From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040510 Description of problem: The portmapper under RHEL3 AS (2.4.21-15.0.3.EL, probably WS, too) appears to have a problem: Jul 27 15:55:51 foo kernel: application bug: portmap(21231) has SIGCHLD set to SIG_IGN but calls wait(). Jul 27 15:57:00 foo kernel: application bug: portmap(21231) has SIGCHLD set to SIG_IGN but calls wait(). Jul 27 15:58:04 foo kernel: application bug: portmap(21231) has SIGCHLD set to SIG_IGN but calls wait(). Jul 27 16:01:36 foo kernel: application bug: portmap(21231) has SIGCHLD set to SIG_IGN but calls wait(). I don't know what triggers those lines and I don't notice any problems with the machine, either. The portmapper appears to work fine. The output of `rpcinfo -p` looks like this: program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 391002 2 tcp 32769 sgi_fam 100024 1 udp 47544 status 100024 1 tcp 59745 status 100011 1 udp 684 rquotad 100011 2 udp 684 rquotad 100011 1 tcp 687 rquotad 100011 2 tcp 687 rquotad 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100021 1 udp 47547 nlockmgr 100021 3 udp 47547 nlockmgr 100021 4 udp 47547 nlockmgr 100021 1 tcp 59747 nlockmgr 100021 3 tcp 59747 nlockmgr 100021 4 tcp 59747 nlockmgr 100005 1 udp 726 mountd 100005 1 tcp 729 mountd 100005 2 udp 726 mountd 100005 2 tcp 729 mountd 100005 3 udp 726 mountd 100005 3 tcp 729 mountd I found only one similar problem (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=117333) and judging by the comment of Chip Turner I thought I'd post this issue here. Version-Release number of selected component (if applicable): portmap-4.0-56 How reproducible: Didn't try Additional info: Let me know if you need any other information. Thanks.
This appears to be fixed in the latest release by this code: #ifdef IGNORE_SIGCHLD /* Lionel Cons <cons.ch> */ (void)signal(SIGCHLD, SIG_IGN); #else (void)signal(SIGCHLD, reap); #endif with reap() doing the wait3() which will not called if IGNORE_SIGCHLD is set.
After further review it appears code is there but... the compile flag IGNORE_SIGCHLD is not being set... and I really don't want to turn it on at this late stage of the release cycle... So I'm closing this as WONTFIX...