]> xenbits.xensource.com Git - osstest.git/log
projects / osstest.git / log
? search:
summary | shortlog | log | commit | commitdiff | tree
first ⋅ prev ⋅ next
osstest.git
9 years agocr-for-branches: Add linux-3.18 flight-57311 flight-57312 flight-57313 flight-57314 flight-57315 flight-57316 flight-57317 flight-57318 flight-57319 flight-57320 flight-57321 flight-57322 flight-57323 flight-57324 flight-57325 flight-57326 flight-57327 flight-57328 flight-57329 flight-57330 flight-57331 flight-57332 flight-57333 flight-57334 flight-57335 flight-57336 flight-57337 flight-57338 flight-57339 flight-57340 flight-57341 flight-57342 flight-57343 flight-57344 flight-57345 flight-57346 flight-57347 flight-57348 flight-57349 flight-57350 flight-57351 flight-57352 flight-57353 flight-57354 flight-57355 flight-57357 flight-57358 flight-57359 flight-57360 flight-57361 flight-57362 flight-57363 flight-57364 flight-57365 flight-57366 flight-57367 flight-57368 flight-57369 flight-57370 flight-57371 flight-57372 flight-57373 flight-57374 flight-57375 flight-57376 flight-57377 flight-57378 flight-57379 flight-57380 flight-57381 flight-57382 flight-57383 flight-57384 flight-57385 flight-57386 flight-57387 flight-57388 flight-57389 flight-57390 flight-57391 flight-57392 flight-57393 flight-57394 flight-57395 flight-57396 flight-57397 flight-57398 flight-57399 flight-57400 flight-57402 flight-57403 flight-57404 flight-57405 flight-57406 flight-57407 flight-57408 flight-57409 flight-57410 flight-57411 flight-57412 flight-57413 flight-57414 flight-57415 flight-57416 flight-57417 flight-57418 flight-57419 flight-57420 flight-57421 flight-57422 flight-57423 flight-57424 flight-57425 flight-57426 flight-57427 flight-57428 flight-57429 flight-57430 flight-57431 flight-57432 flight-57433 flight-57434 flight-57435 flight-57436 flight-57437 flight-57438 flight-57439 flight-57440 flight-57441 flight-57442 flight-57443 flight-57444
commit | commitdiff | tree
Ian Campbell [Tue, 26 May 2015 11:44:09 +0000 (12:44 +0100)]
cr-for-branches: Add linux-3.18

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoMove the code for setting memory size into prep() flight-56996 flight-57080 flight-57081 flight-57082 flight-57083 flight-57084 flight-57085 flight-57086 flight-57087 flight-57088 flight-57089 flight-57090 flight-57091 flight-57092 flight-57093 flight-57094 flight-57095 flight-57096 flight-57097 flight-57098 flight-57099 flight-57100 flight-57101 flight-57102 flight-57103 flight-57104 flight-57105 flight-57106 flight-57107 flight-57108 flight-57109 flight-57110 flight-57111 flight-57112 flight-57113 flight-57115 flight-57116 flight-57117 flight-57118 flight-57119 flight-57120 flight-57121 flight-57122 flight-57123 flight-57124 flight-57125 flight-57126 flight-57127 flight-57128 flight-57129 flight-57130 flight-57131 flight-57132 flight-57133 flight-57134 flight-57135 flight-57136 flight-57137 flight-57138 flight-57139 flight-57140 flight-57141 flight-57142 flight-57143 flight-57144 flight-57145 flight-57146 flight-57147 flight-57148 flight-57149 flight-57150 flight-57151 flight-57152 flight-57153 flight-57154 flight-57155 flight-57156 flight-57157 flight-57158 flight-57159 flight-57160 flight-57161 flight-57162 flight-57163 flight-57164 flight-57165 flight-57166 flight-57167 flight-57168 flight-57169 flight-57170 flight-57171 flight-57172 flight-57173 flight-57174 flight-57175 flight-57176 flight-57177 flight-57178 flight-57179 flight-57180 flight-57181 flight-57182 flight-57183 flight-57184 flight-57185 flight-57186 flight-57187 flight-57188 flight-57189 flight-57190 flight-57191 flight-57192 flight-57193 flight-57194 flight-57195 flight-57196 flight-57197 flight-57198 flight-57199 flight-57200 flight-57201 flight-57202 flight-57203 flight-57204 flight-57205 flight-57206 flight-57207 flight-57208 flight-57209 flight-57210 flight-57211 flight-57212 flight-57213 flight-57214 flight-57215 flight-57216 flight-57217 flight-57218 flight-57219 flight-57220 flight-57221 flight-57222 flight-57223 flight-57224 flight-57225 flight-57226 flight-57227 flight-57228 flight-57229 flight-57230 flight-57231 flight-57232 flight-57233 flight-57234 flight-57235 flight-57236 flight-57237 flight-57238 flight-57239 flight-57240 flight-57241 flight-57242 flight-57243 flight-57244 flight-57245 flight-57246 flight-57247 flight-57248 flight-57249 flight-57250 flight-57251 flight-57252 flight-57253 flight-57254 flight-57255 flight-57256 flight-57257 flight-57258 flight-57259 flight-57260 flight-57261 flight-57262 flight-57263 flight-57264 flight-57265 flight-57266 flight-57267 flight-57268 flight-57269 flight-57270 flight-57271 flight-57272 flight-57273 flight-57274 flight-57275 flight-57276 flight-57277 flight-57278 flight-57279 flight-57280 flight-57281 flight-57282 flight-57283 flight-57284 flight-57285 flight-57286 flight-57287 flight-57288 flight-57289 flight-57290 flight-57291 flight-57292 flight-57293 flight-57294 flight-57295 flight-57296 flight-57297 flight-57298 flight-57299 flight-57300 flight-57301 flight-57302 flight-57303 flight-57304 flight-57305 flight-57306 flight-57307 flight-57308 flight-57310
commit | commitdiff | tree
longtao.pang [Wed, 13 May 2015 03:36:40 +0000 (11:36 +0800)]
Move the code for setting memory size into prep()

Move the code for setting memory size from outside into prep(),
where these $ram_mb operations are only used; And in next patch,
for L2 guest installation, $ram_mb will be overridden by a runvar value,
which needs $gho in scope.

This is pure code motion. No functional change.

Signed-off-by: longtao.pang <longtaox.pang@intel.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoEdit some APIs in TestSupport.pm for nested test
commit | commitdiff | tree
longtao.pang [Wed, 13 May 2015 03:36:39 +0000 (11:36 +0800)]
Edit some APIs in TestSupport.pm for nested test

1. If vif model is defined for guest, use it in guest's configuration
2. In L2 installation context, its host (L1) shall not have a static IP
but stored in runvar in L1 installation before; use this stored runvar
IP.

Signed-off-by: longtao.pang <longtaox.pang@intel.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- resolved conflict with addition of ${viftype}, made
         ${vifmodel} a prefix rather than postfix thing for
         consistency with ${viftype}. ]

9 years agoRefactor installation of overlays
commit | commitdiff | tree
longtao.pang [Wed, 13 May 2015 03:36:38 +0000 (11:36 +0800)]
Refactor installation of overlays

Apply overlay to guests as well as hosts by moving to preseed_base.

We do this because we want to be able to:
       * Use '/etc/init.d/osstest-confirm-booted' for guests as
         well as hosts (in particular when testing nestedhvm
         we sometimes want to treat a guest as a host).
       * Use grub's 20_linux_xen which is harmless unless Xen is
         installed in the guest which is the case only for
         nestedhvm testing when we want these changes.
The other things in the overlay are some initscripts
(xenbridge and xenlightdaemons) which are not enabled for guests and
hence are therefore harmless.

Signed-off-by: longtao.pang <longtaox.pang@intel.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoRevert "cs-bisection-step: allow -bisect blessed flights for basis pass" flight-56820 flight-56821 flight-56822 flight-56823 flight-56824 flight-56825 flight-56826 flight-56827 flight-56828 flight-56829 flight-56830 flight-56831 flight-56832 flight-56833 flight-56834 flight-56835 flight-56836 flight-56837 flight-56838 flight-56839 flight-56840 flight-56841 flight-56842 flight-56843 flight-56844 flight-56845 flight-56846 flight-56847 flight-56848 flight-56849 flight-56850 flight-56851 flight-56852 flight-56853 flight-56854 flight-56855 flight-56856 flight-56857 flight-56858 flight-56859 flight-56860 flight-56861 flight-56862 flight-56863 flight-56864 flight-56865 flight-56866 flight-56867 flight-56868 flight-56869 flight-56870 flight-56871 flight-56872 flight-56873 flight-56874 flight-56875 flight-56876 flight-56877 flight-56878 flight-56879 flight-56880 flight-56881 flight-56882 flight-56883 flight-56884 flight-56885 flight-56886 flight-56887 flight-56888 flight-56889 flight-56890 flight-56891 flight-56892 flight-56893 flight-56894 flight-56895 flight-56896 flight-56897 flight-56898 flight-56899 flight-56900 flight-56901 flight-56902 flight-56903 flight-56904 flight-56905 flight-56906 flight-56907 flight-56908 flight-56909 flight-56910 flight-56911 flight-56912 flight-56913 flight-56914 flight-56915 flight-56916 flight-56917 flight-56918 flight-56919 flight-56920 flight-56921 flight-56923 flight-56924 flight-56925 flight-56926 flight-56927 flight-56928 flight-56929 flight-56930 flight-56931 flight-56932 flight-56933 flight-56934 flight-56935 flight-56936 flight-56937 flight-56938 flight-56939 flight-56940 flight-56941 flight-56942 flight-56943 flight-56944 flight-56945 flight-56946 flight-56947 flight-56948 flight-56949 flight-56950 flight-56951 flight-56952 flight-56953 flight-56954 flight-56955 flight-56956 flight-56958 flight-56959 flight-56960 flight-56961 flight-56962 flight-56963 flight-56964 flight-56965 flight-56966 flight-56967 flight-56968 flight-56969 flight-56970 flight-56971 flight-56972 flight-56973 flight-56974 flight-56975 flight-56976 flight-56977 flight-56978 flight-56979 flight-56980 flight-56981 flight-56982 flight-56983 flight-56984 flight-56985 flight-56986 flight-56987 flight-56988 flight-56989 flight-56990 flight-56991 flight-56992 flight-56993 flight-56994 flight-56995 flight-56997 flight-56998 flight-56999 flight-57000 flight-57001 flight-57002 flight-57003 flight-57004 flight-57005 flight-57006 flight-57007 flight-57008 flight-57009 flight-57010 flight-57011 flight-57012 flight-57013 flight-57014 flight-57015 flight-57016 flight-57017 flight-57018 flight-57019 flight-57020 flight-57021 flight-57022 flight-57023 flight-57024 flight-57025 flight-57026 flight-57027 flight-57028 flight-57029 flight-57030 flight-57031 flight-57032 flight-57033 flight-57034 flight-57035 flight-57036 flight-57037 flight-57038 flight-57039 flight-57040 flight-57041 flight-57042 flight-57043 flight-57044 flight-57045 flight-57046 flight-57047 flight-57048 flight-57049 flight-57050 flight-57051 flight-57052 flight-57053 flight-57054 flight-57055 flight-57056 flight-57057 flight-57058 flight-57059 flight-57060 flight-57061 flight-57062 flight-57063 flight-57064 flight-57065 flight-57066 flight-57067 flight-57068 flight-57069 flight-57070 flight-57071 flight-57072 flight-57073 flight-57074 flight-57075 flight-57076 flight-57077 flight-57078 flight-57079
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 12:43:39 +0000 (13:43 +0100)]
Revert "cs-bisection-step: allow -bisect blessed flights for basis pass"

If the bisector is allowed to consider its own output flights as
candidates for the basis pass, then it will (generally) restart with a
new basis each time it finds a pass.

