An Error Is Returned When I Change the Read-Only Status of an Index
Symptom
When the disk usage of a security cluster exceeds the threshold, all indices will enter a read-only mode (with read_only_allow_delete set to true), preventing further writes. To restore write access, manually set read_only_allow_delete to false by running the following command:
PUT _settings
{
"index": {
"blocks": {
"read_only_allow_delete": "false"
}
}
} The error information is as follows:
{
"error": {
"root_cause": [
{
"type": "security_exception",
"reason": "no permissions for [] and User [name=admin, roles=[admin], requestedTenant=null]"
}
],
"type": "security_exception",
"reason": "no permissions for [] and User [name=admin, roles=[admin], requestedTenant=null]"
},
"status": 403
} Possible Causes
By default, a security cluster has an . opendistro_security index, which cannot be written. You need to skip this index when changing the status of indexes.
Solution
Use wildcards to match specified indexes. (Use a wildcard to replace the indexname in the following example.)
PUT indexname/_settings
{
"index": {
"blocks": {
"read_only_allow_delete": "false"
}
}
}