If you want to start a forcemerge via the Opensearch Dashboard as it says in the instructions: https://opensearch.org/docs/latest/dashboards/im-dashboards/forcemerge then the message appears that the forcemerge has been started, but in reality no one was triggered on the cluster. After some seconds a successful message appears.. If you run the forcemerge via the dev console, it will be executed on the cluster, but will run into a timeout in the dev console. If you run the forcemerge with the same technical user trought the API directly to the cluster, everythings works as expected. I have the feeling that something is being transmitted incorrectly between opensearch dashboard and opensearch cluster. User has "indices:*" index_allowed_actions.
The opensearch dashboard logs are a little bit confused. When the User has "indices:*" under index_allowed_actions I received following message:
Index Management - CommonService - apiCaller StatusCodeError: [status_exception] lronConfig LRON:indices:admin/forcemerge not found
at respond (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:349:15)
at checkRespForFailure (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:306:7)
at HttpConnector. (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/connectors/http.js:173:7)
at IncomingMessage.wrapper (/usr/share/opensearch-dashboards/node_modules/lodash/lodash.js:4991:19) at IncomingMessage.emit (node:events:525:35)
at IncomingMessage.emit (node:domain:489:12)
at endReadableNT (node:internal/streams/readable:1359:12)
at processTicksAndRejections (node:internal/process/task_queues:82:21) {
status: 404,
displayName: 'NotFound',
path: '/_plugins/_im/lron/LRON%3Aindices%3Aadmin%2Fforcemerge',
query: undefined,
body: {
error: {
root_cause: [Array],
type: 'status_exception',
reason: 'lronConfig LRON:indices:admin/forcemerge not found'
},
status: 404 },
statusCode: 404,
response: '{"error":{"root_cause":[{"type":"status_exception","reason":"lronConfig LRON:indices:admin/forcemerge not found"}],"type":"status_exception","reason":"lronConfig LRON:indices:admin/forcemerge not found"},"status":404}',
toString: [Function (anonymous)],
toJSON: [Function (anonymous)]}
What the heck is LRON? When the user has "indices_all" under index_allowed_actions then I receive the following message:
Index Management - CommonService - apiCaller StatusCodeError: [index_not_found_exception] no such index [.opensearch-control-center], with { index=".opensearch-control-center" & resource.id=".opensearch-control-center" & resource.type="index_expression" & index_uuid="na" }
at respond (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:349:15)
at checkRespForFailure (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:306:7)
at HttpConnector. (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/connectors/http.js:173:7)
at IncomingMessage.wrapper (/usr/share/opensearch-dashboards/node_modules/lodash/lodash.js:4991:19)
at IncomingMessage.emit (node:events:525:35)
at IncomingMessage.emit (node:domain:489:12)
at endReadableNT (node:internal/streams/readable:1359:12)
at processTicksAndRejections (node:internal/process/task_queues:82:21) {
status: 404,
displayName: 'NotFound',
path: '/_plugins/_im/lron/LRON%3Aindices%3Aadmin%2Fforcemerge',
query: undefined,
body: {
error: {
root_cause: [Array],
type: 'index_not_found_exception',
reason: 'no such index [.opensearch-control-center]',
index: '.opensearch-control-center',
'resource.id': '.opensearch-control-center',
'resource.type': 'index_expression',
index_uuid: 'na'
},
status: 404
},
statusCode: 404,
response: '{"error":{"root_cause":[{"type":"index_not_found_exception","reason":"no such index [.opensearch-control-center]","index":".opensearch-control-center","resource.id":".opensearch-control-center","resource.type":"index_expression","index_uuid":"na"}],"type":"index_not_found_exception","reason":"no such index [.opensearch-control-center]","index":".opensearch-control-center","resource.id":".opensearch-control-center","resource.type":"index_expression","index_uuid":"na"},"status":404}',
toString: [Function (anonymous)],
toJSON: [Function (anonymous)]}
I don't now what the difference between these permissions are (as documented these should identically), but I think something is completely Bug! It would be very nice to get some help to solve these problems.
yanicksimon
commented
11 months ago 
dblock
Describe the bug
If you want to start a forcemerge via the Opensearch Dashboard as it says in the instructions: https://opensearch.org/docs/latest/dashboards/im-dashboards/forcemerge then the message appears that the forcemerge has been started, but in reality no one was triggered on the cluster. After some seconds a successful message appears.. If you run the forcemerge via the dev console, it will be executed on the cluster, but will run into a timeout in the dev console. If you run the forcemerge with the same technical user trought the API directly to the cluster, everythings works as expected. I have the feeling that something is being transmitted incorrectly between opensearch dashboard and opensearch cluster. User has "indices:*" index_allowed_actions.
