From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vasiliy G Tolstov Subject: Re: RE: "xm create" take more time to return. Date: Wed, 24 Nov 2010 15:26:48 +0300 Message-ID: <1290601608.2890.19.camel@vase> References: <749B9D3DBF0F054390025D9EAFF47F22301CB45C@shsmsx501.ccr.corp.intel.com> <749B9D3DBF0F054390025D9EAFF47F223023AB34@shsmsx501.ccr.corp.intel.com> <1290601106.2890.17.camel@vase> Reply-To: v.tolstov@selfip.ru Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Stefano Stabellini Cc: "xen-devel@lists.xensource.com" , Jackson , Ian@lonpmailmx01.citrite.netIan, Keir Fraser , "Zhang, Yang Z" List-Id: xen-devel@lists.xenproject.org On Wed, 2010-11-24 at 12:23 +0000, Stefano Stabellini wrote: > On Wed, 24 Nov 2010, Vasiliy G Tolstov wrote: > > On Wed, 2010-11-24 at 11:14 +0800, Zhang, Yang Z wrote: > > > No one care about the stress case? Or this issue only happen to me? > > > > > > best regards > > > yang > > > > > > > I'm confirm this. I use SLES latest version (hello suse!) After some > > time (xm create/destroy) xenstored takes 70-80 cpu time. xentop not > > working. strace -ff -v -F say's that xenstored in cicle get records.... > > > > When i try tdbtool /var/lib/xenstored/tdb check it says that database > > correct. > > > > Problem does not dissapeared after all domains is destroyed. > > > > Could you please post the output of xenstore-ls after you start to see > the problem? Ok top: 10446 root 0 -20 8816 1448 616 R 90 0.0 1458:35 xenstored sample strace output: select(48, [4 5 6 12 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 46 47], [39 40 46], NULL, NULL) = 5 (in [19 38], out [39 40 46]) read(38, "\2\0\0\0\0\0\0\0\0\0\0\0\25\0\0\0", 16) = 16 read(19, "/local/domain/614/vm\0", 21) = 21 fcntl(49, F_SETLKW, {type=F_RDLCK, whence=SEEK_SET, start=12660, len=1}) = 0 lseek(49, 12664, SEEK_SET) = 12664 read(49, "\0\0\0\0", 4) = 4 fcntl(49, F_SETLKW, {type=F_UNLCK, whence=SEEK_SET, start=12660, len=1}) = 0 fcntl(49, F_SETLKW, {type=F_RDLCK, whence=SEEK_SET, start=18368, len=1}) = 0 lseek(49, 18372, SEEK_SET) = 18372 read(49, " I suspect xm destroy doesn't remove all the xenstore entries created by > xm create... > -- Vasiliy G Tolstov Selfip.Ru