Date: Mon, 25 Feb 2002 12:50:59 +0100 From: Bernd Walter <ticso@cicely8.cicely.de> To: Christian Weisgerber <naddy@mips.inka.de> Cc: freebsd-alpha@FreeBSD.ORG Subject: Re: -CURRENT lockups Message-ID: <20020225115058.GB72127@cicely8.cicely.de> In-Reply-To: <a5c8uj$fr$1@kemoauc.mips.inka.de> References: <a59h09$149$1@kemoauc.mips.inka.de> <20020224140250.GG64936@cicely8.cicely.de> <a5b5o5$2gg$1@kemoauc.mips.inka.de> <20020224180950.GA69066@cicely8.cicely.de> <a5c8uj$fr$1@kemoauc.mips.inka.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Feb 25, 2002 at 02:53:07AM +0000, Christian Weisgerber wrote: > Bernd Walter <ticso@cicely8.cicely.de> wrote: > > > If it's realy with a lockmgr lock you should enable debugging for it: > > options KTR > > options KTR_EXTEND > > options KTR_ENTRIES=1024 > > options KTR_COMPILE="(KTR_LOCKMGR)" > > options KTR_MASK=KTR_LOCKMGR > > options KTR_CPUMASK=0x3 > > options KTR_VERBOSE > > This doesn't seem realistic. Just how often is lockmgr() called? > > > In case it's to chatty on booting you can set KTR_MASK to 0 and enable > > it later with sysctl debug.ktr.mask=0x00020000. > > Well, after an hour the boot hasn't gone anywhere. Sorry - I should have known. The vebose output slows things to much down. Usually one would disable verboseity and check the buffer from ddb. I have no idea about how difficult it is to read the buffer from srm. Interestingly I had two full hangs with my noname yesterday. SCSI channel was active but the box did not even react on the halt button. All I did was an installworld on a local disc with src and obj on nfs. -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020225115058.GB72127>