The opensearch dashboard logs are a little bit confused. When the User has "indices:*" under index_allowed_actions I received following message:
Index Management - CommonService - apiCaller StatusCodeError: [status_exception] lronConfig LRON:indices:admin/forcemerge not found at respond (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:349:15) at checkRespForFailure (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:306:7) at HttpConnector. (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/connectors/http.js:173:7)
at IncomingMessage.wrapper (/usr/share/opensearch-dashboards/node_modules/lodash/lodash.js:4991:19) at IncomingMessage.emit (node:events:525:35)
at IncomingMessage.emit (node:domain:489:12)
at endReadableNT (node:internal/streams/readable:1359:12)
at processTicksAndRejections (node:internal/process/task_queues:82:21) {
status: 404,
displayName: 'NotFound',
path: '/_plugins/_im/lron/LRON%3Aindices%3Aadmin%2Fforcemerge',
query: undefined,
body: {
error: {
root_cause: [Array],
type: 'status_exception',
reason: 'lronConfig LRON:indices:admin/forcemerge not found'
},
status: 404 },
statusCode: 404,
response: '{"error":{"root_cause":[{"type":"status_exception","reason":"lronConfig LRON:indices:admin/forcemerge not found"}],"type":"status_exception","reason":"lronConfig LRON:indices:admin/forcemerge not found"},"status":404}',
toString: [Function (anonymous)],
toJSON: [Function (anonymous)]}
What the heck is LRON? When the user has "indices_all" under index_allowed_actions then I receive the following message:
Index Management - CommonService - apiCaller StatusCodeError: [index_not_found_exception] no such index [.opensearch-control-center], with { index=".opensearch-control-center" & resource.id=".opensearch-control-center" & resource.type="index_expression" & index_uuid="na" } at respond (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:349:15) at checkRespForFailure (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/transport.js:306:7) at HttpConnector. (/usr/share/opensearch-dashboards/node_modules/elasticsearch/src/lib/connectors/http.js:173:7)
at IncomingMessage.wrapper (/usr/share/opensearch-dashboards/node_modules/lodash/lodash.js:4991:19)
at IncomingMessage.emit (node:events:525:35)
at IncomingMessage.emit (node:domain:489:12)
at endReadableNT (node:internal/streams/readable:1359:12)
at processTicksAndRejections (node:internal/process/task_queues:82:21) {
status: 404,
displayName: 'NotFound',
path: '/_plugins/_im/lron/LRON%3Aindices%3Aadmin%2Fforcemerge',
query: undefined,
body: {
error: {
root_cause: [Array],
type: 'index_not_found_exception',
reason: 'no such index [.opensearch-control-center]',
index: '.opensearch-control-center',
'resource.id': '.opensearch-control-center',
'resource.type': 'index_expression',
index_uuid: 'na'
},
status: 404
},
statusCode: 404,
response: '{"error":{"root_cause":[{"type":"index_not_found_exception","reason":"no such index [.opensearch-control-center]","index":".opensearch-control-center","resource.id":".opensearch-control-center","resource.type":"index_expression","index_uuid":"na"}],"type":"index_not_found_exception","reason":"no such index [.opensearch-control-center]","index":".opensearch-control-center","resource.id":".opensearch-control-center","resource.type":"index_expression","index_uuid":"na"},"status":404}',
toString: [Function (anonymous)],
toJSON: [Function (anonymous)]}
I don't now what the difference between these permissions are (as documented these should identically), but I think something is completely Bug! It would be very nice to get some help to solve these problems.
To Reproduce Steps to reproduce the behavior:
See instructions here: https://opensearch.org/docs/latest/dashboards/im-dashboards/forcemerge
Expected behavior A forcemerge will be triggered at the cluster and I can see the forcemerge in the thread pools or in the queue at the clusters.
OpenSearch Version 2.11.0
Dashboards Version 2.11.0
Plugins
Screenshots
Host/Environment (please complete the following information):
Additional context