This is very slow and wasteful.

There is not much explanation in 01edca47 of the change I am now
reverting, but I think I probably created by some semi-manual process
a flight to serve as the basis.  I now think that my mistake was to
bless that flight `adhoc-bisect' or some such, rather than `adhoc'.

This reverts commit 01edca47be3742a1660b1956c1f06ca934b97352.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agocs-bisection-step: Do not treat repro attempts as flail
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 12:36:15 +0000 (13:36 +0100)]
cs-bisection-step: Do not treat repro attempts as flail

The need_repro machinery deliberarely makes attempts to reproduce
various results.

This can cause the flail detector to trigger when not intended.  In
particular, the bisector may have (for some reason[1]) restarted with
a new baseline, and the temporarally-stripy pass/fail requirement
would then require the basis fail to be repro'd, again.

[1] Currently this happens much more often than is desirable.  This
will be fixed in a moment.

Fix this by only considering, for the purposes of flail, flights which
are no older than the first repro (the basis pass).

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoRevert "cs-bisection-step: Abandon repro attempts after a bit"
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 11:30:19 +0000 (12:30 +0100)]
Revert "cs-bisection-step: Abandon repro attempts after a bit"

This safety catch is unnecessary and unhelpful.

It is unnecessary because 489773b4 "Detect flailing" will detect
attempts by the bisector to repeatedly run the same flight and hope
for different results.

It is unhelpful because it can happen for good reasons that a
particular revision has been tested many times.  In particular:

 - The osstest push gate input tree may have not been advanced for a
   long time and been failing its push gate.

 - The bisector may have (for some reason[1]) restarted with a new
   baseline, and the temporarally-stripy pass/fail requirement would
   then require the basis fail to be repro'd, again.

[1] Currently this happens much more often than is desirable.  This
will be fixed in a moment.

This reverts commit 2676277181599a889657354028b992379aa6142b.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agocs-bisection-step: Flail detection: look only at our blessing
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 12:11:41 +0000 (13:11 +0100)]
cs-bisection-step: Flail detection: look only at our blessing

When looking for identical previously flights, consider only ones
which have the same blessing as our prospective flight will have.

There are good reasons why apparently identical flights might appear
with other in-scope blessings: notably, a single-test-job main branch
might produce many failures from its push gate, which would all have
the main branch blessing (rather than the bisector's blessing).

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agocs-bisection-step: Clarify and correct flailing message
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 12:06:37 +0000 (13:06 +0100)]
cs-bisection-step: Clarify and correct flailing message

There are many other possible reasons for this, besides a bug in the
build version machinery.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agocs-bisection-step: Refer to jobs we create just by job name
commit | commitdiff | tree
Ian Jackson [Wed, 20 May 2015 11:49:05 +0000 (12:49 +0100)]
cs-bisection-step: Refer to jobs we create just by job name

When we make a fresh build job, rather than referring to an existing
job in another flight, pass to the rest of the machinery only the job
name, not <flight>.<job>.

This means that the generated flight refers to its own jobs without
specifying the flight number.  This allows the flail detector to
operate properly: without this, we might have repeated attempts to
test and build the same thing, but they would look identical because
their self-referential runvars would be different due to their
different flight numbers.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agots-host-install: die on unknown options flight-56739 flight-56801 flight-56817 flight-56818 flight-56819
commit | commitdiff | tree
Ian Jackson [Mon, 18 May 2015 12:46:07 +0000 (13:46 +0100)]
ts-host-install: die on unknown options

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
9 years agots-host-install: New --poweron-test-only option
commit | commitdiff | tree
Ian Jackson [Fri, 15 May 2015 16:54:03 +0000 (16:54 +0000)]
ts-host-install: New --poweron-test-only option

We are having a difficulty with one of the test boxes, which can be
most easily reproduced by running ts-host-install to power cycle the
box and then see if it wakes up enough to fetch a preseed file.

Keep this mode of operation in tree in case it's useful in future.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agots-host-install: Fix indentation error
commit | commitdiff | tree
Ian Jackson [Fri, 15 May 2015 16:53:33 +0000 (16:53 +0000)]
ts-host-install: Fix indentation error

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoRevert "allow.all: Ignore FreeBSD local migration failures"
commit | commitdiff | tree
Ian Jackson [Sat, 16 May 2015 00:15:30 +0000 (01:15 +0100)]
Revert "allow.all: Ignore FreeBSD local migration failures"

We think these are solved by the new FreeBSD images, in
"FreeBSD: use custom image containing BSD netfront bugfix"

This reverts commit fcb818eeb543fbf9f7d2cbc572531c0a5bf312ef.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoFreeBSD: use custom image containing BSD netfront bugfix
commit | commitdiff | tree
Ian Jackson [Fri, 15 May 2015 17:03:01 +0000 (18:03 +0100)]
FreeBSD: use custom image containing BSD netfront bugfix

Roger has supplied these images which contain a fix for a netfront bug
in FreeBSD: it would send the gratuitous ARP before netback was ready,
so it could get lost, leading to stochastic migration failures:
  https://svnweb.freebsd.org/base?view=revision&revision=282908

Here is the runvar diff:
 - test-amd64-i386-freebsd10-amd64 freebsd_image FreeBSD-10.1-RELEASE-amd64.raw.xz
 - test-amd64-i386-freebsd10-i386  freebsd_image FreeBSD-10.1-RELEASE-i386.raw.xz
 + test-amd64-i386-freebsd10-amd64 freebsd_image FreeBSD-10.1-CUSTOM-amd64-20150518.raw.xz
 + test-amd64-i386-freebsd10-i386  freebsd_image FreeBSD-10.1-CUSTOM-i386-20150518.raw.xz
I have confirmed that these are the right filenames.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
9 years agoDebian: correct u-boot command to load flask policy flight-56691 flight-56734 flight-56740 flight-56741 flight-56742 flight-56743 flight-56744 flight-56745 flight-56746 flight-56747 flight-56748 flight-56749 flight-56750 flight-56751 flight-56754 flight-56755 flight-56756 flight-56757 flight-56758 flight-56759 flight-56760 flight-56761 flight-56762 flight-56763 flight-56764 flight-56765 flight-56766 flight-56767 flight-56768 flight-56769 flight-56770 flight-56771 flight-56772 flight-56773 flight-56774 flight-56775 flight-56776 flight-56777 flight-56779 flight-56780 flight-56781 flight-56782 flight-56783 flight-56784 flight-56785 flight-56786 flight-56787 flight-56788 flight-56789 flight-56790 flight-56791 flight-56792 flight-56793 flight-56794 flight-56795 flight-56796 flight-56797 flight-56798 flight-56799 flight-56800 flight-56802 flight-56803 flight-56804 flight-56805 flight-56806 flight-56807 flight-56808 flight-56809 flight-56810 flight-56811 flight-56812 flight-56813 flight-56814 flight-56815 flight-56816
commit | commitdiff | tree
Ian Campbell [Fri, 15 May 2015 18:54:36 +0000 (19:54 +0100)]
Debian: correct u-boot command to load flask policy

The use of the $flaskpolocy variable is escaped only once, meaning it
is interpreted by the shell on the test host, a context in which it
is not set (it is set in Perl, and it would be set in u-boot env by
the line above). The symptom was that the ext2load command got given
no file to load and so tries to load the default "pxelinux.0":

May 14 20:47:47.108243 ## Executing script at 43100000
May 14 20:47:47.111115 Loading dtbs/sun7i-a20-cubietruck.dtb
May 14 20:47:47.114489 22007 bytes read in 114 ms (188.5 KiB/s)
May 14 20:47:47.252237 820116 bytes read in 114 ms (6.9 MiB/s)
May 14 20:47:47.396931 Loaded xen-4.6-unstable to 0x41000000 (c8394)
May 14 20:47:47.400989 command line: [...]
May 14 20:47:47.413380 4963576 bytes read in 214 ms (22.1 MiB/s)
May 14 20:47:47.653005 Loaded vmlinuz-3.16.7-ckt4+ to 0x42000000 (4bbcf8)
May 14 20:47:47.657999 command line: [...]
May 14 20:47:47.671376 12501593 bytes read in 510 ms (23.4 MiB/s)
May 14 20:47:48.203299 Loaded initrd.img-3.16.7-ckt4+ to 0x43300000 (bec259)
May 14 20:47:48.208406 ** File not found pxelinux.0 **
May 14 20:47:48.294291 Loaded xenpolicy-4.6-unstable to 0x41200000 (bec259)

The filesize in the log message remains the same as the initrd because
nothing is actually loaded so the variable isn't updated.

Rather than adding additional backticks to quote it to be interpreted
by u-boot, remove the quoting so it is interpreted by Perl, making
the setting of the u-boot var (which in any case used the wrong
syntax) unnecessary. This matches what we do for the kernel etc too.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Wei Liu <Wei.Liu2@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoRevert "ts-logs-capture: collect resize2fs.log"
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 17:11:23 +0000 (18:11 +0100)]
Revert "ts-logs-capture: collect resize2fs.log"

This is no longer generated.

This reverts commit 9f299b936d1812bfd958d2771676a8de3cf9c060.

9 years agots-xen-build-prep: mkfs a new /home/osstest, don't resize2fs
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 20:38:27 +0000 (21:38 +0100)]
ts-xen-build-prep: mkfs a new /home/osstest, don't resize2fs

Online resize is 40x slower than mkfs.  It appears that the
backgrounded resize2fs can starve build tasks of IO bandwidth.

So instead, use mkfs to make a new filesystem for /home/osstest.
We use rsync to copy in the old contents.

For convenience of (a) review (b) possible reversion, we keep (for
now) the lvextend machinery.  So we create a new 1-extent LV for the
lvextend machinery to work on.

But we don't call resize2fs when we extend it, because now it doesn't
have a fs on it yet.  We make the filesystem later.

We move the ccache_setup until after this is done because it's a bit
pointless to put things in the to-be-removed /home/osstest when they
could be put in the new one after it had been set up.

We take slight care to make the rune slightly idempotent: if it
completed successfully we detect this and do not run it again.  But if
it didn't, things may be messed up and running it again is unlikely to
help and may make things worse.

I have tested this on rice-weevil and the whole new target command
(including rsync, mkfs, mount etc.) takes 126s.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agots-xen-build-prep: Break out determine_vg_lv
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 20:37:02 +0000 (21:37 +0100)]
ts-xen-build-prep: Break out determine_vg_lv

And make $lv a global.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoTestSupport: break out lv_dev_mapper
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 17:28:55 +0000 (18:28 +0100)]
TestSupport: break out lv_dev_mapper

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
9 years agots-xen-build-prep: Abolish "background" parameter to lvextend
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 17:13:10 +0000 (18:13 +0100)]
ts-xen-build-prep: Abolish "background" parameter to lvextend

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
9 years agoRevert "ts-xen-build-prep: Run the backgrounded resize2fs with ionice"
commit | commitdiff | tree
Ian Jackson [Wed, 11 Jun 2014 17:10:43 +0000 (18:10 +0100)]
Revert "ts-xen-build-prep: Run the backgrounded resize2fs with ionice"

This did not really help.  We are going to do something different.

