powered by Jive Software

Invalid username or password - no admin logon either

This morning, one of my users came to me and said they’re getting an error message when trying to log into our Openfire system with Spark.
It said, “Invalid username or password”.

I then attempted to log in from my workstation, and got the same result.
Basically, when I click the ‘Login’ button in the lower right corner of the Spark GUI, the message text on the bottom says ‘Authenticating’, and the spinner spins…and keeps spinning and just sits there like that for 30-60 seconds and then pops up an error message that says, “Invalid username or password” (see attached jpg).

Then, I tried to log into the admin console via web browser, and the browser just sits there for a few minutes, and eventually just times out without resulting in logging in.

Does anyone have any idea why, for a system that has been in use for over a year with no such problems, there would be this kind of problem all of a sudden?



you could take a look at the log files.

Maybe Openfire did loose the connection to the database unless you are using the internal one.


I think I’m using the internal DB.

This is on Ubuntu Linux.

Which log files should I check, and where are they?

What other details do you need to help me locate them?

The log files are in /path-to/openfire/logs/ where /path-to may be /usr/local on Linux. The error log should have a current timestamp. Do you have still free space in the filesystem used by Openfire?


OK, I located the error log file.

The first two lines in the most recent entry are as follows…

“2010.06.01 11:09:18 [org.jivesoftware.openfire.ldap.LdapAuthProvider.authenticate(LdapAuthProvider. java:122)] Error connecting to LDAP server
javax.naming.CommunicationException: server.domain.edu:389 [Root exception is java.net.ConnectException: Connection timed out]”

I checked and found that “server.domain.edu” is not responding on the network.
Other domain controllers are though.

Is it possible that Openfire has this one domain controller hard-coded somewhere?

If so, does anyone know where I’d go to edit that?


p.s. There are lots more lines of messages in the most recent entry in the log file, but I’m just starting here. If this is causing the problem, then perhaps the rest of the errors are irrelevant. If getting this taken care of doesn’t ‘fix’ it, I will continue on with the rest of the messages.


the ldap server is stored in the database. If you want to edit it you need to stop Openfire and then either use the hsqldb viewer or use “vi”, “sed” or “perl” to replace the ldap URL.

openfire.script and openfire.log are the files of the embedded database, one should not delete them and modify them with care.

For example like this:

cd embedded-db
mv openfire.script openfire.script.org
grep -c server.domain.edu openfire.script.org
# this should return "1", otherwise the next command may change too much
cat openfire.script.org | sed s#server.domain.edu#newserver#g' >openfire.script


Actually, I looked around a little more and found a file that had the name of the domain controller in it (openfire.script I think).
I replaced it with one of the others, and restarted the Linux box.

I will report back in a few minutes when it gets back up and I have a chance to check on it again…

Wow, editing that file by changing the name of the domain controller from one to another fixed the problem; I can now log into the admin console and users can now log into it vis Spark.

Here’s where I found the config file…

Thanks for your help locating the logs.