Issues fixed in MarkLogic Server 11.0.2
- Last Updated: July 6, 2026
- 6 minute read
- MarkLogic Server
- Documentation
Released: Mar 3, 2023
The following issues have been addressed in this release:
| Platform | Component | Bug ID | Details |
|---|---|---|---|
| all | BUG-60070 | When an invalid @Sort directive is used (such as missing the direction property), an internal error message is returned instead of a valid GraphQL response When an invalid @Sort directive is used (such as missing the direction property), then an internal error message is returned instead of a valid GraphQL response that includes the error in the "errors" property. |
|
| all | BUG-60126 | The admin UI takes a long time to render cluster or database status for large clusters The admin UI cluster status and database status pages take a long time to load and even time out in some cases when the running against a cluster with many hosts and/or databases with 100s of forests. |
|
| all | BUG-60156 | Backup jobs to S3 takes longer to initiate in MarkLogic 11 In MarkLogic 11, backup jobs that write to S3 take longer to initiate than they did in MarkLogic 10. |
|
| all | BUG-60026 | PDF conversion sometimes reports that the config file cannot be found In some cases, PDF conversion is reporting that the config file cannot be found; ICN-CANTOPENCONFIG |
|
| all | BUG-60069 | The @sort directive is unrecognized unless it is capitalized as @Sort. The directives should be normalized to all be lower-case. However, @Sort was missed. |
|
| all | BUG-60134 | Invalid directives could cause unexpected results The GraphQL implementation was ignoring invalid directives, so users might have a typo in a directive, or they might attempt to use an invalid directive. In both those cases, the query might function "properly", but not return the expected results. |
|
| all | BUG-60215 | Environment UI messages with /n do not display correctly In MarkLogic 10.0-9 and later, environment UI messages configured via the /cluster-ui-settings.xml document that have /n characters are not rendered correctly. |
|
| all | BUG-60064 | An empty "errors" array is included in the JSON when there are no errors When there are no errors, the response JSON includes an "errors" property with an empty array as the value. This is a violation of the GraphQL specification. |
|
| all | BUG-60173 | Optic cluster statistics do not include version numbers When you run Optic queries with a Trace option (op:result(..,"trace=Identifier")) or use the "Optic" trace event, the queries do not log the cluster effective version, version and software version in the cluster statistics XML. |
|
| all | BUG-58309 | No information is printed when PDF conversion cannot find the config file PDF conversion will sometimes fail because it cannot find the converter configuration file but there are no log messages indicating that this is the problem. |
|
| all | CLD-665 | jackson-databind security vulnerability Upgraded jackson-databind in the AWS Managed Cluster feature to address CVE-2022-42004. |
|
| all | BUG-60186 | Users receive a server error when adding certificate templates without required fields In the Admin UI, when adding a certificate template, if some required fields are not filled in, the user will receive a server error rather than the UI validating the fields before sending them. |
|
| all | BUG-60160 | Inconsistent capitalization for labels in database field create form of the Admin UI The Admin UI has Inconsistent capitalization for labels in the database field create form as well as number of other forms. |
|
| all | BUG-60224 | LEFT JOIN with IS NULL condition on a nullable column unexpectedly returning no results LEFT JOIN in optic query with IS NULL (op.not(op.isDefined())) condition unexpectedly returns no results |
|
| all | BUG-60112 | SPARQL queries with multiple OPTIONALs could return the wrong results SPARQL queries that contain multiple OPTIONALs could return the the wrong results in some cases. |
|
| all | BUG-60093 | /v1/healthcheck requires an HTTP accept header The /v1/healthcheck endpoint added in MarkLogic 11 requires an HTTP accept header instead of defaulting to XML as the other REST endpoints do. |
|
| all | BUG-60218 | The /v1/values REST API errors out when the constraint references multiple json-property values The /v1/values REST API does does not work with constraints that reference multiple json-property values. |
|
| all | BUG-60105 | Unexpected duplicated content from XSLT special case Templates are applied too many times if a child node satisfies the "match" |