This reverts commit ee31a32af1fc9a38ad08b4a36ee69e1c58ea0828.

9 years agoDebian: Log which kernel/initrd we failed to copy.
commit | commitdiff | tree
Ian Campbell [Fri, 6 Mar 2015 10:05:34 +0000 (10:05 +0000)]
Debian: Log which kernel/initrd we failed to copy.

Helpful to trackdown which config var I've got wrong.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoi18n/l10n: Make Timezone configurable and change the default
commit | commitdiff | tree
Ian Jackson [Fri, 15 May 2015 10:31:35 +0000 (11:31 +0100)]
i18n/l10n: Make Timezone configurable and change the default

* Introduce a new config option Timezone
* Replace hardcoded Europe/London everywhere with $c{Timezone}
* The default is UTC
* But in production-config-cambridge set it to Europe/London

The overall effect is:
* No change in Cambridge
* Default timezone changes to UTC but can now be overridden
* Production instance timezone changes to UTC

(It appears that there is no reasonable way to find out the Olson TZ
name of the controller host's default timezone.  If there were, or we
discover one, we should arrange that the default is set
appropriately.)

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agots-logs-capture: Collect /var/log/messages and /var/log/debug
commit | commitdiff | tree
Ian Jackson [Thu, 14 May 2015 17:41:21 +0000 (18:41 +0100)]
ts-logs-capture: Collect /var/log/messages and /var/log/debug

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agoproduction-config: Use /home/logs, not /home/osstest/pub
commit | commitdiff | tree
Ian Jackson [Thu, 14 May 2015 17:12:30 +0000 (18:12 +0100)]
production-config: Use /home/logs, not /home/osstest/pub

The logs and images (including .../logs, .../results, etc.) are now on
their own filesystem on the production osstest VM, which I have called
/home/logs.

Changing this in production config will allow us to tidy up by
removing the symlink I left behind.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
9 years agosg-run-job: Repeat rumpuserxen-demo-xenstorels test 150 times flight-56372 flight-56412 flight-56474 flight-56475 flight-56476 flight-56477 flight-56478 flight-56479 flight-56480 flight-56481 flight-56482 flight-56483 flight-56484 flight-56485 flight-56486 flight-56487 flight-56488 flight-56489 flight-56490 flight-56492 flight-56493 flight-56494 flight-56495 flight-56496 flight-56497 flight-56498 flight-56499 flight-56500 flight-56501 flight-56502 flight-56503 flight-56504 flight-56505 flight-56506 flight-56507 flight-56508 flight-56509 flight-56510 flight-56511 flight-56512 flight-56513 flight-56514 flight-56515 flight-56516 flight-56517 flight-56518 flight-56519 flight-56520 flight-56521 flight-56522 flight-56523 flight-56524 flight-56525 flight-56526 flight-56527 flight-56528 flight-56529 flight-56530 flight-56531 flight-56532 flight-56533 flight-56534 flight-56535 flight-56536 flight-56537 flight-56538 flight-56539 flight-56540 flight-56541 flight-56543 flight-56544 flight-56545 flight-56546 flight-56547 flight-56548 flight-56549 flight-56550 flight-56551 flight-56552 flight-56553 flight-56554 flight-56555 flight-56556 flight-56557 flight-56558 flight-56559 flight-56560 flight-56561 flight-56562 flight-56563 flight-56564 flight-56565 flight-56566 flight-56567 flight-56568 flight-56569 flight-56570 flight-56571 flight-56572 flight-56573 flight-56574 flight-56575 flight-56576 flight-56577 flight-56578 flight-56579 flight-56580 flight-56582 flight-56583 flight-56584 flight-56585 flight-56586 flight-56587 flight-56588 flight-56589 flight-56590 flight-56591 flight-56592 flight-56593 flight-56594 flight-56595 flight-56596 flight-56597 flight-56598 flight-56599 flight-56600 flight-56601 flight-56603 flight-56604 flight-56605 flight-56606 flight-56607 flight-56608 flight-56609 flight-56610 flight-56611 flight-56612 flight-56613 flight-56614 flight-56615 flight-56616 flight-56617 flight-56618 flight-56619 flight-56620 flight-56621 flight-56622 flight-56623 flight-56624 flight-56625 flight-56626 flight-56627 flight-56628 flight-56629 flight-56630 flight-56631 flight-56632 flight-56633 flight-56635 flight-56636 flight-56637 flight-56638 flight-56639 flight-56640 flight-56641 flight-56642 flight-56643 flight-56644 flight-56645 flight-56646 flight-56647 flight-56648 flight-56649 flight-56650 flight-56651 flight-56652 flight-56653 flight-56654 flight-56655 flight-56656 flight-56657 flight-56658 flight-56659 flight-56660 flight-56661 flight-56662 flight-56663 flight-56664 flight-56665 flight-56666 flight-56667 flight-56668 flight-56669 flight-56670 flight-56671 flight-56672 flight-56673 flight-56674 flight-56675 flight-56676 flight-56677 flight-56679 flight-56680 flight-56681 flight-56682 flight-56683 flight-56684 flight-56685 flight-56686 flight-56687 flight-56689 flight-56690 flight-56692 flight-56693 flight-56694 flight-56695 flight-56696 flight-56697 flight-56698 flight-56699 flight-56700 flight-56701 flight-56702 flight-56703 flight-56704 flight-56705 flight-56706 flight-56707 flight-56708 flight-56709 flight-56710 flight-56711 flight-56712 flight-56713 flight-56714 flight-56715 flight-56716 flight-56717 flight-56718 flight-56719 flight-56720 flight-56721 flight-56722 flight-56723 flight-56724 flight-56725 flight-56726 flight-56727 flight-56728 flight-56729 flight-56730 flight-56731 flight-56732 flight-56733 flight-56735 flight-56736 flight-56737 flight-56738
commit | commitdiff | tree
Ian Campbell [Fri, 17 Apr 2015 10:01:29 +0000 (11:01 +0100)]
sg-run-job: Repeat rumpuserxen-demo-xenstorels test 150 times

With 50 we are still seeing occasional spurious passes.

Looking at some logs the current 50 iterations takes around 4 minutes
to run (3-4s per iteration).

I think we can justify spending 10 minutes running this test, which is
150x4s.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agomake-flight: Drop pointless jobs from seabios and ovmf flights
commit | commitdiff | tree
Ian Campbell [Tue, 12 May 2015 08:10:33 +0000 (09:10 +0100)]
make-flight: Drop pointless jobs from seabios and ovmf flights

These are both firmware which is used only with qemuu tests. There is
no point testing either PV guets or HVM with qemut (which always uses
rombios).

Comparing before and after of:

./standalone-generate-dump-flight-runvars | \
    awk '// {printf "%-32s %s\n", $1, $2}' | \
   sort -u

Results in:

    @@ -615,48 +615,19 @@
     ovmf                             build-i386-libvirt
     ovmf                             build-i386-pvops
     ovmf                             build-i386-xsm
    -ovmf                             test-amd64-amd64-libvirt
    -ovmf                             test-amd64-amd64-libvirt-xsm
    -ovmf                             test-amd64-amd64-pair
    -ovmf                             test-amd64-amd64-xl
    -ovmf                             test-amd64-amd64-xl-credit2
    -ovmf                             test-amd64-amd64-xl-multivcpu
    -ovmf                             test-amd64-amd64-xl-pvh-amd
    -ovmf                             test-amd64-amd64-xl-pvh-intel
    -ovmf                             test-amd64-amd64-xl-qemut-debianhvm-amd64
    -ovmf                             test-amd64-amd64-xl-qemut-debianhvm-amd64-xsm
    -ovmf                             test-amd64-amd64-xl-qemut-win7-amd64
    -ovmf                             test-amd64-amd64-xl-qemut-winxpsp3
     ovmf                             test-amd64-amd64-xl-qemuu-debianhvm-amd64
     ovmf                             test-amd64-amd64-xl-qemuu-debianhvm-amd64-xsm
     ovmf                             test-amd64-amd64-xl-qemuu-ovmf-amd64
     ovmf                             test-amd64-amd64-xl-qemuu-win7-amd64
     ovmf                             test-amd64-amd64-xl-qemuu-winxpsp3
    -ovmf                             test-amd64-amd64-xl-sedf
    -ovmf                             test-amd64-amd64-xl-sedf-pin
    -ovmf                             test-amd64-amd64-xl-xsm
    -ovmf                             test-amd64-i386-freebsd10-amd64
    -ovmf                             test-amd64-i386-freebsd10-i386
    -ovmf                             test-amd64-i386-libvirt
    -ovmf                             test-amd64-i386-libvirt-xsm
    -ovmf                             test-amd64-i386-pair
    -ovmf                             test-amd64-i386-qemut-rhel6hvm-amd
    -ovmf                             test-amd64-i386-qemut-rhel6hvm-intel
     ovmf                             test-amd64-i386-qemuu-rhel6hvm-amd
     ovmf                             test-amd64-i386-qemuu-rhel6hvm-intel
    -ovmf                             test-amd64-i386-xl
    -ovmf                             test-amd64-i386-xl-qemut-debianhvm-amd64
    -ovmf                             test-amd64-i386-xl-qemut-debianhvm-amd64-xsm
    -ovmf                             test-amd64-i386-xl-qemut-win7-amd64
    -ovmf                             test-amd64-i386-xl-qemut-winxpsp3
    -ovmf                             test-amd64-i386-xl-qemut-winxpsp3-vcpus1
     ovmf                             test-amd64-i386-xl-qemuu-debianhvm-amd64
     ovmf                             test-amd64-i386-xl-qemuu-debianhvm-amd64-xsm
     ovmf                             test-amd64-i386-xl-qemuu-ovmf-amd64
     ovmf                             test-amd64-i386-xl-qemuu-win7-amd64
     ovmf                             test-amd64-i386-xl-qemuu-winxpsp3
     ovmf                             test-amd64-i386-xl-qemuu-winxpsp3-vcpus1
    -ovmf                             test-amd64-i386-xl-xsm
     qemu-mainline                    build-amd64
     qemu-mainline                    build-amd64-libvirt
     qemu-mainline                    build-amd64-pvops
    @@ -906,48 +877,19 @@
     seabios                          build-i386-libvirt
     seabios                          build-i386-pvops
     seabios                          build-i386-xsm
    -seabios                          test-amd64-amd64-libvirt
    -seabios                          test-amd64-amd64-libvirt-xsm
    -seabios                          test-amd64-amd64-pair
    -seabios                          test-amd64-amd64-xl
    -seabios                          test-amd64-amd64-xl-credit2
    -seabios                          test-amd64-amd64-xl-multivcpu
    -seabios                          test-amd64-amd64-xl-pvh-amd
    -seabios                          test-amd64-amd64-xl-pvh-intel
    -seabios                          test-amd64-amd64-xl-qemut-debianhvm-amd64
    -seabios                          test-amd64-amd64-xl-qemut-debianhvm-amd64-xsm
    -seabios                          test-amd64-amd64-xl-qemut-win7-amd64
    -seabios                          test-amd64-amd64-xl-qemut-winxpsp3
     seabios                          test-amd64-amd64-xl-qemuu-debianhvm-amd64
     seabios                          test-amd64-amd64-xl-qemuu-debianhvm-amd64-xsm
     seabios                          test-amd64-amd64-xl-qemuu-ovmf-amd64
     seabios                          test-amd64-amd64-xl-qemuu-win7-amd64
     seabios                          test-amd64-amd64-xl-qemuu-winxpsp3
    -seabios                          test-amd64-amd64-xl-sedf
    -seabios                          test-amd64-amd64-xl-sedf-pin
    -seabios                          test-amd64-amd64-xl-xsm
    -seabios                          test-amd64-i386-freebsd10-amd64
    -seabios                          test-amd64-i386-freebsd10-i386
    -seabios                          test-amd64-i386-libvirt
    -seabios                          test-amd64-i386-libvirt-xsm
    -seabios                          test-amd64-i386-pair
    -seabios                          test-amd64-i386-qemut-rhel6hvm-amd
    -seabios                          test-amd64-i386-qemut-rhel6hvm-intel
     seabios                          test-amd64-i386-qemuu-rhel6hvm-amd
     seabios                          test-amd64-i386-qemuu-rhel6hvm-intel
    -seabios                          test-amd64-i386-xl
    -seabios                          test-amd64-i386-xl-qemut-debianhvm-amd64
    -seabios                          test-amd64-i386-xl-qemut-debianhvm-amd64-xsm
    -seabios                          test-amd64-i386-xl-qemut-win7-amd64
    -seabios                          test-amd64-i386-xl-qemut-winxpsp3
    -seabios                          test-amd64-i386-xl-qemut-winxpsp3-vcpus1
     seabios                          test-amd64-i386-xl-qemuu-debianhvm-amd64
     seabios                          test-amd64-i386-xl-qemuu-debianhvm-amd64-xsm
     seabios                          test-amd64-i386-xl-qemuu-ovmf-amd64
     seabios                          test-amd64-i386-xl-qemuu-win7-amd64
     seabios                          test-amd64-i386-xl-qemuu-winxpsp3
     seabios                          test-amd64-i386-xl-qemuu-winxpsp3-vcpus1
    -seabios                          test-amd64-i386-xl-xsm
     xen-4.0-testing                  build-amd64
     xen-4.0-testing                  build-amd64-libvirt
     xen-4.0-testing                  build-amd64-pvops

