From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 4.0) Description of problem: The syslog daemon (syslogd only, not klogd) crashes when a logfile reaches 2Gb (exactly 2147483648 bytes). It will also refuse to start up, until the offending file is removed. It hangs for a couple of minutes and than exits. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. create 2Gb file (exactly 2147483648 bytes) or let a log file reach this size. 2. configure this file as a log file in syslog.conf 3. /etc/init.d/syslog start Actual Results: syslogd hanging for couple of minutes and then exits. During start-up, two syslogd processes are seen. One process spawns another. The spawned process is <defunct>. If the file is bigger than 2Gb (2147483648 bytes) syslog will start, but will not log to the file. Expected Results: syslogd to start up normally. Additional info: syslogd version 1.4-1 kernel version 2.4-2 root@syslogcs:/var> ls -l syslog.log -rw-r--r-- 1 root root 2147483647 Nov 7 21:44 syslog.log root@syslogcs:/var> ps -fe |grep syslogd root 20273 20263 0 21:55 pts/0 00:00:00 initlog -q -c syslogd -m 0 -r root 20274 20273 0 21:55 pts/0 00:00:00 syslogd -m 0 -r root 20275 20274 0 21:55 ? 00:00:00 [syslogd <defunct>] root 20280 4578 0 21:55 pts/1 00:00:00 grep syslogd
This is fixed in the sysklogd that shipped with Red Hat Linux 7.2.