Error saving draft in Confluence while editing a Page due to "Error saving draft. No draft has been saved."
Platform Notice: Server and Data Center Only - This article only applies to Atlassian products on the server and data center platforms.
While editing a page in Confluence, the following errors may appear.
Error saving draft. No draft has been saved.
Unable to communicate with server. Saving is not possible at the moment.
Screenshot of the issue:
This only occurs to specific pages, while other pages can be created or modified.
Confluence Server or Data Center with Collaborative Editing enabled.
Tomcat logs (
catalina.out) repeatedly shows the following warning message:
INFO [http-nio-8080-exec-198] org.apache.coyote.http11.Http11Processor.service Error parsing HTTP request header Note: further occurrences of HTTP request parsing errors will be logged at DEBUG level. java.lang.IllegalArgumentException: Request header is too large
maxHttpHeaderSize is too small for the current deployment.
According to the Apache Tomcat reference:
The maximum size of the request and response HTTP header, specified in bytes. If not specified, this attribute is set to 8192 (8 KB).
maxHttpHeaderSize value on the necessary Tomcat connector as described in the steps below.
If you are running Confluence Data Center with multiple nodes, this change must be applied to each node.
- Navigate to
- Make a backup copy of the
server.xmland include the
maxHttpHeaderSizeparameter on the main Tomcat connector.
If the parameter is already in place, just edit the value for it.
- Restart Confluence following your standard procedure.
Example of a Tomcat connector with the suggested parameter:
<Connector port="8090" maxHttpHeaderSize="32768" connectionTimeout="20000" maxThreads="48" minSpareThreads="10" enableLookups="false" acceptCount="10" debug="0" URIEncoding="UTF-8" protocol="org.apache.coyote.http11.Http11NioProtocol" />
This parameter does not define a limit and can be increased if needed.
We recommend doubling this value each time it's needed: 8k, 16k, 32k, 64k and so on.