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoDebian: Apply uboot size_hex_prefix workaround to XSM policy too
commit | commitdiff | tree
Ian Campbell [Mon, 4 May 2015 08:18:54 +0000 (09:18 +0100)]
Debian: Apply uboot size_hex_prefix workaround to XSM policy too

Worksaround "Sorry, I could not convert "be700c>"" messages.

The big comment and code which determines $size_hex_prefix is moved
unchanged.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agots-xen-*: Avoid use of uninitialised $r{"enable_xsm"}
commit | commitdiff | tree
Ian Campbell [Thu, 2 Apr 2015 07:33:40 +0000 (08:33 +0100)]
ts-xen-*: Avoid use of uninitialised $r{"enable_xsm"}

Use of uninitialized value $r{"enable_xsm"} in pattern match (m//) at ./ts-xen-install line 49.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoToolstack::libvirt: Support for ACPI fallback for shutdown
commit | commitdiff | tree
Ian Campbell [Thu, 7 May 2015 15:37:18 +0000 (16:37 +0100)]
Toolstack::libvirt: Support for ACPI fallback for shutdown

This is the libvirt counterpart to "Toolstack::xl: Support for ACPI
fallback for shutdown". Currently there are no jobs which test HVM
guests with libvirt and so this is completely untested in the context
of osstest (but at least should be harmless to current jobs).

This relies on an assumption that "virsh shutdown" behaves the same as
"virsh reboot" and accepts a comma separated list of methods to try
given to the --mode argument, which Jim has tested and confirmed to be
true.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Acked by: Jim Fehlig <jfehlig@suse.com>

9 years agoToolstack::xl: Support for ACPI fallback for shutdown
commit | commitdiff | tree
Ian Campbell [Thu, 7 May 2015 09:07:04 +0000 (10:07 +0100)]
Toolstack::xl: Support for ACPI fallback for shutdown

HVM guests which do not include PV drivers will not shutdown after a
simple "xl shutdown". Add a runvar to indicate that the guest will
shutdown in response to an ACPI power event and apply this to the win7
and winxp test jobs.

Tested with:
    test-amd64-amd64-xl-qemuu-winxpsp3
    test-amd64-amd64-xl-qemuu-win7-amd64
    test-amd64-amd64-xl-qemuu-debianhvm-amd64
    test-amd64-amd64-xl

The first two correctly used the extra flag and the domains shutdown,
the other two correctly didn't use the extra flag and still shutdown.

Not currently implemented for libvirt, since we currently have no
libvirt tests I could try it with.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agomake-flight: Handle $BUILD_LVEXTEND_MAX in mfi-common:create_build_jobs()
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:38 +0000 (10:56 +0100)]
make-flight: Handle $BUILD_LVEXTEND_MAX in mfi-common:create_build_jobs()

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agots-debian-hvm-install: Only apply EFI workaround for Wheezy
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:37 +0000 (10:56 +0100)]
ts-debian-hvm-install: Only apply EFI workaround for Wheezy

The previous refactoring of preseed hooks makes this easy to do.

The underlying issue is lack of persistent variable store in our OVMF
setup, which we workaround by placing a copy of grub at the removable
media path. Add a comment saying this since I initially thought this
was just a Wheezy bug.

In Jessie the extra copy of grub can be achieved by preseeding
grub-installer/force-efi-extra-removable (since various real h/w has
similar limitations/bugs) however I haven't tested that so I didn't
add it to the preseed yet, I just mention it in the code comment.

Currently this script hardcodes Wheezy, refactor to use the 'suite'
guest_var (or $c{GuestDebianSuite})

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: wei.liu2@citrix.com
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoDebian: add preseed_create_guest helper
commit | commitdiff | tree
Ian Campbell [Fri, 25 Apr 2014 20:29:24 +0000 (21:29 +0100)]
Debian: add preseed_create_guest helper

Creates a preseed file suitable for use in a PV guest

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoDebian: Refactor installation of overlays, so it can be used for guests too
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:35 +0000 (10:56 +0100)]
Debian: Refactor installation of overlays, so it can be used for guests too

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: longtaox.pang@intel.com
9 years agoDebian: refactor preseeding of .ssh directories
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:34 +0000 (10:56 +0100)]
Debian: refactor preseeding of .ssh directories

Causes known_hosts to be consistently created as well as ~osstest/.ssh
to be consistently populated (it previsouly wasn't for HVM guests).

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: longtaox.pang@intel.com
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoDebian: refactor code to add preseed commands to the preseed file
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:33 +0000 (10:56 +0100)]
Debian: refactor code to add preseed commands to the preseed file

Call it from ts-debian-hvm-install.

This means that, in future, ts-debian-di-install can use
preseed_hook_command and preseed_hook_installscript.

The existing opencoded use of d-i/late_command in the guest preseed
needs to become a preseed_hook_command so as not to clash with the use
of preseed_hook_cmds().

This requires also adding a #! line and the "set -ex" boilerplate
which in turn requires slightly rewriting the /boot/EFI handling part
to also work if the system is not installed for EFI (in which case
grubx64.efi isn't installed). Previously this would have needlessly
created the directory and then ignored the error from cp.

The ssh authorized keys bit isn't touched since it works as is and
will go away in a subsequent patch.

Apart from no longer creating /target/boot/efi/EFI/boot when it is not
needed there is no functional change for now.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: longtaox.pang@intel.com
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agocreate_webfile: Support use with guests as well as hosts.
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:32 +0000 (10:56 +0100)]
create_webfile: Support use with guests as well as hosts.

In particular make the path unique by ensuring it includes the host
and guest name in the guest case.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
9 years agoTestSupport: allow caller of prepareguest_part_xencfg to specify viftype
commit | commitdiff | tree
Ian Campbell [Wed, 6 May 2015 09:56:31 +0000 (10:56 +0100)]
TestSupport: allow caller of prepareguest_part_xencfg to specify viftype

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoMerge remote-tracking branch 'cam/production' flight-55390 flight-55391 flight-55392 flight-55393 flight-55394 flight-55395 flight-55396 flight-55399 flight-55400 flight-55401 flight-55402 flight-55403 flight-55404 flight-55405 flight-55406 flight-55407 flight-55409 flight-55410 flight-55411 flight-55412 flight-55413 flight-55414 flight-55415 flight-55416 flight-55417 flight-55418 flight-55419 flight-55420 flight-55421 flight-55422 flight-55423 flight-55424 flight-55425 flight-55426 flight-55427 flight-55428 flight-55429 flight-55430 flight-55431 flight-55432 flight-55433 flight-55434 flight-55435 flight-55436 flight-55437 flight-55438 flight-55439 flight-55440 flight-55441 flight-55442 flight-55443 flight-55444 flight-55445 flight-55446 flight-55447 flight-55448 flight-55449 flight-55450 flight-55451 flight-55452 flight-55453 flight-55454 flight-55455 flight-55456 flight-55457 flight-55458 flight-55459 flight-55460 flight-55509 flight-55511 flight-55513 flight-55514 flight-55527 flight-55528 flight-55529 flight-55530 flight-55867 flight-55872 flight-55873 flight-55874 flight-55875 flight-55876 flight-55877 flight-55878 flight-55879 flight-55880 flight-55882 flight-55883 flight-55884 flight-56363 flight-56364 flight-56365 flight-56366 flight-56367 flight-56368 flight-56369 flight-56370 flight-56371 flight-56373 flight-56374 flight-56375 flight-56376 flight-56377 flight-56378 flight-56379 flight-56380 flight-56381 flight-56382 flight-56383 flight-56384 flight-56385 flight-56386 flight-56387 flight-56388 flight-56389 flight-56390 flight-56391 flight-56392 flight-56393 flight-56394 flight-56395 flight-56396 flight-56397 flight-56398 flight-56399 flight-56400 flight-56401 flight-56402 flight-56403 flight-56404 flight-56405 flight-56406 flight-56407 flight-56408 flight-56409 flight-56410 flight-56411 flight-56413 flight-56414 flight-56415 flight-56416 flight-56417 flight-56418 flight-56419 flight-56420 flight-56421 flight-56422 flight-56423 flight-56424 flight-56425 flight-56426 flight-56427 flight-56428 flight-56429 flight-56430 flight-56431 flight-56432 flight-56433 flight-56434 flight-56435 flight-56436 flight-56437 flight-56438 flight-56439 flight-56440 flight-56441 flight-56442 flight-56443 flight-56444 flight-56445 flight-56446 flight-56447 flight-56449 flight-56450 flight-56451 flight-56452 flight-56453 flight-56454 flight-56455 flight-56456 flight-56457 flight-56458 flight-56459 flight-56460 flight-56461 flight-56462 flight-56463 flight-56464 flight-56465 flight-56467 flight-56468 flight-56469 flight-56470 flight-56471 flight-56472 flight-56473
commit | commitdiff | tree
Ian Jackson [Tue, 12 May 2015 12:36:27 +0000 (13:36 +0100)]
Merge remote-tracking branch 'cam/production'

10 years agoDocs (in new Docs directory) for cr-disk-report
commit | commitdiff | tree
Ian Jackson [Tue, 12 May 2015 10:41:12 +0000 (11:41 +0100)]
Docs (in new Docs directory) for cr-disk-report

The PDF is automatically generated and linked to, when applicable.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agocr-disk-report: New script
commit | commitdiff | tree
Ian Jackson [Mon, 11 May 2015 15:49:18 +0000 (16:49 +0100)]
cr-disk-report: New script

This generates a report on the disk usage in the logs directory.

