The problem of Cyrillic characters in the description field when importing from Trac.

[Alexander]

After importing data from Trac to JIRA. I received an incorrect description field, although the summary field was transferred normally! I do not understand this because both of these fields are taken from the same table from trac and have the same encoding. I also tried to install a clean trac with both the built-in sqlite database and the mysql database, the problem remains.

Can I attach a picture somehow?

[Andy Heinzer · Accepted solution]

With Jira, it's not just the database encoding that you have to worry about.  You also have to consider that the Java JVM Jira is using to load up could be using a different character encoding than expected.  We expect Jira to use a UTF-8 character encoding for the application/java itself.   Please see the details in JIRA Application internationalisation and encoding troubleshooting.  Specifically the parts I think you will need to do are:

• Start JIRA with the variable -Dfile.encoding=utf-8 and -Dsun.jnu.encoding=UTF-8. See Setting properties and options on startup.

I have seen lots of Windows environments that can default into a cp1252 encoding, or Linux/Unix environments that default to an ANSI encoding because that tends to be the operating system's default.  In cases like that, even if your database is setup correctly, when Jira is running, it could be misinterpreting the character encoding because the application settings are set that way.

I'd recommend settings these startup options and restarting Jira to let them take effect.  Once that is done, try to do the import again and see if the data is character encoded correctly.

[Alexander]

thanks, it helped