Since JIRA 5.0.3 comment length can be limited by using the jira.text.field.character.limit property (see JRA-27668). However, it is still possible to crash JIRA by adding many comments to an issue (or when the limit is disabled, which is the default).

Steps to reproduce:

• Create a huge issue by adding lots of large comments or a large number of comments
• open the issue in the browser, clicking "Refresh" a few times so that it's being loaded in several parallel requests
• JIRA will crash with OutOfMemoryException

We need to be able to handle large issues like this. Some options for fixing that come to mind:

• only rendering the first N comments (where N may vary depending on the comment limit)
• calculate the size of the comments before loading them into memory, and do something different depending on whether they are above or below a certain threshold

Those are just 2 of many possible options. The important thing is that JIRA doesn't crash, and that we handle both dimensions (comment size and number of comments).

Resolution - JIRA 6.4 and above

For JIRA 6.4.x customers

All the code changes to support the limit throughout JIRA will be included in JIRA 6.4.1.

However, the limit is still off by default, customers wishing to enforce the limit can manually set the jira.text.field.character.limit advanced setting. We recommend to set the value to 32767 which will be new default from JIRA 7.0 and onwards.

For JIRA 7.0 and Cloud customers

All the code changes to support the limit throughout JIRA will be included in the next Cloud release and JIRA 7.0

jira.text.field.character.limit will be set to 32767 by default unless you have explicitly changed this setting to a different value.

Workaround

Extract large comments and set jira.text.field.character.limit.
Please see our OutOfMemoryError Caused by Large Comments or Descriptions in an Issue KB for further information on this.