10 years agoLogs: Break out logs_select etc. into new Osstest::Management
commit | commitdiff | tree
Ian Jackson [Mon, 11 May 2015 14:24:50 +0000 (15:24 +0100)]
Logs: Break out logs_select etc. into new Osstest::Management

We are going to want to reuse these minor bits of
cr-ensure-disk-space.  Break them out into a new perl module.

We also need to rename some things to make them have names more
suitable for a wider namespace, even if only selectively exported:
 * @logsshopts from @sshopts (it is not the same variable as
   Osstest::TestSupport::sshopts).
 * $loghost and $logdir from $pubhost and $pubdir).
 * onloghost from ontarget.
 * logcfg from dircfg.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agocs-bisection-step: Detect flailing
commit | commitdiff | tree
Ian Jackson [Mon, 11 May 2015 13:30:34 +0000 (14:30 +0100)]
cs-bisection-step: Detect flailing

Specifically, search for previous identical flights (on the same
branch with the relevant blessings).  Here `identical' means it has
exactly the same set of runvars (and, hence, the same set of jobs,
assuming each job has at least one runvar).

This detects various forms of looping which aren't stopped any other
way.

Specifically, one relevant situation occurs if attempts to build
revision (A,B,C) actually build (A,B,C') due to bugs in the build
machinery (which could be bugs in osstest or in the software under
test).  In this case the bisector would never spot its previous
attempts as relevant; instead, it would disregard them due to the
mismatched versions.  It would then retry the same version, until
something happened to stop it.

As written here, we do not consider osstest revision as a relevant
factor in `identical'.  So if reason for the looping is a bug in
osstest we would need to manually un-bless affected flights, as well
as removing the stamp files which are used to record completion of
attempted bisection.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME.planner: Document the resource planning system
commit | commitdiff | tree
Ian Jackson [Thu, 20 Nov 2014 18:07:16 +0000 (18:07 +0000)]
README.planner: Document the resource planning system

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoproduction-config: Correct syntax for clearing OsstestUpstream. flight-55275 flight-55276 flight-55277 flight-55278 flight-55279 flight-55280 flight-55291 flight-55301 flight-55302 flight-55303 flight-55304 flight-55305 flight-55306 flight-55307 flight-55308 flight-55309 flight-55310 flight-55311 flight-55312 flight-55313 flight-55314 flight-55315 flight-55316 flight-55317 flight-55318 flight-55319 flight-55320 flight-55321 flight-55322 flight-55323 flight-55324 flight-55325 flight-55326 flight-55327 flight-55328 flight-55329 flight-55330 flight-55331 flight-55332 flight-55333 flight-55334 flight-55335 flight-55336 flight-55337 flight-55338 flight-55339 flight-55340 flight-55341 flight-55342 flight-55343 flight-55344 flight-55345 flight-55346 flight-55347 flight-55348 flight-55349 flight-55350 flight-55351 flight-55352 flight-55353 flight-55354 flight-55355 flight-55356 flight-55357 flight-55358 flight-55359 flight-55360 flight-55361 flight-55362 flight-55363 flight-55364 flight-55365 flight-55366 flight-55367 flight-55368 flight-55369 flight-55370 flight-55371 flight-55372 flight-55373 flight-55374 flight-55375 flight-55376 flight-55377 flight-55378 flight-55379 flight-55380 flight-55381 flight-55382 flight-55383 flight-55384 flight-55385 flight-55386 flight-55387 flight-55388 flight-55389
commit | commitdiff | tree
Ian Campbell [Mon, 11 May 2015 10:37:33 +0000 (11:37 +0100)]
production-config: Correct syntax for clearing OsstestUpstream.

Previous syntax actually sets it to the literal string containing two
quotation marks.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoallow.all: Ignore FreeBSD local migration failures
commit | commitdiff | tree
Ian Campbell [Mon, 11 May 2015 10:35:56 +0000 (11:35 +0100)]
allow.all: Ignore FreeBSD local migration failures

Until the issues with them are resolved

Rerunning sg-report-flight for 54309 results in:

@@ -5,11 +9,11 @@

 Tests which did not succeed and are blocking,
 including tests which could not be run:
- test-amd64-i386-freebsd10-amd64 13 guest-localmigrate     fail REGR. vs. 50405
  build-amd64-xsm               5 xen-build                 fail REGR. vs. 50405

 Regressions which are regarded as allowable (not blocking):
  test-armhf-armhf-xl-sedf-pin  3 host-install(3)         broken REGR. vs. 50405
+ test-amd64-i386-freebsd10-amd64 13 guest-localmigrate     fail REGR. vs. 50405
  test-amd64-i386-freebsd10-i386 13 guest-localmigrate           fail like 50405
  test-armhf-armhf-libvirt     11 guest-start                  fail   like 50405

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoAutomerge of git://xenbits.xen.org/osstest.git master into production
commit | commitdiff | tree
xen.org [Sun, 10 May 2015 16:35:00 +0000 (17:35 +0100)]
Automerge of git://xenbits.xen.org/osstest.git master into production

10 years agocambridge: Reduce reliance on woking and configdb
commit | commitdiff | tree
Ian Campbell [Thu, 7 May 2015 15:41:21 +0000 (16:41 +0100)]
cambridge: Reduce reliance on woking and configdb

The hosts which were previsouly attached to woking's serial ports are
now attached to osstser1 which is our own box.

All hosts which previously used configdb (and by extension statedb)
are now configured directly via the host properties entries for PDU
control (direct to PDU) and for Ethernet address, so disable configdb.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoMerge branch 'from-cambridge/2015-05-05' of xenbits.xen.org:/home/xen/git/people... flight-53859 flight-54286 flight-55171 flight-55172 flight-55173 flight-55174 flight-55225 flight-55226 flight-55227 flight-55228 flight-55229 flight-55230 flight-55239 flight-55240 flight-55241 flight-55242 flight-55243 flight-55244 flight-55245 flight-55246 flight-55247 flight-55248 flight-55249 flight-55250 flight-55251 flight-55252 flight-55253 flight-55254 flight-55255 flight-55256 flight-55257 flight-55258 flight-55259 flight-55260 flight-55261 flight-55262 flight-55263 flight-55264 flight-55265 flight-55266 flight-55267 flight-55268 flight-55269 flight-55270 flight-55271 flight-55272 flight-55273 flight-55274
commit | commitdiff | tree
Ian Jackson [Thu, 7 May 2015 14:06:59 +0000 (15:06 +0100)]
Merge branch 'from-cambridge/2015-05-05' of xenbits.xen.org:/home/xen/git/people/ianc/osstest into fast

10 years agocs-bisection-step: Abandon repro attempts after a bit flight-53720 flight-53721 flight-53722 flight-53723 flight-53725 flight-53726 flight-53727 flight-53728 flight-53729 flight-53730 flight-53731 flight-53732 flight-53733 flight-53734 flight-53735 flight-53736 flight-53737 flight-53739 flight-53740 flight-53741 flight-53742 flight-53743 flight-53744 flight-53745 flight-53746 flight-53747 flight-53748 flight-53749 flight-53750 flight-53751 flight-53752 flight-53753 flight-53754 flight-53756 flight-53757 flight-53758 flight-53759 flight-53760 flight-53761 flight-53762 flight-53763 flight-53764 flight-53765 flight-53766 flight-53767 flight-53768 flight-53769 flight-53770 flight-53771 flight-53772 flight-53773 flight-53774 flight-53775 flight-53777 flight-53778 flight-53779 flight-53780 flight-53781 flight-53782 flight-53783 flight-53785 flight-53786 flight-53787 flight-53789 flight-53790 flight-53791 flight-53792 flight-53793 flight-53794 flight-53795 flight-53796 flight-53797 flight-53798 flight-53799 flight-53800 flight-53803 flight-53804 flight-53805 flight-53806 flight-53807 flight-53809 flight-53810 flight-53811 flight-53812 flight-53813 flight-53814 flight-53815 flight-53816 flight-53817 flight-53818 flight-53819 flight-53820 flight-53821 flight-53822 flight-53823 flight-53824 flight-53825 flight-53826 flight-53827 flight-53828 flight-53829 flight-53830 flight-53831 flight-53832 flight-53833 flight-53834 flight-53835 flight-53836 flight-53837 flight-53838 flight-53839 flight-53840 flight-53841 flight-53842 flight-53843 flight-53844 flight-53845 flight-53846 flight-53847 flight-53848 flight-53849 flight-53850 flight-53851 flight-53852 flight-53853 flight-53854 flight-53855 flight-53856 flight-53857 flight-53858 flight-53860 flight-53861 flight-53862 flight-53863 flight-53864 flight-53865 flight-53866 flight-53867 flight-53868 flight-53869 flight-53870 flight-53871 flight-53872 flight-53873 flight-53874 flight-53875 flight-53876 flight-53877 flight-53878 flight-53879 flight-53880 flight-53881 flight-53882 flight-53883 flight-53884 flight-53885 flight-53886 flight-53887 flight-53888 flight-53889 flight-53890 flight-53891 flight-53892 flight-53893 flight-53894 flight-53895 flight-53896 flight-53897 flight-53898 flight-53899 flight-53900 flight-53901 flight-53902 flight-53903 flight-53904 flight-53905 flight-53906 flight-53907 flight-53908 flight-53909 flight-53910 flight-53911 flight-53912 flight-53913 flight-53914 flight-53915 flight-53916 flight-53917 flight-53918 flight-53919 flight-53920 flight-53921 flight-53922 flight-53923 flight-53924 flight-53925 flight-53926 flight-53927 flight-53928 flight-53929 flight-53930 flight-53931 flight-53933 flight-53934 flight-53935 flight-53936 flight-53937 flight-53938 flight-53939 flight-53940 flight-53941 flight-53942 flight-53943 flight-53944 flight-53945 flight-53946 flight-53947 flight-53948 flight-53949 flight-53950 flight-53951 flight-53952 flight-53953 flight-53954 flight-53955 flight-53956 flight-53957 flight-53958 flight-53959 flight-53960 flight-53961 flight-53963 flight-53964 flight-54085 flight-54092 flight-54093 flight-54094 flight-54095 flight-54279 flight-54283 flight-54284 flight-54285 flight-54287 flight-54288 flight-54289 flight-54290 flight-54291 flight-54292 flight-54293 flight-54294 flight-54295 flight-54296 flight-54297 flight-54298 flight-54299 flight-54300 flight-54301 flight-54302 flight-54303 flight-54304 flight-54305 flight-54306 flight-54307 flight-54308 flight-54309 flight-54310 flight-54311 flight-54312 flight-54313 flight-54692 flight-54693 flight-54694 flight-54695 flight-54696 flight-54819 flight-54829 flight-54831 flight-54832 flight-54835 flight-54836 flight-54837 flight-54838 flight-54839 flight-54840 flight-55168 flight-55169 flight-55170
commit | commitdiff | tree
Ian Jackson [Wed, 6 May 2015 22:47:30 +0000 (23:47 +0100)]
cs-bisection-step: Abandon repro attempts after a bit

If we have had a number of attempts at a repro, and none of them have
produced a pass or fail, something is probably wrong and we should
give up rather than carrying on.