|
| all | BUG-60157 | Unexpected errors encountered when backing up to S3 with MarkLogic 11 When running backups up to S3 with MarkLogic 11, there may be unexpected errors that may cause the backups to fail. |
|
| all | BUG-60220 | Weekly scheduled backups that are set to run every two weeks run once a week Scheduled backups setup to run every two weeks will still run once a week. |
|
| all | Semantics | BUG-57043 | sem:query-results-serialize produces invalid JSON when serializing constructed triples When using sem:query-results-serialize to serialize constructed triples as JSON, the JSON produced may be invalid. |
| all | BUG-60155 | A maximum of 1000 results are returned from xdmp:filesystem-directry() with S3 paths in MarkLogic 11 In MarkLogic 11, when using xdmp:filesystem-directry() with S3 paths, a maximum of 1000 results will be returned. |
|
| all | BUG-60074 | cts:queries are not properly serialized in Optic plans Optic plans that contain cts:queries are not properly serialized so they cannot be deserialized and executed correctly. |
|
| all | BUG-60068 | When an invalid date is used for one of the range filters, an internal error message is returned instead of a valid GraphQL response When an invalid date is used for one of the range filters (such as @greaterThan), then an internal error message is returned instead of a valid GraphQL response that includes the error in the "errors" property. |
|
| all | BUG-60249 | Serialized Optic plans from MarkLogic 10 that use fromSearchDocs() cannot be used in MarkLogic 11 Serialized Optic plans that use op.fromSearchDocs() that were generated in MarkLogic 10 (e.g. used for query-based views), may fail to work in MarkLogic 11. |
|
| all | BUG-60109 | op.fromSearchDocs() performance could be improved Optic queries that use op.fromSearchDocs() run a lot slower than equivalent cts.search() calls. This can also be seen in Optic queries that use op.joinDocCols(). |
|
| all | adminGUI | BUG-58330 | Forest Status page Replica Table shows data in the wrong columns for number of large replica forests The columns in the forest status page replica table shift and causing data to be displayed under the wrong columns when there is a large list of replica forests. |
| all | BUG-60163 | Lack of logging to diagnose SSL socket errors When SSL socket errors occur, it is difficult to diagnose. Additional logging should be added. |
|
| all | Admin API | BUG-60036 | admin:ui-set-banner() defaults the message to "Message of the day" If no message is provided, the admin:ui-set-banner() function defaults the message to "Message of the day". This message is displayed as a modal dialog when the user logs in. |
| all | BUG-60219 | XQuery mapping operator ! does not work for recursive calls The XQuery mapping operator ! throws an error when used for recursively calling functions in a sequence. |
|
| all | Managed Cluster | BUG-60028 | In AWS, the /var/tmp/marklogic.host file is written with permissions of rw-rw-rw- (666) When running the AWS AMI, the /var/tmp/marklogic.host file is written with permissions of rw-rw-rw- (666) which is flagged by security scanners. |
| all | BUG-60166 | Optic trace does not output optimization information of a cached plan The optimization information of a cached plan not cached together with the plan. If you run op:result(...,"trace=Identifier") or xdmp:sql(...",trace=Identifier") or sem.sparql(...,"trace=Identifier") and the query plan is cached then the optimization information will not be logged. |
|
| all | BUG-60148 | The cluster status page on the admin UI can time out for large clusters The cluster status page on the admin UI can time out when viewing it for large clusters or clusters with 100s of forests. |
|
| all | BUG-60089 | The rewriter for enhanced heath check is not set correctly when upgrading to MarkLogic 11 When upgrading from MarkLogic 9 or 10 to MarkLogic 11, the rewriter for the new enhanced heath check is not set correctly so the new health check will not work correctly. |
|
| all | BUG-60039 | op.xpath() does not support context node XPaths op.xpath() always makes the XPath argument into a rooted path (i.e. adds a leading "/"). This prevents relative paths that start with "./" or "../" from working. |
|
| all | BUG-60006 | Detecting indexes can cause backups to take a long time to start up When starting a backup, the index detection process can cause a significant delay when there a large number of forests in the database. |