From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from adore.lightlink.com (kimoto@adore.lightlink.com [205.232.34.20]) by kvack.org (8.8.7/8.8.7) with ESMTP id QAA30900 for ; Sun, 21 Jun 1998 16:19:59 -0400 From: Paul Kimoto Message-ID: <19980621161940.A18093@adore.lightlink.com> Date: Sun, 21 Jun 1998 16:19:40 -0400 Subject: Re: update re: fork() failures in 2.1.103 References: <19980618235448.18503@adore.lightlink.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: ; from Rik van Riel on Fri, Jun 19, 1998 at 09:33:54AM +0200 Sender: owner-linux-mm@kvack.org To: Linux MM List-ID: RECAP: In 2.1.99, 2.1.101, 2.1.103, and 2.1.104-pre1, my system has been usable for only ~1 day with 32 MB of memory, or ~2.5 days with 48 MB. Then my system has trouble forking, typically with EAGAIN. The situation can be alleviated temporarily by killing off a few processes, but the errors always reappear soon thereafter. I have sent in the results of Shift-ScrollLock, which Rik thinks are not typical of excessive memory fragmentation. Now, I have scripts that run "ifconfig ppp0" hourly (to check whether PPP is "UP"). Recently I joined the modern era by changing from net-tools 1.432 to 1.45. The forking errors have gone away (at least for uptimes twice the above). When I changed these scripts to run "/sbin/ifconfig.old ppp0" instead, they came back. Running the old ifconfig (when the problem arises) would put "kmod: fork failed, errno 11" messages in the logfiles. The new ifconfig doesn't. Running strace on "ifconfig ppp0" shows that the old version makes the following system calls that the new one doesn't: > socket(PF_??? (0x4), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x4), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x4), SOCK_DGRAM, , 0) = -1 EINVAL (Invalid argument) > socket(PF_??? (0x3), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x3), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x3), SOCK_DGRAM, , 0) = -1 EINVAL (Invalid argument) > socket(PF_??? (0x5), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x5), SOCK_DGRAM, , 0) = -1 ENOSYS (Function not implemented) > socket(PF_??? (0x5), SOCK_DGRAM, , 0) = -1 EINVAL (Invalid argument) (I am not sure whether these system calls have been taken out of the new ifconfig, or whether I merely configured net-tools to be ignorant of appletalk, etc.) Something about my old ifconfig must be triggering a bug (or hardware error?) somewhere. I am willing to take further suggestions for experiments to try, if anyone is still interested. -Paul (please cc: relevant messages to me)