Handle this with the machinery we use for conflicting test results.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agocs-bisection-step: Report conflict if basis pass/fail are wrong
commit | commitdiff | tree
Ian Jackson [Wed, 6 May 2015 22:41:00 +0000 (23:41 +0100)]
cs-bisection-step: Report conflict if basis pass/fail are wrong

It can happen that the (for example) supposed basis pass (originally
only tested on another host) failed (when reproduced, or for some
other reason).  When that happens do not attempt to get it to pass;
instead, treat it the same way we would if we had actually got
conflicting results at that revision.

(Conversely, do not attempt to get a basis fail if the basis fail has
already passed on the selected host.  This is, as it happens,
impossible in a bisection triggered by sg-report-flight with the
current invocation arrangements - but cs-bisection-step should
handle it correctly.)

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agocs-bisection-step: Break out report_conflict
commit | commitdiff | tree
Ian Jackson [Wed, 6 May 2015 22:39:18 +0000 (23:39 +0100)]
cs-bisection-step: Break out report_conflict

We are going to want to reuse this.

Also, explicitly return 0 (for false) from conflicted_warning when it
doesn't apply.

No resulting functional change.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agomake-flight: Drop qemut tests in qemuu branches
commit | commitdiff | tree
Ian Jackson [Tue, 5 May 2015 18:31:10 +0000 (19:31 +0100)]
make-flight: Drop qemut tests in qemuu branches

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agomake-flight: Drop unspecified-qemu tests
commit | commitdiff | tree
Ian Jackson [Tue, 5 May 2015 18:25:59 +0000 (19:25 +0100)]
make-flight: Drop unspecified-qemu tests

The default since 4.3 onwards is qemuu only, and we do not expect
it to change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agostandalone: noreinstall -> reinstall in help string
commit | commitdiff | tree
Wei Liu [Fri, 24 Apr 2015 10:08:25 +0000 (11:08 +0100)]
standalone: noreinstall -> reinstall in help string

Signed-off-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agosg-report-flight: Report stepno and testid of first worst fail
commit | commitdiff | tree
Ian Jackson [Fri, 1 May 2015 14:54:03 +0000 (15:54 +0100)]
sg-report-flight: Report stepno and testid of first worst fail

This makes reading the scoreboard considerably easier.

We abuse the local variable @worst slightly, pushing the extra info we
are going to print onto the end of it.

We also have to defer printing the cells, because we compute the cell
to duplicate in column order but we have to output them in row order.
For symmetry we accumulate both rows rather than only the second row.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-flight: Include actual status in column-head cells
commit | commitdiff | tree
Ian Jackson [Fri, 1 May 2015 14:47:44 +0000 (15:47 +0100)]
sg-report-flight: Include actual status in column-head cells

Adjust the $cell_html anonymous sub to include the actual step status
unless it is told not to (by passing it a new $core_only argument).

Change the main body call site to pass this option.  As a result the
column headings now include the step status.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-flight: Better choice of column-head cell
commit | commitdiff | tree
Ian Jackson [Fri, 1 May 2015 14:46:36 +0000 (15:46 +0100)]
sg-report-flight: Better choice of column-head cell

We are more interested in the first worst failure.  But if everything
passed, then the last pass.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-flight: Break colour of out $cell_html->() results
commit | commitdiff | tree
Ian Jackson [Fri, 1 May 2015 13:40:07 +0000 (14:40 +0100)]
sg-report-flight: Break colour of out $cell_html->() results

No functional change.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agosg-report-job-history: Avoid full runvars table scan (!)
commit | commitdiff | tree
Ian Jackson [Thu, 30 Apr 2015 15:23:56 +0000 (16:23 +0100)]
sg-report-job-history: Avoid full runvars table scan (!)

sg-report-job-history wants to know the potential names of runvars
relating to hosts.  To do this it tries to find a list of distinct
runvar names which exist in the flights it's processing.

However, it fails to limit the runvar query appropriately, and as a
result postgresql must scan almost the complete runvars table to
produce an answer.  This is very slow if the table is bigger than the
database server's RAM.

Fix this by limiting the runvars table query to relevant flights.

Specifically:

 * Break the `100' from the LIMIT clause on the flights search
   into a local variable $limit.
 * Break the bulk of the flights search sql statement text into
   a local variable $fromstuff.
 * In the runvars statement, add a condition on flights which uses
   LIMIT and OFFSET, based on results of the the flights query.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agocs-bisection-step: Use pbm tools, not graphicsmagick/imagemagick
commit | commitdiff | tree
Ian Jackson [Thu, 30 Apr 2015 15:20:51 +0000 (16:20 +0100)]
cs-bisection-step: Use pbm tools, not graphicsmagick/imagemagick

Graphicsmagick / imagemagick have very poor performance with images
with large pixel sizes.  The bisector can generate some very large
images.

In an example I have seen, a 21595x21048 png, occupying only 2.6Mby of
disk space.  An invocation of `convert' to resize this was using 3Gby
of RAM and lots of CPU.  Whereas, the pbm utilities can process this
with much less memory and a tiny fraction of the cpu time.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoAutomerge of git://xenbits.xen.org/osstest.git master into production
commit | commitdiff | tree
xen.org [Mon, 4 May 2015 10:20:14 +0000 (11:20 +0100)]
Automerge of git://xenbits.xen.org/osstest.git master into production

10 years agots-kernel-build: Enable CONFIG_SCSI_SAS_ATA flight-53073 flight-53718 flight-53719
commit | commitdiff | tree
Ian Jackson [Wed, 29 Apr 2015 15:06:29 +0000 (16:06 +0100)]
ts-kernel-build: Enable CONFIG_SCSI_SAS_ATA

(Some) SAS storage controller drivers do not recognise attached SATA
disks when this option is not set.  It is inexplicably not set by
default in Linux 3.14.36 (at least).

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agotarget_cmd_build: Delete build-ok-stamp before starting
commit | commitdiff | tree
Ian Jackson [Tue, 21 Apr 2015 16:39:24 +0000 (17:39 +0100)]
target_cmd_build: Delete build-ok-stamp before starting

Many of the callers of target_cmd_build use a build-ok-stamp idiom to
detect failed builds.  This idiom does not work if the stamp file
exists already, so delete it.

In the future we may move more of the test build-ok-stamp, echo ok,
into TestSupport, but this will do for now.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agots-kernel-build: Enable x86 IOMMU options
commit | commitdiff | tree
Ian Jackson [Tue, 21 Apr 2015 15:28:37 +0000 (16:28 +0100)]
ts-kernel-build: Enable x86 IOMMU options

This has a variety of beneficial implications:

 * The kernel becomes more like the kind of distro kernels that Xen
   users are probably using.

 * We are more likely to discover any bugs in Linux where Linux
   running under Xen (eg as dom0) fights with Xen for control of io
   mediation resources or otherwise mishandles the situation.

 * A pleasant side effect is that in a kernel which does not yet have
   "config: Enable NEED_DMA_MAP_STATE when SWIOTLB is selected"
   (a bugfix), enabling INTEL_IOMMU has the side effect of enabling
   NEED_DMA_MAP_STATE and thus working around the bug.

The list of options to enable was derived by eyeballing
drivers/iommu/Kconfig from 3.14.34.

I will leave the question of whether to enable any ARM IOMMU options
for the Xen ARM folks to consider.

Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
CC: Boris Ostrovsky <boris.ostrovsky@oracle.com>
CC: David Vrabel <david.vrabel@citrix.com>
CC: Andrew Cooper <andrew.cooper3@citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
10 years agots-libvirt-build: set coredump ulimit to unlimited for libvirtd
commit | commitdiff | tree
Ian Campbell [Wed, 25 Mar 2015 11:30:36 +0000 (11:30 +0000)]
ts-libvirt-build: set coredump ulimit to unlimited for libvirtd

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-host-install: Arrange for ssh logins to have no corefile size limit
commit | commitdiff | tree
Ian Campbell [Wed, 25 Mar 2015 14:29:50 +0000 (14:29 +0000)]
ts-host-install: Arrange for ssh logins to have no corefile size limit

Collect the output of "cat /proc/self/limits" so we get some clue if
this isn't working for some reason.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoArrange for core dumps to be placed in /var/core and collect them
commit | commitdiff | tree
Ian Campbell [Tue, 24 Mar 2015 14:23:58 +0000 (14:23 +0000)]
Arrange for core dumps to be placed in /var/core and collect them

Refactor the $kvp_replace helper in ts-xen-install into a generic
helper (which requires using ::EO and ::EI for namespacing) for use
with target_editfile and use it to edit /etc/sysctl.conf to set
kernel.core_pattern on boot.

Tested in standalone mode by installing and running a C program
containing "*(int *)0 = 1;" which, after running "ulimit -c unlimited"
produces the expected core file. ts-logs-capture when run in
standalone mode then picks them up.

I've not yet figured out how to make the desired rlimit take affect
for all processes (including e.g. daemons spawned on boot). Likely
this will involve some combination of pam_limits.so PAM module and
adding explicit ulimit calls to the initscripts which we care about
(primarily xencommons and libvirt initscripts).

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agocambridge: Stop publishing logs to chiark
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 13:24:00 +0000 (14:24 +0100)]
cambridge: Stop publishing logs to chiark

http://osstest.cam.xci-test.com/~osstest/testlogs already exists and
points to the live logs directory, so switch PubBaseUrl to that in the
Cambridge config such that email reports etc contain it. This won't be
externally accessible but I think that won't matter now that the
master production instance is elsewhere.

Arrange that cr-publish-flight-logs doesn't publish the corresponding
thing if either LogsPublish or ResultsPublish is not set, and unset
them in the Cambridge config.

Likewise arrange that cr-ensure-disk-space doesn't do anything if the
configuration variable passed as an option is not set, and unset
Publish (the base for {Logs,Results}Publish) in the Cambridge config.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v2: Check the config variable and not its name.
v3: Adjust for control VM move to xs.citrite.net

10 years agocambridge: Do not try to push harness to XenProject instance output
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 13:12:51 +0000 (14:12 +0100)]
cambridge: Do not try to push harness to XenProject instance output

By arranging for cr-publish-flight-logs to ignore --push-harness if
either of HarnessPublishGitRepoDir or HarnessPublishGitUserHost are
not specified

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
---
v2:
 - Avoid logm which isn't available here, wasn't saying much of use
   anyway.
 - Syntax fix (is not a function, so exit not return)

Perhaps we should have our own tree for such things, but for now just
nobble it.

10 years agocambridge: Add a crontab
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 13:08:59 +0000 (14:08 +0100)]
cambridge: Add a crontab

Right now all it does is keep osstest itself up to date.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoHandle osstest's own local push gate in non-master production instances
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 12:50:50 +0000 (13:50 +0100)]
Handle osstest's own local push gate in non-master production instances

We want to arrange that the master XenProject instance continues to
test its own pretest branch while any downstream instances will pickup
changes from the master instance's production (i.e. tested) branch,
which is published at git://xenbits.xen.org/osstest.git#master. We
want to also be able to use local pretest for local changes (which may
or may not get merged back upstream).

Add a new configuration option OsstestUpstream which by default is
"git://xenbits.xen.org/osstest.git master" and which is cleared to
nothing on the master instance via production-config.

If the option is not set then the existing behaviour is unchanged.

If the option is set then osstest branch flights will still prefer to
test the local pretest branch, but if nothing is pending there then it
will proceed by merging the upstream branch into the local production
branch and testing the result.

