[SOLVED] XEN Error: /etc/xen/scripts/block: Failed to find an unused loop device

bsd · 05-18-2016, 12:26 PM

Hi,

We're using Xen for about 2 years. We use CentOS 6.7 as host. We always install Xen with the directives on https://wiki.centos.org/HowTos/Xen/Xen4QuickStart

We use libvirt and virt-manager for installing new VMs.

There's no problem on our old servers which are using Xen 4.4. But on our recently installed servers with Xen 4.6, we cannot create VMs more than 5-6.

The error says:

Code:

logger: /etc/xen/scripts/block: Failed to find an unused loop device

We see that our old servers with Xen 4.4 are using tapdisk. But somehow although we follow the same tutorial, the recently installed servers are using loop system. And we don't know how to fix that.

Here's an old server with Xen 4.4 without any error:

Code:

release                : 3.18.17-13.el6.x86_64
version                : #1 SMP Tue Jul 21 17:17:17 UTC 2015
machine                : x86_64
nr_cpus                : 8
max_cpu_id             : 7
nr_nodes               : 1
cores_per_socket       : 4
threads_per_core       : 2
cpu_mhz                : 3591
hw_caps                : bfebfbff:2c100800:00000000:00007f00:77fafbff:00000000:00000021:00002fbb
virt_caps              : hvm
total_memory           : 32645
free_memory            : 12664
sharing_freed_memory   : 0
sharing_used_memory    : 0
outstanding_claims     : 0
free_cpus              : 0
xen_major              : 4
xen_minor              : 4
xen_extra              : .3-3.el6
xen_version            : 4.4.3-3.el6
xen_caps               : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 hvm-3.0-x86_32p hvm-3.0-x86_64 
xen_scheduler          : credit
xen_pagesize           : 4096
platform_params        : virt_start=0xffff800000000000
xen_changeset          : Wed Oct 21 16:18:30 2015 +0100 git:5418e02-dirty
xen_commandline        : dom0_mem=1024M,max:1024M loglvl=all guest_loglvl=all
cc_compiler            : gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
cc_compile_by          : mockbuild
cc_compile_domain      : centos.org
cc_compile_date        : Thu Oct 29 12:18:52 UTC 2015
xend_config_format     : 4

Xen 4.6 server with problem:

Code:

release                : 3.18.25-19.el6.x86_64
version                : #1 SMP Wed Mar 16 19:11:01 UTC 2016
machine                : x86_64
nr_cpus                : 8
max_cpu_id             : 63
nr_nodes               : 1
cores_per_socket       : 4
threads_per_core       : 2
cpu_mhz                : 3292
hw_caps                : bfebfbff:2c100800:00000000:00007f00:77fafbff:00000000:00000021:00002fbb
virt_caps              : hvm hvm_directio
total_memory           : 32733
free_memory            : 25079
sharing_freed_memory   : 0
sharing_used_memory    : 0
outstanding_claims     : 0
free_cpus              : 0
xen_major              : 4
xen_minor              : 6
xen_extra              : .1-5.el6
xen_version            : 4.6.1-5.el6
xen_caps               : xen-3.0-x86_64 xen-3.0-x86_32p hvm-3.0-x86_32 hvm-3.0-x86_32p hvm-3.0-x86_64 
xen_scheduler          : credit
xen_pagesize           : 4096
platform_params        : virt_start=0xffff800000000000
xen_changeset          : Tue Mar 29 11:02:43 2016 +0100 git:2e19973-dirty
xen_commandline        : dom0_mem=1024M,max:1024M cpuinfo com1=115200,8n1 console=com1,tty loglvl=all guest_loglvl=all
cc_compiler            : gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-16)
cc_compile_by          : mockbuild
cc_compile_domain      : centos.org
cc_compile_date        : Tue Mar 29 12:12:37 UTC 2016
xend_config_format     : 4

I'd be very happy if someone help me to solve this problem.

Best Regards.

dijetlo · 05-21-2016, 01:34 AM

OK, your kernel has a loop device module and it creates a preset number of local loop devices. It can be manually set by editing/creating /etc/modprobe.d/local-loop to read

options loop max_loop=512

Reboot your Xen host and you should have more loop devices than you need.

bsd · 05-21-2016, 04:01 AM

Quote:

Originally Posted by dijetlo

OK, your kernel has a loop device module and it creates a preset number of local loop devices. It can be manually set by editing/creating /etc/modprobe.d/local-loop to read

options loop max_loop=512

Reboot your Xen host and you should have more loop devices than you need.

First of all, thank you for your reply.

I've already tried what you suggest but it doesn't work. Creating a modprobe config file and rebooting the server doesn't effect.

However, I've just tried to add it to the grup.conf and it worked.

Code:

module /vmlinuz-3.18.32-20.el6.x86_64 ro root=UUID=7f717261-b48b-4734-be07-8a3e10c600cd rd_NO_LUKS rd_NO_LVM LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 rd_MD_UUID=ddc4c099:eda4fb73:e4c22436:59825f8e crashkernel=auto  KEYBOARDTYPE=pc KEYTABLE=us rd_NO_DM rhgb quiet max_loop=512

So, modprobe configuration doesn't work but grup.conf configuration with max_loop=512 works.

Thank you very much for your help. I'm updating the thread as solved.

Best Regards.