[
https://issues.apache.org/jira/browse/SOLR-13166?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Amrit Sarkar updated SOLR-13166:
--------------------------------
Description:
While working with Solr, schema and configuration changes without understanding
can result in severe node failures, and much effort and time get consumed to
fix such situations.
Few such problematic situations can be:
* Too many fields in the schema
* Too many commits: too short auto commit
* Spellchecker, suggester issues. Build suggester index on startup or on every
commit causes memory pressure and latency issues
-- Schema mess-ups
* Text field commented out and Solr refuses to reload core
* Rename field type for unique key or version field
* Single-valued to multivalued and vice versa
* Switching between docvalues on/off
* Changing text to string type because user wanted to facet on a text field
The intention is to add a layer above Schema and Config API to have some checks
and let the end user know the ramifications of the changes he/she intends to do.
was:
While working with Solr, schema and configuration changes without understanding
can result in severe node failures, and much effort and time get consumed to
fix such situations.
Few such problematic situations can be:
* Too many fields in the schema
* Too many commits: too short auto commit
* Too high cache sizes set which bloats heap memory
* Spellchecker, suggester issues. Build suggester index on startup or on every
commit causes memory pressure and latency issues
-- Schema mess-ups
* Text field commented out and Solr refuses to reload core
* Rename field type for unique key or version field
* Single-valued to multivalued and vice versa
* Switching between docvalues on/off
* Copy/pasting from old solr schema examples and trying them on new versions
* Changing text to string type because user wanted to facet on a text field
* CDCR: if user forgets turning off buffer and the target goes down, the tlog
accumulates until node runs out of disk space or has huge recovery time.
The intention is to add a layer above Schema and Config API to have some checks
and let the end user know the ramifications of the changes he/she intends to do.
> Add smart checks for Config and Schema API in Solr to avoid malicious updates
> -----------------------------------------------------------------------------
>
> Key: SOLR-13166
> URL: https://issues.apache.org/jira/browse/SOLR-13166
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Components: config-api, Schema and Analysis
> Reporter: Amrit Sarkar
> Priority: Major
>
> While working with Solr, schema and configuration changes without
> understanding can result in severe node failures, and much effort and time
> get consumed to fix such situations.
> Few such problematic situations can be:
> * Too many fields in the schema
> * Too many commits: too short auto commit
> * Spellchecker, suggester issues. Build suggester index on startup or on
> every commit causes memory pressure and latency issues
> -- Schema mess-ups
> * Text field commented out and Solr refuses to reload core
> * Rename field type for unique key or version field
> * Single-valued to multivalued and vice versa
> * Switching between docvalues on/off
> * Changing text to string type because user wanted to facet on a text field
> The intention is to add a layer above Schema and Config API to have some
> checks and let the end user know the ramifications of the changes he/she
> intends to do.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
