Re: [PATCH OSSTEST v4] Set {ident}_suite runvar when install a Debian guest.

[Ian Campbell <ian.campbell@citrix.com>]

On Tue, 2015-11-24 at 15:14 +0000, Ian Campbell wrote:
> ---
> I've run an adhoc test with:
> OSSTEST_JOBS_ONLY=
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:build-amd64
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:build-amd64-pvops
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:test-amd64-amd64-qemuu-nested # nested
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:test-amd64-amd64-xl-qemuu-debianhvm-amd64 # regular HVM
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:test-amd64-amd64-xl-qcow2 # DI install
> OSSTEST_JOBS_ONLY=$OSSTEST_JOBS_ONLY:test-amd64-amd64-xl # Normal PV
> export OSSTEST_JOBS_ONLY
> ( ;-) )
> 
> The three HVM ones have passed and the PV one is well past the point where any
> of this would make a difference (it's at 17.ts-guest-localmigrate)

I did notice one oddity, which I think is unrelated to this series which is
that test-amd64-amd64-qemuu-nested/18.ts-leak-check.log contains:

    2015-11-24 14:49:41 Z [host l1] L1 host l1: guest l1 (in scape-moth) 5a:36:0e:bd:00:03 10.80.238.243 
    2015-11-24 14:49:41 Z executing ssh ...     root@10.80.238.243     xl list 
    2015-11-24 14:49:41 Z executing ssh ...     root@10.80.238.243     ps -wwef 
    2015-11-24 14:49:42 Z executing ssh ...     root@10.80.238.243     ps -wwef 
    2015-11-24 14:49:42 Z LEAKED [process 2000 /usr/sbin/ntpd] process: ntp       2000     1  0 14:30 ?        00:00:00 /usr/sbin/ntpd -p /var/run/ntpd.pid -g -c /var/lib/ntp/ntp.conf.dhcp -u 107:114 
    2015-11-24 14:49:42 Z executing ssh ...     root@10.80.238.243     xenstore-ls -fp 
    2015-11-24 14:49:42 Z executing ssh ...     root@10.80.238.243     losetup -a 
    2015-11-24 14:49:42 Z executing ssh ...     root@10.80.238.243     find /tmp /var/run /var/tmp /var/lib/xen /var/core ! -type d -print0 -ls -printf '\0' 
    2015-11-24 14:49:42 Z FAILURE: 1 leaked object(s) 

While e.g. http://logs.test-lab.xenproject.org/osstest/logs/65050/test-amd6
4-amd64-qemuu-nested-intel/info.html did not suffer from this.

In leak-basis-l1.guest.osstest in my flight I see:
    root      1531     1  0 14:30 ?        00:00:00 /bin/sh /etc/init.d/ntp start
compared with 65050 where I see 
    ntp       4154     1  0 07:47 ?        00:00:00 /usr/sbin/ntpd -p /var/run/ntpd.pid -g -u 103:107

So this seems like a race of some sort.

L1 serial console ends:
         Starting LSB: exim Mail Transport Agent...^M
         Starting LSB: Apache2 web server...^M
         Starting LSB: Confirm fully booted...^M
         Starting LSB: Start NTP daemon...^M
         Starting LSB: Start/stop xenstored and xenconsoled...^M
         Starting D-Bus System Message Bus...^M
[  OK  ] Started D-Bus System Message Bus.^M
         Starting System Logging Service...^M
         Starting Permit User Sessions...^M
         Starting ACPI event daemon...^M
[  OK  ] Started ACPI event daemon.^M
[  OK  ] Started /etc/rc.local Compatibility.^M
[  OK  ] Started LSB: Confirm fully booted.^M
[  OK  ] Started Permit User Sessions.^M
         Starting Getty on tty1...^M
[  OK  ] Started Getty on tty1.^M
         Starting Serial Getty on hvc0...^M
[  OK  ] Started Serial Getty on hvc0.^M
[  OK  ] Reached target Login Prompts.^M
[  OK  ] Started Login Service.^M
[  OK  ] Started System Logging Service.^M
[  OK  ] Started LSB: exim Mail Transport Agent.^M
[  OK  ] Started LSB: Apache2 web server.^M
[  OK  ] Started LSB: Start/stop xenstored and xenconsoled.^M

(I had to strip a bunch of colour/escaping, applying the FANCYTTY=0 to
these guests would be useful, so would timestamps).

Ian.

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel