VMware Greenplum Text configuration parameters can be overridden by setting a new value in a VMware Greenplum session. Changes made to configuration parameters only affect future VMware Greenplum Text operations; existing indexes use the parameter values that were set when they were created.
See Changing VMware Greenplum Text Server Configuration Parameters for information about changing configuration parameters and examples.
The following table lists the VMware Greenplum Text configuration parameters with their defaults and value constraints.
Parameter Name | Description | Minimum | Maximum | Default |
---|---|---|---|---|
admin_timeout |
Timeout, in seconds, for admin requests (create_index, etc.). | 30 |
INT_MAX |
3600 |
commit_timeout |
Timeout, in seconds, for prepare commit and commit operations. | 30 |
INT_MAX |
3600 |
delete_timeout |
Timeout, in seconds, for delete requests. | 30 |
INT_MAX |
3600 |
extension_factor |
Maximum number of replicas that can be added for an index per VMware Greenplum Text node after the index is created. | 0 |
10 |
2 |
facet_timeout |
Timeout, in seconds, for faceting queries. | 30 |
INT_MAX |
3600 |
failover_factor |
Minimum ratio of Solr nodes that must be up in order to create a new index. (SolrNodesUp/TotalSolrNodes ) |
0.0 |
1.0 |
0.8 |
hl_post_tag |
Markup that gptext.highlight() and gptext.highlight_instant_content() inserts after terms in search results. |
'</em>' |
||
hl_pre_tag |
Markup that gptext.highlight() and gptext.highlight_instant_content() inserts before terms in search results. |
'<em>' |
||
idx_buffer_size |
Size of indexing buffer in bytes. | 16777216 |
536870912 |
134217728 |
idx_delim |
Delimiter to use during indexing. | comma
|
||
idx_encapsulator |
The character optionally used to surround values to preserve characters such as the CSV separator or whitespace. | quote
|
||
idx_escape |
Escape character to use for indexing. | backslash
|
||
idx_num_shards |
The number of shards to create for indexes that use the Solr CompositeId router. When 0, the default, the number of shards is the same as the number of VMware Greenplum segments. When the index allows non-unique document IDs, the Solr Implicit router is used, and the number of shards is always equal to the number of VMware Greenplum primary segments. | 0 | 1024 | 0 |
index_timeout |
Timeout, in seconds, for receiving response to indexing operation. | 30 |
INT_MAX |
3600 |
optimize_timeout |
Timeout, in seconds, for optimize operations. | 30 |
INT_MAX |
3600 |
ping_timeout |
Timeout, in seconds, for ping requests. | 30 |
INT_MAX |
120 |
replication_factor |
The number of replicas per shard for a newly created index. | 1 |
10 |
2 |
replication_timeout |
Timeout, in seconds, for replication operations (backup, restore). | 30 |
INT_MAX |
43200 |
rollback_timeout |
Timeout, in seconds, for rollback operations. | 30 |
INT_MAX |
3600 |
search_batch_size |
Batch size for search requests. | 1 |
INT_MAX |
2500000 |
search_buffer_size |
Buffer size for search results, in bytes. | 4096 |
67108864 |
16777216 |
search_param_separator |
Delimiter to use in the options parameter of the gptext.search() UDF. |
'&' |
||
search_post_buffer_size |
Post buffer size for search, in bytes. | 512 |
4194304 |
4096 |
search_timeout |
Timeout, in seconds, for searches. | 30 | INT_MAX | 600 |
stats_timeout |
Timeout, in seconds, for obtaining statistics. | 30 |
INT_MAX |
600 |
idx_segment_error_limit |
Limit for indexing errors per segment. If this value is exceeded on any segment, the indexing operation is stopped. | 1 |
INT_MAX |
10 |
terms_batch_size |
Batch size for terms operations. | 1 |
INT_MAX |
1000 |