Releases: gchq/stroom
v7.8-beta.19
v7.8-beta.19
-
Issue #4855 : Fix Plan B session state range.
-
Improve log message when proxy HTTP forwarding fails. Now includes the time taken and the bytes sent.
-
Issue #4858 : Fix Plan B query field error.
-
Issue #4859 : Fix query on open for embedded queries.
-
Issue #4861 : Allow embedded queries to have a set page size.
-
Issue #4748 : Add user group icon to API key page.
-
Issue #4806 : Add filter/group/sort indicators to hidden field menu items.
v7.9-beta.8
v7.9-beta.8
-
Uplift BCrypt lib to 0.4.3.
-
Add BCrypt as a hashing algorithm to data feed keys. Change Data feed key auth to require the header as configured by
dataFeedKeyOwnerMetaKey
. ChangehashAlgorithmId
tohashAlgorithm
in the data feed keys json file.
v7.9-beta.7
v7.9-beta.7
-
Issue #4831 : Fix Data Retention -> Impact Summary not showing any data.
-
Issue #4829 : Fix stuck searches.
-
Issue #4830 : Use a cache rather than sessions to maintain auth flow state to avoid creating unnecessary sessions.
-
Issue #4842 : Fix null session when doing OIDC code flow with KeyCloak.
-
Issue #4844 : Fix issue where vis parent table filters are not applied to the right data values.
-
Issue #4837 : Change the fetching of OIDC config to use jersey client instead of Apache http client. The yaml properties
appConfig.security.authentication.openId.httpClient
andproxyConfig.security.authentication.openId.httpClient
have been removed. Configuration of the jersey client is now done usingjerseyClients.OPEN_ID.
(see https://gchq.github.io/stroom-docs/docs/install-guide/configuration/stroom-and-proxy/common-configuration/#jersey-http-client-configuration). -
Issue #4849 : Fix the default forwarding queue config so that it retries for HTTP and not for FILE. Add the config prop
queue.queueAndRetryEnabled
to control whether forwarding is queued with retry handling or not. Add the config propatomicMoveEnabled
toforwardFileDestinations
items to allow disabling of atomic file moves when using a remote file system that doesn't support atomic moves.
v7.9-beta.6
v7.9-beta.6
-
Issue #4109 : Add
receive
config propertiesx509CertificateHeader
,x509CertificateDnHeader
andallowedCertificateProviders
to control the use of certificates and DNs placed in the request headers by load balancers or reverse proxies that are doing the TLS termination. Header keys were previously hard coded.allowedCertificateProviders
is an allow list of FQDN/IPs that are allowed to use the cert/DN headers. -
Add Dropwizard Metrics to proxy.
-
Change proxy to use the same caching as stroom.
-
Remove unused proxy config property
maxAggregateAge
.aggregationFrequency
controls the aggregation age/frequency. -
Stroom-Proxy instances that are making remote feed status requests using an API key or token, will now need to hold the application permission
Check Receipt Status
in Stroom. This prevents anybody with an API key from checking feed statuses. -
Issue #4312 : Add Data Feed Keys to proxy and stroom to allow their use in data receipt authentication. Replace
proxyConfig.receive.(certificateAuthenticationEnabled|tokenAuthenticationEnabled)
withproxyConfig.receive.enabledAuthenticationTypes
that takes values:DATA_FEED_KEY|TOKEN|CERTIFICATE
(whereTOKEN
means an oauth token or an API key). The feed status check endpoint/api/feedStatus/v1
has been deprecated. Proxies with a version >=v7.9 should now use/api/feedStatus/v2
. -
Replace proxy config prop
proxyConfig.eventStore.maxOpenFiles
withproxyConfig.eventStore.openFilesCache
. -
Add optional auto-generation of the
Feed
attribute using propertyproxyConfig.receive.feedNameGenerationEnabled
. This is used alongside propertiesproxyConfig.receive.feedNameTemplate
(which defines a template for the auto-generated feed name using meta keys and their values) andfeedNameGenerationMandatoryHeaders
which defines the mandatory meta headers that must be present for a auto-generation of the feed name to be possible. -
Add a new Content Templates screen to stroom (requires
Manage Content Templates
application permission). This screen is used to define rules for matching incoming data where the feed does not exist and creating content to process data for that feed. -
Feed status check calls made by a proxy into stroom now require the application permission
Check Receipt Status
. This is to stop anyone with an API key from discovering the feeds available in stroom. Any existing API keys used for feed status checks on proxy will need to haveCheck Receipt Status
granted to the owner of the key. -
Issue #4844 : Fix issue where vis parent table filters are not applied to the right data values.
v7.8-beta.18
v7.8-beta.18
-
Issue #4831 : Fix Data Retention -> Impact Summary not showing any data.
-
Issue #4829 : Fix stuck searches.
-
Issue #4830 : Use a cache rather than sessions to maintain auth flow state to avoid creating unnecessary sessions.
-
Issue #4842 : Fix null session when doing OIDC code flow with KeyCloak.
-
Issue #4844 : Fix issue where vis parent table filters are not applied to the right data values.
-
Issue #4837 : Change the fetching of OIDC config to use jersey client instead of Apache http client. The yaml properties
appConfig.security.authentication.openId.httpClient
andproxyConfig.security.authentication.openId.httpClient
have been removed. Configuration of the jersey client is now done usingjerseyClients.OPEN_ID.
(see https://gchq.github.io/stroom-docs/docs/install-guide/configuration/stroom-and-proxy/common-configuration/#jersey-http-client-configuration). -
Issue #4849 : Fix the default forwarding queue config so that it retries for HTTP and not for FILE. Add the config prop
queue.queueAndRetryEnabled
to control whether forwarding is queued with retry handling or not. Add the config propatomicMoveEnabled
toforwardFileDestinations
items to allow disabling of atomic file moves when using a remote file system that doesn't support atomic moves.
v7.9-beta.5
v7.9-beta.5
-
Issue #4831 : Fix Data Retention -> Impact Summary not showing any data.
-
Issue #4829 : Fix stuck searches.
-
Issue #4830 : Use a cache rather than sessions to maintain auth flow state to avoid creating unnecessary sessions.
-
Issue #4839 : Change record count filter to allow counting of records at a custom depth to match split filter.
-
Issue #4842 : Fix null session when doing OIDC code flow with KeyCloak.
v7.9-beta.4-annotations3
v7.9-beta.4-annotations3
-
Issue #4831 : Fix Data Retention -> Impact Summary not showing any data.
-
Issue #4829 : Fix stuck searches.
-
Issue #4830 : Use a cache rather than sessions to maintain auth flow state to avoid creating unnecessary sessions.
-
Issue #4839 : Change record count filter to allow counting of records at a custom depth to match split filter.
-
Annotations 2.0.
v7.9-beta.4-annotations.4
v7.9-beta.4-annotations.4
- Annotations 2.0.
v7.9-beta.4-annotations2
v7.9-beta.4-annotations2
- Annotations 2.0.