-
Bug
-
Resolution: Fixed
-
Medium
-
5.0, 5.2.1
-
None
-
5
-
This issue deals with how JIRA manages HTTP sessions for web Crawlers. The related issue JRA-27047 deals with session management for stateless requests to the REST/SOAP API.
JIRA creates a new session for every request that comes in from the Google bot, and any other crawler that doesn't accept cookies.
This is a problem for public-facing sites, since it strains the server unnecessarily (these sessions hang around for 5 hours). The Apache guys have solved this on http://issues.apache.org by using a Tomcat valve, see Crawler Session Manager Valve for their write-up.
To fix this we should refrain from creating new sessions for crawlers.
- is duplicated by
-
JRASERVER-27406 Get the Bot-Killer bundled in OnDemand for use in JIRA
- Closed
- is related to
-
JRASERVER-27047 JIRA creates sessions for "stateless" requests
- Closed
- relates to
-
JRASERVER-31322 REST session not terminated
- Closed