to be expanded for custom search bridges.
Default: ome.model.core.Image,ome.model.containers.Project,ome.model.containers.Dataset,ome.model.screen.Plate,ome.model.screen.Screen
Maximum file size for text indexing (bytes) If a file larger than this is attached, e.g. to an image, the indexer will simply ignore the contents of the file when creating the search index. This should not be set to more than half of the Indexer heap space.
Note
If you set the max file size to greater than 1/2 the size of the indexer’s heap (256 MB by default), you may encounter Out of Memory errors in the Indexer process or you may cause the search index to become corrupt. Be sure that you also increase the heap size accordingly (see OutOfMemoryError / PermGen space errors in OMERO.server logs).
Default: 131072000
Number of objects to load in a single indexing window. The larger this value the fewer times a single object will be indexed unnecessarily. Each object uses roughly 100 bytes of memory.
Default: 1000000
Maximum number of OR-clauses to which a single search can expand
Default: 4096
Instead, it is possible to tell the server to run more indexing reptitions, each of which gets completely committed before the next. This will only occur when there is a substantial backlog of searches to perform. (More than 1 hours worth)
Default: 1
Periodically the completion percentage will be printed. The calculation can be expensive and so is not done frequently.
Default: 100
A keystore is a database of private keys and their associated X.509 certificate chains authenticating the corresponding public keys. A keystore is mostly needed if you are doing client-side certificates for authentication against your LDAP server.
Default: [empty]
Sets the password of the keystore
Default: [empty]
Controls whether the server will allow creation of user accounts with an empty password. If set to true (default, strict mode), empty passwords are disallowed. This still allows the guest user to interact with the server.
Default: true
A truststore is a database of trusted entities and their associated X.509 certificate chains authenticating the corresponding public keys. The truststore contains the Certificate Authority (CA) certificates and the certificate(s) of the other party to which this entity intends to send encrypted (confidential) data. This file must contain the public key certificates of the CA and the client’s public key certificate.
Default: [empty]
Sets the password of the truststore
Default: [empty]