http://issues.apache.org/SpamAssassin/show_bug.cgi?id=5518
------- Additional Comments From [EMAIL PROTECTED] 2007-06-15 18:17 ------- > I would like to see someone verify it on a BSD system that > demonstrated the problem of bug 3586 before we fixed it > to make sure that this patch doesn't regress that bug. Looks good on: FreeBSD xxx FreeBSD 6.2-RELEASE-p5 #0: ... amd64 [82058] dbg: config: copying current conf to backup [82058] info: spamd: server started on port 783/tcp (running version 3.2.1) [82058] info: spamd: server pid: 82058 [82058] info: spamd: server successfully spawned child process, pid 82060 [82058] dbg: prefork: child 82060: entering state 0 [82058] dbg: prefork: new lowest idle kid: none [82060] dbg: prefork: sysread(8) not ready, wait max 300 secs [82058] info: spamd: server successfully spawned child process, pid 82061 [82058] dbg: prefork: child 82061: entering state 0 [82058] dbg: prefork: new lowest idle kid: none root 82058 58022 3.4 1.2 74720 48820 S+ 0:00.93 /usr/local/bin/perl -T -w /usr/local/bin/spamd --virtual-config-dir=/home/mark/00 -u nobody -D nobody 82060 82058 0.0 1.2 74728 48852 S+ 0:00.01 spamd child (perl) nobody 82061 82058 0.0 1.2 74728 48904 S+ 0:00.01 spamd child (perl) but it was fine before as well (nobody is 65534). If I revert the change in bug 3586 (on 3.2.1), it is again fine, but a workaround-in-use is reported: [87675] info: spamd: server successfully spawned child process, pid 87677 [87675] dbg: prefork: child 87677: entering state 0 [87675] dbg: prefork: new lowest idle kid: none [87677] dbg: spamd: initial attempt to change real uid failed, trying BSD workaround [87675] info: spamd: server successfully spawned child process, pid 87678 [87677] dbg: prefork: sysread(8) not ready, wait max 300 secs root 87675 87445 5.5 1.2 74716 48840 S+ 0:00.93 /usr/local/bin/perl -T -w /usr/local/bin/spamd --virtual-config-dir=/home/mark/00 -u nobody -D nobody 87677 87675 0.0 1.2 74720 48916 S+ 0:00.01 spamd child (perl) nobody 87678 87675 0.0 1.2 74720 48872 S+ 0:00.01 spamd child (perl) So I can say the new patch didn't break anything on FreeBSD. ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee.
