Description of problem: We are using proftp to provide a SFTP service. We noticed that since the latest version (we just upgraded to it), that uploads using SFTP on a fully up to date C7 causing the daemon to segfault. This causes corrupted data. Version-Release number of selected component (if applicable): proftpd-1.3.5b-1.el7.x86_64 How reproducible: Almost always (> 70%) Steps to Reproduce: 1. Install the latest proftpd on a fully up to date C7 2. Upload multiple packages using SFTP 3. After the first package corruption can occur Actual results: Files of correct size, but corrupted content on the server Expected results: Completely intact files on the server Additional info: This line is visible in the /var/log/messages: May 20 10:56:28 localhost proftpd[11719]: 0.0.0.0 (10.57.25.155[10.57.25.155]) - ProFTPD terminating (signal 11)
Some more additionel info, going back to 1.3.5a-2 fixed the segfault.
Looks a bit like http://bugs.proftpd.org/show_bug.cgi?id=4203 Can you give this scratch build a try and see if it helps? http://koji.fedoraproject.org/koji/taskinfo?taskID=14208851
Hello Paul, I've got a bit delayed dealing with other stuff. But I've upgraded our server today with that build from koji and I can confirm that it fixes the segfault. So proftpd no longer crashes and files are no longer corrupted. This build can be pushed our IMHO. Thanks, Tim
Hi Tim, thanks for the feedback; I'll push out updates for EPEL and Fedora now.
proftpd-1.3.5b-2.el7 has been submitted as an update to Fedora EPEL 7. https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2016-e9000a1ec5
proftpd-1.3.5b-2.el7 has been pushed to the Fedora EPEL 7 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2016-e9000a1ec5
proftpd-1.3.5b-2.el7 has been pushed to the Fedora EPEL 7 stable repository. If problems still persist, please make note of it in this bug report.