This merge must be done:
 - in a clone not in the main testing.git in order to avoid inserting
   merge conflict markers into the active set of scripts.
 - in a non-bare repo because git merge requires it.

$repos/osstest is a bare repo which we want to keep that way because
using repo_tree_rev_fetch_git to fetch the remote branch is
convenient.

So we use $repos/osstest-merge as a temporary merge repo and reclone
from the active local repo each time.

All of this happens in ap-fetch-version.

As part of this arrange that the result is always left in the ap-fetch
branch of the for-osstest.git repo (even for existing cases) and the
sha1 is produced as output. Resetting to that revision is handled by
cr-daily-branch.

If the merge fails then manual intervention (i.e. a manual merge and
push to the _local_ pretest) will be required. Likewise if local
pretest and local production have diverged manual intervention will be
required.

In ap-push we stop pushing to xenbits#master except for the master
instance if an upstream is defined. At some point it might be useful
to add a configuration option for where to push to but I don't have
that requirement right now.

ap-fetch-version-old requires no changes.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
---
v4:
 - Use git update-ref properly, i.e. with the full ref name, otherwise
   it creates random .git/ap-fetch refs

v3:
 - Only merge from upstream if there is nothing pending locally.
 - Always update ap-fetch.

v2:
 - Arrange for $OSSTEST_USE_HEAD=y to take precendence
 - drop LOCALREV (which was wrong anyway) in favour of inline
   branchname
 - Rename OSSTEST_REVISION_MERGE as revision_merge to avoid implying
   it can be set and will be honoured.
 - Git in Debian Squeeze lacks -C and --no-edit, adjust accordingly.

full name to update-ref

10 years agoOsstest/Debian.pm: Use Fqdn hostprop when collecting host keys
commit | commitdiff | tree
Ian Campbell [Fri, 1 May 2015 10:20:45 +0000 (11:20 +0100)]
Osstest/Debian.pm: Use Fqdn hostprop when collecting host keys

Otherwise hosts which are not in the same DnsDomain are not processed,
resulting in log messages such as:

2015-05-01 10:06:19 Z skipping host key for nonexistent host marilith-n4.xs.citrite.net
2015-05-01 10:06:20 Z skipping host key for nonexistent host lace-bug.xs.citrite.net

The practical impact of this appears to be that the pair migration
tests can fail with:

2015-05-01 13:18:03 Z executing ssh ... root@10.80.250.26 xl migrate debian.guest.osstest moss-bug
Host key verification failed.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agocambridge: Switch configuration to use osstest.xs.citrite.net
commit | commitdiff | tree
Ian Campbell [Wed, 29 Apr 2015 10:14:58 +0000 (11:14 +0100)]
cambridge: Switch configuration to use osstest.xs.citrite.net

The VM has moved to different infrastructure and its new name is
osstest.citrite.net.

Update ExecutiveDbnamePat. The DB is still in the XC infrastructure so
using DnsDomain (the default) no longer works.

Set {Owner,Queue}DaemonHost to refer to the new VM host and not the
default ControlDaemonHost value of control-daemons.osstest.cam.xci-test.com
(which will be removed later).

We set both variables rather than just ControlDaemonHost in case we
ever want to move one but not the other.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoallow instance specific settings
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 10:54:47 +0000 (11:54 +0100)]
allow instance specific settings

cri-args-hostlists and invoke-daemon now check for
$HOME/.xen-osstest/settings which can contain things like "export
OSSTEST_CONFIG=production-config-cambridge" to tailor things for a
particular instance of osstest running in production mode.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
[ ijc -- add invoke-daemon too and reword commit message accordingly ]

