Skip to main content

[Resolved] Sporadic stale file handles & potential error messages when logging in to CentOS 7 environment

Posted by on Tuesday, September 18, 2018 in Cluster Status Notice.

Update 9/21/2018: It appears this issue has been resolved. Please open a Helpdesk ticket if you experience any problems.

This morning we began sporadically seeing stale file handle messages in our CentOS 7 environment, which may have also impacted logins to We initially thought this due to a bad hard drive on one of our gateways, but it now appears that a user’s analysis processes running on the gateway may be the cause. We are working with this user to understand the root cause of the problem to prevent this from occurring again.

We added another gateway to the rotation this morning when we believed a bad hard drive was causing problems. We did this quickly and unfortunately did not ensure that a key associated with the new gateway was consistent with the other gateway in our rotation. As a result, you may see messages like the following when you attempt to ssh to

The ECDSA host key for has changed
These can be safely ignored and you can follow the instructions at the bottom of the message to resolve this issue. Specifically, look in your ~/.ssh/known_hosts file and remove the line containing The next time you attempt to login to, you will be asked to verify the authenticity of the host, to which you can simply reply yes.
Please open a Helpdesk ticket with us if you have any questions.