Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mel.connect.com.au!munnari.OZ.AU!news.hawaii.edu!ames!lll-winken.llnl.gov!sol.ctr.columbia.edu!news.mindlink.net!uniserve!usenet From: tom@uniserve.com (Tom Samplonius) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: sendto() buffer exhausted in 2.1.0-RELEASE... Date: 20 Jan 1996 07:19:51 GMT Organization: UNIServe Online Lines: 30 Distribution: world Message-ID: <4dq52n$gec@atlas.uniserve.com> References: <4dotbj$2fo@NNTP.MsState.Edu> NNTP-Posting-Host: pc.sdf.com Mime-Version: 1.0 Content-Type: Text/Plain; charset=ISO-8859-1 X-Newsreader: WinVN 0.99.6 In article <4dotbj$2fo@NNTP.MsState.Edu>, simmons@aris.com¯ says... > >I've recently had a server running FreeBSD 2.1.0-RELEASE >fail because the named locked up (and this machine depends >on its named for name resolution). A ktrace revealed that >the sendto() system call did not have access to buffer space: > > 10273 named CALL sendto(0x3,0xefbfd034,0x90,0,0,0) > 10273 named RET sendto -1 errno 55 No buffer space available > >Which the sendto(2) man page describes as: > > [ENOBUFS] The system was unable to allocate an internal buffer. The > operation may succeed when buffers become available. > >I had to reboot the machine to clear up the problem. What would cause >the kernel buffers to become exhausted? Is there a known bug or memory >leak in 2.1.0-RELEASE that would lead to this? We just upgraded the >kernel to 2.1.0-RELEASE recently. > >Thanks in advance for any insight... You just need more buffers (mbufs actually) for your application. Compile a kernel with the maxuser setting higher than you have now. Handling of a mbuf shortage is poor right now. Tom