Red Hat Bugzilla – Bug 29788
tftp service deactivates because of looping
Last modified: 2007-04-18 12:31:48 EDT
From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows 98) tftp-server-0.17-7.i386.rpm xinetd reports error: 01/2/26@22:13:31 xinetd[29434]: ERROR: tftp service was deactivated because of looping 01/2/26@22:13:31 xinetd[29434]: WARNING: recv: Bad file descriptor (errno = 9)Yoy?Puy?Ypu?aoy?uGoyeoy?+Yoyooy?uyooy? tftpd will not respond any further until xinetd is restarted. Reproducible: Sometimes Steps to Reproduce: 1. upgrade tftpd tftp-server-0.17-7.i386.rpm 2. enable xinetd logging via editing /etc/rc.d/init.d/xinetd script daemon xinetd -filelog /var/log/xinetd.log -reuse - pidfile /var/run/xinetd.pid 3. configure tftpd via editing /etc/xinet.d/tftp server_args = -cs /var/tftpd 4. upload files around 1 megabyte or above several times (let tftpd create file) 5. tftp responded correctly with many small files, but tftping larger files produced looping crash Actual Results: tftpd stops with looping errror tftpd does not die during the file transfer though.... tftpd will not respond again until xinetd is restarted Expected Results: tftpd should have exited normally and waited for future connections
tftp package in 7.0 incorrectly specified 'wait=no' in xinetd.d/tftp. This should be 'wait=yes'. This has been corrected in later package versions. If you change only that setting, will the problem go away?
*** This bug has been marked as a duplicate of 23923 ***