similar to: Bug#686778: xcp-xapi: what user should xapi run as

Package: xcp-xapi Version: 1.3-15 Severity: normal By default, xapi listens on the loopback interface. To make it available over the network, its inventory needs to be updated with the bridged interface details. For example, if xenbr0 is the bridged interface, it needs to be set it: lnx200-39:~# cat /etc/xcp/inventory BUILD_NUMBER='0' CURRENT_INTERFACES='xenbr0'

Package: xcp-xapi Version: 1.3-15 Severity: wishlist The current initscript has the possibility of slowing down the boot process by 50 seconds. If this is required, it probably should be backgrounded. # Wait for xapi to write its "init complete" cookie: after here it's safe to modify templates. wait_for_xapi() { MAX_RETRIES=50 RETRY=0 while [ ${RETRY} -lt ${MAX_RETRIES}

Package: xcp-xapi Version: 1.3-15 Severity: normal My packages were initially installed from the deb repository at download.xensource.org. Eventually, I upgraded them to the debian repository ones. Right now, xcp-xapi fails to start with the following messages. lnx200-39:~# /etc/init.d/xcp-xapi start Error: Connection refused (calling connect ) lnx200-39:~# -- System Information: Debian

Package: xcp-xapi Severity: important The README doc mentions: 4/ Configuring Xen and XCP -------------------------- 4.1/ Switch using xl instead of xm and xend to control Xen As of writing, you should do that: sed -i "s/TOOLSTACK=/TOOLSTACK=xl/" /etc/default/xen Where as the init script checks for: # Exit cleanly if TOOLSTACK != xapi if [ "${TOOLSTACK}" !=

Package: xcp-xapi Version: 1.3.2-6 Severity: important Looks like on Wheezy, /etc/default/xen is not created by xen-utils-common. Since xcp-xapi depends on its existence and further declaration of the toolstack in it, it should take extra care to ensure that it is present. root at debian:~# aptitude install xcp-xapi The following NEW packages will be installed: blktap-dkms{a} blktap-utils{a}

Package: xcp-xapi Severity: normal Tags: upstream Are there good reasons why xapi by default prefers to bidn to 80 and 443? This will lead to xapi installation/start failure on every server machine that might (and will probably) have apache installed. I know we can remove apache or change its port or change xapi's port (But I've had a hard time to easily find out where to specify the

Package: xcp-networkd Version: 1.3.2-11 Severity: normal Given that openvswitch is one of the networking modes, it should be in the Recommends: than being in the Depends: -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (990, 'testing'), (500, 'unstable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-3-amd64

I just noticed OpenXenCenter (aka OpenXenManager) that is an open source replacement for Citrix's XenCenter. This tool works with XenServer. But as per the blog entry: http://blog.xen.org/index.php/2010/02/09/xen-cloud-platform-installation/ it is also supposed to work with Xen Cloud Platform. XCP is licensed GPL2 and is a combination of multiple packages. Has the team looked at the

Package: xen-utils-common Version: 4.1.2-3 Severity: important /etc/default/xen, which carries the TOOLSTACK setting, is not honored by xend init script. The xend initscript doesn't check for the value of TOOLSTACK. It just populates it in its initscript. lnx200-39:/etc/init.d# grep TOOLSTACK * xend:TOOLSTACK=$(/usr/lib/xen-common/bin/xen-toolstack 2>/dev/null) xend: if [ -z

Package: xcp-xapi Version: 1.3.2-15 Severity: normal I am trying to pass an SRIOV virtual function in a virtual machine but the attempt fails with the following errors shown in the xcp-xapi.log [20130829T19:29:17.909Z|debug|sriov1|314 UNIX /var/lib/xcp/xapi|VM.start R:9e8e10bd31bb|pciops] PCI devices from other-config:pci to attach: 0/0000:04:00.1 [20130829T19:29:18.012Z|debug|sriov1|314 UNIX

Package: xcp-xapi Version: 1.3.2-8 Severity: important Tags: patch Found race condition (specific only to slave hosts): xcp-xapi can start earlier than xcp-networkd due lack of dependency in init.d script. Syptoms: After reboot slave's xcp does not work: xe (anything): The host failed to acquire an IP address on its management interface and therefore cannot contact the master. log:

Package: xcp-xapi Version: 1.3.2-8 Severity: important During startup /etc/init.d/xendomains script is called. That script is installed with xen-utils-common witch xcp-xapi is depends on. That script calls /usr/sbin/xen as xen list &> /dev/null This 'xen' seems to call xapi: xen list ^Z [1]+ Stopped xen list root at lab-xh3:~# ps f PID TTY STAT TIME

There's no date defined yet, but Wheezy freeze will happen soon. How do we plan on concluding on this bug report? Recap for the actual bug I reported: If you want to start xapi, execute the following steps: 1) Start /etc/init.d/xend 2) Stop /etc/init.d/xend 3) Start /etc/init.d/xcp-xapi Step 1 is required because you need access to /proc/xen/ Step 2 is required because xapi cannot run

Package: xcp-xapi Version: 1.3.2-6 Severity: minor Tags: upstream If xe pif-configure-ip called twice with different IP, it did not remove old IP from interface (but didn't store both addresses in database). (Checked for management interface) Steps to reproduce: xe pif-reconfigure-ip uuid=... ip=IP1 .... xe pif-reconfigure-ip uuid=(same) ip=IP2 ... 'ip addr show' shows both

Package: xcp-xapi Version: 1.3.2-8 Severity: minor During upgrade from 1.3.2-7 to 1.3.2-8 got follwing message on slave host (upgrade was simulatanious on slave and master): ... Setting up xcp-xapi (1.3.2-8) ... The server failed to handle your request, due to an internal error. The given message may give details useful for debugging the problem. message: INTERNAL_ERROR: [

Package: xcp-xapi Version: 1.3.2-6 Severity: important I done normal reboot: xe host-disable host=this_host xe host-reboot host=this_host After reboot host stays in 'disable' state and when I've tried to make it enable I got: xe host-enable uuid=ab523ee3-0546-5400-e2db-32f340292fff The specified host is disabled and cannot be re-enabled until after it has rebooted. host:

Package: xcp-xapi Version: 1.3.2-9 Severity: normal Hi guys, Just a couple of things I noticed running through /usr/share/doc/xcp-xapi/README.Debian.gz as referenced by http://wiki.debian.org/XCP In "4/ Configuring Xen and XCP": sed -i "s/TOOLSTACK=/TOOLSTACK=xapi/" /etc/default/xen should be sed -i "s/TOOLSTACK=.*/TOOLSTACK=xapi/" /etc/default/xen Otherwise

Package: xcp-xapi Version: 1.3.2-6 Severity: important Configuration: Two xcp-xapi hosts (default sid/xcp-xapi installation) in single pool. Network backend is bridge. After migration of 'idle' (doning nothing) virtual machine it stops to reply to ping for long time. AFAIK that caused by mac-learning table on upper level (hardware) switch, which route all frames to wrong port

Package: xcp-xapi Version: 1.3.2-7 Severity: normal xcp-xapi fails to eject host from pool with following error: xe pool-eject host-uuid=a6806a39-442d-f17d-b7a0-59161d18f56f WARNING: Ejecting a host from the pool will reinitialise that host's local SRs. WARNING: Any data contained with the local SRs will be lost. Type 'yes' to continue yes The server failed to handle your request,

Package: xcp-xapi Version: 1.3.2-6 Severity: minor Tags: upstream Builin utility for lowlevel vm boot repairment xe-edit-bootloader contains wrong path to xe. IS: XE="/usr/lib/xcp/bin/xe" MUST BE: XE="/usr/bin/xe" -- System Information: Debian Release: wheezy/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kernel: Linux