Date: Thu, 13 May 1999 16:55:22 -0400 (EDT) From: Jim Cassata <jim@web-ex.com> To: Benjamin Gavin <gavinb@supranet.net> Cc: freebsd-stable@FreeBSD.ORG Subject: Re: BIND 8.2 Problems when reloading Message-ID: <Pine.BSF.4.10.9905131651120.10040-100000@Homer.Web-Ex.com> In-Reply-To: <4.1.19990513153114.00bbec50@mail.supranet.net>
next in thread | previous in thread | raw e-mail | index | archive | help
we downgraded from 8.2 to 8.1.2 for problems under bsd/os and freebsd 2.2.8. CPU usage went through the roof with 8.2. Jim Cassata 516.421.6000 jim@web-ex.com Web Express 20 Broadhollow Road Suite 3011 Melville, NY 11747 On Thu, 13 May 1999, Benjamin Gavin wrote: > Hi all, > I was just writing to see if anyone else has experienced this problem > under FreeBSD 3.1-STABLE. I am trying to run BIND-8.2 (from ports) and am > getting some really weird behavior. The name server seems to come up fine > and serves up names unhindered for an indefinite period (probably > "forever", but it hasn't gotten that far.). However, every once in a while > (OK, probably once every couple of days), someone wants to make a change to > their zone files. This works fine, then a named.reload is run. > > named.reload is a wrapper script that we have customized to grab some > files with "mirror" the to run "ndc reload". This works fine and dandy > about 2 of 3 times. Every third time or so the program hangs and the > following is dumped to the server logs: > > May 13 15:14:04 ox named[81161]: ctl_wrtimeout: /var/run/ndc: write > timeout, closing > May 13 15:14:04 ox named[81161]: ctl_srvr.c:660: REQUIRE(sess->state == > writing) failed. > May 13 15:14:04 ox named[81161]: ctl_srvr.c:660: REQUIRE(sess->state == > writing) failed. > May 13 15:14:04 ox /kernel: pid 81161 (named), uid 0: exited on signal 6 > (core dumped) > > This same set of messages will appear without delay approximately every > third time, if not by then it happens very shortly after. This doesn't > seem to have anything to do with how close the commands are run after each > other. I have successfully run 5 reloads in a row (within 1 sec of each > other), then the 6th dies. On another occasion the first one I run (first > time it has been reloaded in about 10 hrs) it dies straight out. The > messages are always the same. Has anyone seen this problem before. I have > another system running BIND-8.2 and 3.2-BETA that doesn't seem to exhibit > this problem. I have included the system information below: > > Pentium 133 > 128MB RAM > SCSI Hard Drives > 3Com 3C905B Ethernet > 3.1-STABLE (CVSUP/kernel reinstall 4/26/1999) > > The only difference between the working machine and the one that doesn't > work seems to be the OS version and the number of zones being served. It > seems to be a problem with named connecting to a file-based control socket. > Any ideas?? > > Thanks, > Ben Gavin > > /--------------------------------------------------------------------------/ > Benjamin Gavin - Senior Consultant > > *********** NO SPAM!! ************ > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.10.9905131651120.10040-100000>