Hi All, this is my first post on xen-users, so accept my apologies for my errors in posting... I''ve just installed a SuSE 10.3 64 bit and I''m trying to create three domUs to migrate three existing servers. I''ve used Xen from SuSE distro, no modification. The domUs configuration is / and swap on file, and /var on software raid1 exported as partition. They are old SuSE servers I''m trying to consolidate on a single server. If one of the domUs go under load, randomly the whole system starts to slow down and then to freeze, and I need to hardware reset the server. Sometimes (if I see it slowing down) I can connect to the domUs via ssh to send shutdown -h now and so gain a little bit of time to reboot the dom0. The server is a Core 2 Quad on a Asus P5K-VM, 6Gb RAM. During tests domUs are 2 vcpu / 512Mb ram / 1Gb swap 2 vcpu / 512Mb ram / 1Gb swap 4 vcpu / 2Gb ram / 2 Gb swap Can someone help me to solve this? There is something else I can say to help identify the problem? Thanks, Valerio Granato _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
I have the same problems on different hardware and os. If i run a domU with only one core, all works fine. Allready test it with only one core? Valerio Granato schrieb:> Hi All, > > this is my first post on xen-users, so accept my apologies > for my errors in posting... > > I''ve just installed a SuSE 10.3 64 bit and I''m trying to create > three domUs to migrate three existing servers. > I''ve used Xen from SuSE distro, no modification. > > The domUs configuration is / and swap on file, and /var on > software raid1 exported as partition. > They are old SuSE servers I''m trying to consolidate on a single server. > > If one of the domUs go under load, randomly the whole system > starts to slow down and then to freeze, and I need to hardware > reset the server. Sometimes (if I see it slowing down) I can > connect to the domUs via ssh to send shutdown -h now and > so gain a little bit of time to reboot the dom0. > > The server is a Core 2 Quad on a Asus P5K-VM, 6Gb RAM. > During tests domUs are > > 2 vcpu / 512Mb ram / 1Gb swap > 2 vcpu / 512Mb ram / 1Gb swap > 4 vcpu / 2Gb ram / 2 Gb swap > > Can someone help me to solve this? There is something else > I can say to help identify the problem? > > Thanks, > Valerio Granato > > > > _______________________________________________ > Xen-users mailing list > Xen-users@lists.xensource.com > http://lists.xensource.com/xen-users >_______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
I suggest you to upgrade to the last xen an kernel packages. I had the same problem with an unlucky combination of xen package and kernel. Marco 2008/5/14 MS <ms@toefke.com>:> I have the same problems on different hardware and os. > If i run a domU with only one core, all works fine. > Allready test it with only one core? > > Valerio Granato schrieb: >> >> Hi All, >> >> this is my first post on xen-users, so accept my apologies >> for my errors in posting... >> >> I''ve just installed a SuSE 10.3 64 bit and I''m trying to create >> three domUs to migrate three existing servers. >> I''ve used Xen from SuSE distro, no modification. >> >> The domUs configuration is / and swap on file, and /var on >> software raid1 exported as partition. >> They are old SuSE servers I''m trying to consolidate on a single server. >> >> If one of the domUs go under load, randomly the whole system >> starts to slow down and then to freeze, and I need to hardware >> reset the server. Sometimes (if I see it slowing down) I can >> connect to the domUs via ssh to send shutdown -h now and >> so gain a little bit of time to reboot the dom0. >> >> The server is a Core 2 Quad on a Asus P5K-VM, 6Gb RAM. >> During tests domUs are >> >> 2 vcpu / 512Mb ram / 1Gb swap >> 2 vcpu / 512Mb ram / 1Gb swap >> 4 vcpu / 2Gb ram / 2 Gb swap >> >> Can someone help me to solve this? There is something else >> I can say to help identify the problem? >> >> Thanks, >> Valerio Granato >> >> >> >> _______________________________________________ >> Xen-users mailing list >> Xen-users@lists.xensource.com >> http://lists.xensource.com/xen-users >> > > > _______________________________________________ > Xen-users mailing list > Xen-users@lists.xensource.com > http://lists.xensource.com/xen-users >_______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
