From owner-freebsd-stable Sun Dec 30 8:31: 1 2001 Delivered-To: freebsd-stable@freebsd.org Received: from bastion.internal.lustygrapes.net (dhcp065-024-083-096.columbus.rr.com [65.24.83.96]) by hub.freebsd.org (Postfix) with ESMTP id 02A8D37B416 for ; Sun, 30 Dec 2001 08:30:58 -0800 (PST) Received: from nivomede.internal.lustygrapes.net (nivomede.internal.lustygrapes.net [192.168.10.65]) by bastion.internal.lustygrapes.net (Postfix) with ESMTP id 7A6435A3E; Sun, 30 Dec 2001 11:30:55 -0500 (EST) Date: Sun, 30 Dec 2001 11:30:54 -0500 (EST) From: Brian McDonald X-X-Sender: To: Chris Johnson Cc: Subject: Re: syslogd blocking => can't su to root In-Reply-To: <20011230081106.A98698@palomine.net> Message-ID: <20011230111949.G2732-100000@nivomede.internal.lustygrapes.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Sun, 30 Dec 2001, Chris Johnson wrote: > Date: Sun, 30 Dec 2001 08:11:06 -0500 > From: Chris Johnson > To: stable@FreeBSD.ORG > Subject: syslogd blocking => can't su to root > [snip] > > I discovered this morning that I was unable to su to root, though I had been > able to last night. After I enter the password (whether it is correct or > incorrect), su just hangs. This is on a FreeBSD 4.4-RELEASE box. I've seen this on 4.1-RELEASE boxes as well. Was hoping it would just go away when I upgrade them to 4.5. I noticed it when it blocked other processes waiting for syslog, such as named. [snip] > Questions: > > 1. What might have caused this behavior? I've never seen it on any of the > dozens of FreeBSD boxes I've administered over a period of several years. Dunno. In both cases, syslogd had been running for over 200+ days. Perhaps there is a leak? If you have it happening often, it might help to ktrace syslogd or kill -6 it to get a core file. > 2. Can I kill syslogd and restart it to solve my immediate problem? I killed syslogd and removed /var/run/log (the socket). I found that if I didn't remove the socket file, it didn't always come back. [snip] > 4. Is there any way to fix the problem remotely? I suspect not, but I can have > someone sit down at a console to fix it in a few days. I didn't see anything obvious. Normally syslog is so stable I wouldn't worry about it, but restricting the ability to assume root and fix problems because syslog isn't running correctly seems bad. Mmmm, junior hacker project.. Brian -- Brian McDonald, MCP Klein bottle for sale. Inquire within. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message