---- Original Message ---- From: "MS" <ms@toefke.com> To: <xen-users@lists.xensource.com> Sent: Wednesday, May 14, 2008 8:09 AM Subject: Re: [Xen-users] Xen randomly freezes under load> I have the same problems on different hardware and os. > If i run a domU with only one core, all works fine. > Allready test it with only one core?Yes, tried to test using more memory, less memory, one vcpu, same vcpu count on all domains and so on. On the last test it made, after posting, it seems that killing hald and powersaved on the dom0 helps to recover from freeze. I''ve disabled them and the server has been up for about 8 hours, until I tried to create a 4Gb file for a loopback device: dom0 crashed and I had to hw reset the server. Pheraphs I''ve got the 2.3 xenfaq point (hangs under heavy I/O)? Valerio _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
---- Original Message ---- From: "Marco Strullato" <marco.strullato@gmail.com> To: "MS" <ms@toefke.com> Cc: <xen-users@lists.xensource.com> Sent: Wednesday, May 14, 2008 8:31 AM Subject: Re: [Xen-users] Xen randomly freezes under load> I suggest you to upgrade to the last xen an kernel packages. I had the > same problem with an unlucky combination of xen package and kernel.I''m using Xen version 3.1.0_15042-51.3 (abuild@suse.de) (gcc version 4.2.1 (SUSE Linux)) Thu Dec 20 19:57:34 UTC 2007 Latest ChangeSet: 15042 Linux version 2.6.22.17-0.1-xen (geeko@buildhost) (gcc version 4.2.1 (SUSE Linux)) #1 SMP 2008/02/10 20:01:04 UTC (suse standard packages) Which version you use successfully? Valerio _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
> Which version you use successfully?I''m using xen 3.2.0 rebuilt from the src.rpm and the kernel kernel-xen-2.6.18-53.1.14.el5.centos.plus Marco _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users
Valerio Granato
2008-May-15 09:04 UTC
Re: [Xen-users] kernel BUG at... (was: Xen randomly freezes under load)
---- Original Message ---- From: "Valerio Granato" <contact@livenet.it> To: <xen-users@lists.xensource.com> Sent: Wednesday, May 14, 2008 9:25 AM Subject: Re: [Xen-users] Xen randomly freezes under load> On the last test it made, after posting, it seems that killing > hald and powersaved on the dom0 helps to recover from freeze.Ok, without these two processes the server has been up for 24h without freezes or slowdowns. This morning I''ve added another domU, that is saying kernel: kernel BUG at mm/slab.c:1673! kernel: invalid opcode: 0000 [1] SMP kernel: last sysfs file: /kernel/uevent_seqnum kernel: CPU 2 kernel: Modules linked in: ipv6 reiserfs xenblk xennet kernel: Pid: 24, comm: events/2 Tainted: G N 2.6.22.17-0.1-xen #1 kernel: RIP: e030:[<ffffffff80277e5b>] [<ffffffff80277e5b>] kmem_freepages+0xb6/0xfb kernel: RSP: e02b:ffff880000f01df0 EFLAGS: 00010206 kernel: RAX: 0000000000010028 RBX: ffff8800026c7948 RCX: 0000000000000124 kernel: RDX: 0000000000000000 RSI: 0000000000000009 RDI: ffffffff804ab580 kernel: RBP: 0000000000000001 R08: ffffffff804ab74a R09: ffff880000f5a7d0 kernel: R10: ffff880000f67110 R11: 0000000000000002 R12: ffff880004e3c980 kernel: R13: ffff88003134f000 R14: ffff880004e3c980 R15: ffff880001bfdc00 kernel: FS: 00002ad9c46c16f0(0000) GS:ffffffff804dc100(0000) knlGS:0000000000000000 kernel: CS: e033 DS: 0000 ES: 0000 kernel: Process events/2 (pid: 24, threadinfo ffff880000f00000, task ffff880000f5a7d0) kernel: Stack: ffff880000f67110 ffff880004e3c980 ffff88007134f000 0000000000000001 kernel: ffff880001bfdbe0 ffffffff802782c4 0000000000000000 ffff880001bfdbc0 kernel: ffff88007134f000 ffffffff802785e6 00000003010173b0 ffff880001bfdbc0 kernel: Call Trace: kernel: [<ffffffff802782c4>] slab_destroy+0x3c/0x5b kernel: [<ffffffff802785e6>] drain_freelist+0xcb/0xeb kernel: [<ffffffff802797d9>] cache_reap+0x0/0x104 kernel: [<ffffffff80279888>] cache_reap+0xaf/0x104 kernel: [<ffffffff802388d9>] run_workqueue+0xa9/0x167 kernel: [<ffffffff80239306>] worker_thread+0xe1/0xf0 kernel: [<ffffffff8023c28e>] autoremove_wake_function+0x0/0x2e kernel: [<ffffffff80239225>] worker_thread+0x0/0xf0 kernel: [<ffffffff8023c157>] kthread+0x47/0x74 kernel: [<ffffffff80209e48>] child_rip+0xa/0x12 kernel: [<ffffffff8023c110>] kthread+0x0/0x74 kernel: [<ffffffff80209e3e>] child_rip+0x0/0x12 kernel: kernel: Code: 0f 0b eb fe 0f ba 33 07 48 83 c3 38 48 ff ca 48 83 fa ff 75 kernel: RIP [<ffffffff80277e5b>] kmem_freepages+0xb6/0xfb kernel: RSP <ffff880000f01df0> What this one could be? Valerio _______________________________________________ Xen-users mailing list Xen-users@lists.xensource.com http://lists.xensource.com/xen-users