10 years agoFreeBSD: Cleanups relating to guest images and ts-freebsd-install script flight-50508 flight-50509 flight-50511 flight-50512 flight-50513 flight-50514 flight-50515 flight-50516 flight-50517 flight-50518 flight-50519 flight-50520 flight-50521 flight-50522 flight-50523 flight-50524 flight-50525 flight-50526 flight-50527 flight-50528 flight-50529 flight-50530 flight-50531 flight-50532 flight-50533 flight-50534 flight-50535 flight-50536 flight-50537 flight-50538 flight-50539 flight-50540 flight-50541 flight-50542 flight-50543 flight-50544 flight-50545 flight-50546 flight-50547 flight-50548 flight-50549 flight-50550 flight-50551 flight-50552 flight-50553 flight-50554 flight-50555 flight-50556 flight-50557 flight-50558 flight-50559 flight-50560 flight-50561 flight-50562 flight-50563 flight-50564 flight-50565 flight-50566 flight-50567 flight-50568 flight-50569 flight-50570 flight-50571 flight-50572 flight-50573 flight-50574 flight-50575 flight-50576 flight-50577 flight-50578 flight-50579 flight-50580 flight-50581 flight-50582 flight-50583 flight-50584 flight-50585 flight-50586 flight-50587 flight-50588 flight-50589 flight-50590 flight-50591 flight-50592 flight-50593 flight-50594 flight-50595 flight-50596 flight-50597 flight-50598 flight-50599 flight-50600 flight-50601 flight-50602 flight-50603 flight-50604 flight-50605 flight-50606 flight-50607 flight-50608 flight-50609 flight-50610 flight-50611 flight-50612 flight-50613 flight-50614 flight-50615 flight-50616 flight-50617 flight-50618 flight-50619 flight-50620 flight-50621 flight-50622 flight-50623 flight-50624 flight-50625 flight-50626 flight-50627 flight-50628 flight-50629 flight-50630 flight-50631 flight-50632 flight-50633 flight-50634 flight-50635 flight-50636 flight-50637 flight-50638 flight-50639 flight-50640 flight-50641 flight-50642 flight-50643 flight-50644 flight-50645 flight-50646 flight-50647 flight-50648 flight-50649 flight-50650 flight-50651 flight-50652 flight-50653 flight-50654 flight-50655 flight-50656 flight-50657 flight-50658 flight-50659 flight-50660 flight-50661 flight-50662 flight-50663 flight-50664 flight-50665 flight-50666 flight-50667 flight-50668 flight-50669 flight-50670 flight-50671 flight-50672 flight-50673 flight-50674 flight-50675 flight-50676 flight-50677 flight-50678 flight-50679 flight-50680 flight-50681 flight-50682 flight-50683 flight-50684 flight-50685 flight-50686 flight-50687 flight-50688 flight-50689 flight-50690 flight-50691 flight-50693 flight-50694 flight-50695 flight-50696 flight-50697 flight-50698 flight-50699 flight-50700 flight-50701 flight-50702 flight-50703 flight-50704 flight-50705 flight-50706 flight-50707 flight-50708 flight-50709 flight-50710 flight-50711 flight-50712 flight-50713 flight-50714 flight-50715 flight-50716 flight-50717 flight-50718 flight-50719 flight-50720 flight-50721 flight-50722 flight-50723 flight-50724 flight-50725 flight-50726 flight-50728 flight-50729 flight-50731 flight-50732 flight-50734 flight-50735 flight-50736 flight-50737 flight-50738 flight-50739 flight-50740 flight-50741 flight-50742 flight-50743 flight-50744 flight-50749 flight-50750 flight-50751 flight-50752 flight-50753 flight-50758 flight-50761 flight-50765 flight-50766 flight-50767 flight-50770 flight-50771 flight-51910 flight-51911 flight-51912 flight-51913 flight-51914 flight-51915 flight-51916 flight-51917 flight-51929 flight-51931 flight-51932 flight-51933 flight-51934 flight-51935 flight-51936 flight-51937 flight-51938 flight-51939 flight-51940 flight-51941 flight-51942 flight-51943 flight-51944 flight-51945 flight-51946 flight-51947 flight-51948 flight-51949 flight-51950 flight-51951 flight-51952 flight-51953 flight-51954 flight-51955 flight-51956 flight-51957 flight-51958 flight-51959 flight-51960 flight-51961 flight-51962 flight-51963 flight-51964 flight-51965 flight-51966 flight-51967 flight-51968 flight-51969 flight-51970 flight-51971 flight-51972 flight-51973 flight-51974 flight-51979 flight-51980 flight-51981 flight-51982 flight-51983 flight-51984 flight-51985 flight-51986 flight-51987 flight-51988 flight-51989 flight-51990 flight-51991 flight-51992 flight-51993 flight-51994 flight-51995 flight-51996 flight-51997 flight-51998 flight-51999 flight-52000 flight-52001 flight-52002 flight-52003 flight-52004 flight-52005 flight-52006 flight-52007 flight-52008 flight-52009 flight-52010 flight-52011 flight-52012 flight-52013 flight-52014 flight-52015 flight-52016 flight-52017 flight-52018 flight-52019 flight-52020 flight-52021 flight-52022 flight-52023 flight-52024 flight-52025 flight-52026 flight-52027 flight-52028 flight-52029 flight-52030 flight-52031 flight-52032 flight-52033 flight-52034 flight-52035 flight-52036 flight-52037 flight-52038 flight-52039 flight-52040 flight-52041 flight-52042 flight-52043 flight-52044 flight-52045 flight-52046 flight-52047 flight-52048 flight-52049 flight-52050 flight-52052 flight-52054 flight-52055 flight-52056 flight-52057 flight-52058 flight-52059 flight-52060 flight-52061 flight-52062 flight-52063 flight-52064 flight-52065 flight-52066 flight-52067 flight-52068 flight-52069 flight-52070 flight-52071 flight-52072 flight-52073 flight-52074 flight-52075 flight-52076 flight-52077 flight-52078 flight-52079 flight-52080 flight-52082 flight-52083 flight-52084 flight-52085 flight-52086 flight-52087 flight-52088 flight-52089 flight-52090 flight-52091 flight-52092 flight-52093 flight-52094 flight-52095 flight-52096 flight-52097 flight-52098 flight-52099 flight-52100 flight-52101 flight-52102 flight-52103 flight-52104 flight-52107 flight-52108 flight-52109 flight-52110 flight-52111 flight-52112 flight-52113 flight-52114 flight-52115 flight-52116 flight-52117 flight-52118 flight-52119 flight-52120 flight-52121 flight-52122 flight-52123 flight-52124 flight-52125 flight-52126 flight-52127 flight-52128 flight-52129 flight-52130 flight-52131 flight-52132 flight-52133 flight-52134 flight-52135 flight-52136 flight-52137 flight-52138 flight-52139 flight-52140 flight-52141 flight-52142 flight-52143 flight-52144 flight-52145 flight-52146 flight-52147 flight-52148 flight-52149 flight-52150 flight-52151 flight-52152 flight-52153 flight-52154 flight-52155 flight-52156 flight-52157 flight-52158 flight-52159 flight-52160 flight-52161 flight-52162 flight-52163 flight-52164 flight-52165 flight-52166 flight-52167 flight-52168 flight-52169 flight-52170 flight-52171 flight-52172 flight-52173 flight-52174 flight-52175 flight-52176 flight-52177 flight-52178 flight-52179 flight-52180 flight-52181 flight-52182 flight-52183 flight-52184 flight-52185 flight-52186 flight-52187 flight-52188 flight-52189 flight-52190 flight-52191 flight-52192 flight-52193 flight-52194 flight-52195 flight-52196 flight-52197 flight-52198 flight-52199 flight-52200 flight-52201 flight-52202 flight-52203 flight-52204 flight-52205 flight-52206 flight-52207 flight-52208 flight-52209 flight-52210 flight-52211 flight-52212 flight-52213 flight-52214 flight-52215 flight-52216 flight-52217 flight-52218 flight-52219 flight-52220 flight-52221 flight-52222 flight-52223 flight-52224 flight-52225 flight-52226 flight-52227 flight-52228 flight-52229 flight-52230 flight-52231 flight-52232 flight-52233 flight-52234 flight-52235 flight-52236 flight-52237 flight-52238 flight-52239 flight-52240 flight-52241 flight-52242 flight-52243 flight-52244 flight-52245 flight-52246 flight-52247 flight-52248 flight-52249 flight-52250 flight-52251 flight-52252 flight-52253 flight-52254 flight-52255 flight-52256 flight-52257 flight-52258 flight-52259 flight-52260 flight-52261 flight-52262 flight-52263 flight-52264 flight-52265 flight-52266 flight-52267 flight-52268 flight-52269 flight-52270 flight-52271 flight-52272 flight-52273 flight-52274 flight-52275 flight-52276 flight-52277 flight-52278 flight-52279 flight-52280 flight-52281 flight-52282 flight-52283 flight-52284 flight-52285 flight-52286 flight-52287 flight-52288 flight-52289 flight-52290 flight-52291 flight-52292 flight-52293 flight-52294 flight-52295 flight-52296 flight-52297 flight-52299 flight-52300 flight-52301 flight-52302 flight-52303 flight-52304 flight-52305 flight-52306 flight-52307 flight-52308 flight-52309 flight-52310 flight-52311 flight-52312 flight-52313 flight-52314 flight-52315 flight-52316 flight-52317 flight-52318 flight-52319 flight-52320 flight-52321 flight-52322 flight-52323 flight-52324 flight-52325 flight-52326 flight-52327 flight-52328 flight-52329 flight-52330 flight-52331 flight-52332 flight-52333 flight-52334 flight-52335 flight-52336 flight-52337 flight-52338 flight-52339 flight-52340 flight-52341 flight-52342 flight-52343 flight-52344 flight-52345 flight-52346 flight-52347 flight-52348 flight-52349 flight-52350 flight-52351 flight-52352 flight-52353 flight-52354 flight-52355 flight-52356 flight-52357 flight-52358 flight-52359 flight-52360 flight-52361 flight-52362 flight-52363 flight-52364 flight-52365 flight-52366 flight-52367 flight-52368 flight-52369 flight-52370 flight-52371 flight-52372 flight-52373 flight-52374 flight-52375 flight-52376 flight-52377 flight-52378 flight-52379 flight-52380 flight-52382 flight-52383 flight-52384 flight-52385 flight-52386 flight-52387 flight-52388 flight-52389 flight-52401 flight-52402 flight-52403 flight-52404 flight-52405 flight-52406 flight-52407 flight-52408 flight-52409 flight-52410 flight-52411 flight-52412 flight-52413 flight-52414 flight-52415 flight-52416 flight-52417 flight-52418 flight-52419 flight-52420 flight-52421 flight-52422 flight-52423 flight-52424 flight-52425 flight-52426 flight-52427 flight-52428 flight-52429 flight-52430 flight-52431 flight-52432 flight-52433 flight-52434 flight-52435 flight-52436 flight-52437 flight-52438 flight-52439 flight-52440 flight-52441 flight-52442 flight-52443 flight-52444 flight-52445 flight-52446 flight-52447 flight-52448 flight-52449 flight-52450 flight-52451 flight-52452 flight-52453 flight-52454 flight-52455 flight-52456 flight-52457 flight-52458 flight-52459 flight-52460 flight-52461 flight-52462 flight-52463 flight-52464 flight-52465 flight-52466 flight-52467 flight-52468 flight-52469 flight-52470 flight-52471 flight-52472 flight-52473 flight-52474 flight-52475 flight-52476 flight-52477 flight-52478 flight-52479 flight-52480 flight-52481 flight-52482 flight-52483 flight-52484 flight-52485 flight-52486 flight-52487 flight-52488 flight-52489 flight-52490 flight-52491 flight-52492 flight-52494 flight-52496 flight-52497 flight-52498 flight-52499 flight-52500 flight-52501 flight-52502 flight-52503 flight-52504 flight-52505 flight-52506 flight-52507 flight-52508 flight-52509 flight-52510 flight-52511 flight-52512 flight-52513 flight-52514 flight-52515 flight-52516 flight-52517 flight-52518 flight-52519 flight-52520 flight-52521 flight-52522 flight-52523 flight-52524 flight-52525 flight-52526 flight-52528 flight-52529 flight-52530 flight-52531 flight-52532 flight-52533 flight-52535 flight-52536 flight-52537 flight-52538 flight-52539 flight-52540 flight-52541 flight-52542 flight-52543 flight-52544 flight-52545 flight-52546 flight-52547 flight-52548 flight-52549 flight-52550 flight-52551 flight-52552 flight-52553 flight-52554 flight-52556 flight-52558 flight-52559 flight-52560 flight-52561 flight-52562 flight-52563 flight-52564 flight-52565 flight-52610 flight-52611 flight-52612 flight-52613 flight-52615 flight-52616 flight-52618 flight-52620 flight-52621 flight-52622 flight-52623 flight-52624 flight-52625 flight-52626 flight-52627 flight-52628 flight-52629 flight-52630 flight-52631 flight-52632 flight-52633 flight-52634 flight-52635 flight-52636 flight-52637 flight-52638 flight-52639 flight-52640 flight-52641 flight-52642 flight-52643 flight-52644 flight-52645 flight-52646 flight-52647 flight-52648 flight-52649 flight-52650 flight-52651 flight-52652 flight-52653 flight-52655 flight-52656 flight-52657 flight-52658 flight-52660 flight-52661 flight-52662 flight-52663 flight-52664 flight-52665 flight-52666 flight-52667 flight-52668 flight-52669 flight-52670 flight-52671 flight-52672 flight-52673 flight-52674 flight-52675 flight-52676 flight-52677 flight-52678 flight-52679 flight-52680 flight-52681 flight-52682 flight-52683 flight-52684 flight-52685 flight-52686 flight-52687 flight-52688 flight-52689 flight-52690 flight-52691 flight-52692 flight-52693 flight-52694 flight-52695 flight-52696 flight-52697 flight-52698 flight-52699 flight-52700 flight-52701 flight-52702 flight-52703 flight-52704 flight-52705 flight-52706 flight-52707 flight-52708 flight-52709 flight-52710 flight-52711 flight-52712 flight-52713 flight-52714 flight-52715 flight-52716 flight-52717 flight-52718 flight-52719 flight-52720 flight-52721 flight-52722 flight-52723 flight-52724 flight-52725 flight-52726 flight-52727 flight-52728 flight-52729 flight-52730 flight-52731 flight-52732 flight-52733 flight-52734 flight-52735 flight-52736 flight-52739 flight-52740 flight-52741 flight-52742 flight-52743 flight-52744 flight-52745 flight-52746 flight-52747 flight-52748 flight-52749 flight-52750 flight-52751 flight-52752 flight-52753 flight-52754 flight-52755 flight-52756 flight-52757 flight-52758 flight-52759 flight-52760 flight-52761 flight-52762 flight-52763 flight-52764 flight-52765 flight-52766 flight-52767 flight-52768 flight-52769 flight-52770 flight-52771 flight-52772 flight-52773 flight-52774 flight-52775 flight-52776 flight-52777 flight-52778 flight-52779 flight-52780 flight-52781 flight-52782 flight-52783 flight-52784 flight-52785 flight-52786 flight-52787 flight-52788 flight-52789 flight-52790 flight-52791 flight-52792 flight-52793 flight-52794 flight-52795 flight-52796 flight-52797 flight-52798 flight-52799 flight-52800 flight-52801 flight-52802 flight-52803 flight-52804 flight-52805 flight-52806 flight-52807 flight-52808 flight-52809 flight-52810 flight-52811 flight-52812 flight-52813 flight-52814 flight-52815 flight-52816 flight-52817 flight-52818 flight-52819 flight-52820 flight-52821 flight-52822 flight-52823 flight-52824 flight-52825 flight-52826 flight-52827 flight-52828 flight-52829 flight-52830 flight-52831 flight-52832 flight-52833 flight-52834 flight-52835 flight-52836 flight-52837 flight-52838 flight-52839 flight-52840 flight-52841 flight-52842 flight-52843 flight-52844 flight-52845 flight-52846 flight-52847 flight-52848 flight-52849 flight-52850 flight-52851 flight-52852 flight-52853 flight-52855 flight-52856 flight-52857 flight-52858 flight-52859 flight-52860 flight-52861 flight-52862 flight-52863 flight-52864 flight-52865 flight-52866 flight-52867 flight-52868 flight-52869 flight-52870 flight-52871 flight-52872 flight-52873 flight-52874 flight-52875 flight-52876 flight-52877 flight-52878 flight-52879 flight-52880 flight-52881 flight-52882 flight-52883 flight-52884 flight-52885 flight-52887 flight-52888 flight-52889 flight-52890 flight-52891 flight-52892 flight-52893 flight-52894 flight-52895 flight-52896 flight-52897 flight-52898 flight-52899 flight-52900 flight-52901 flight-52902 flight-52903 flight-52909 flight-52910 flight-52911 flight-52912 flight-52913 flight-52914 flight-52915 flight-52916 flight-52917 flight-52918 flight-52919 flight-52920 flight-52921 flight-52922 flight-52923 flight-52924 flight-52925 flight-52926 flight-52927 flight-52928 flight-52929 flight-52930 flight-52931 flight-52932 flight-52933 flight-52934 flight-52935 flight-52936 flight-52937 flight-52938 flight-52939 flight-52940 flight-52941 flight-52942 flight-52943 flight-52944 flight-52945 flight-52946 flight-52947 flight-52948 flight-52949 flight-52950 flight-52951 flight-52952 flight-52953 flight-52954 flight-52955 flight-52956 flight-52957 flight-52958 flight-52959 flight-52960 flight-52961 flight-52962 flight-52963 flight-52964 flight-52965 flight-52966 flight-52967 flight-52968 flight-52969 flight-52970 flight-52971 flight-52972 flight-52973 flight-52974 flight-52975 flight-52976 flight-52977 flight-52978 flight-52979 flight-52980 flight-52981 flight-52982 flight-52983 flight-52984 flight-52985 flight-52986 flight-52987 flight-52988 flight-52989 flight-52991 flight-52992 flight-52993 flight-52994 flight-52996 flight-52997 flight-52998 flight-52999 flight-53000 flight-53001 flight-53005 flight-53006 flight-53007 flight-53008 flight-53009 flight-53010 flight-53011 flight-53012 flight-53013 flight-53014 flight-53015 flight-53016 flight-53017 flight-53018 flight-53019 flight-53020 flight-53021 flight-53022 flight-53023 flight-53024 flight-53025 flight-53026 flight-53027 flight-53028 flight-53029 flight-53030 flight-53031 flight-53033 flight-53036 flight-53037 flight-53038 flight-53039 flight-53040 flight-53041 flight-53042 flight-53043 flight-53044 flight-53045 flight-53047 flight-53048 flight-53049 flight-53050 flight-53051 flight-53052 flight-53053 flight-53054 flight-53055 flight-53056 flight-53057 flight-53058 flight-53059 flight-53060 flight-53061 flight-53063 flight-53064 flight-53065 flight-53066 flight-53067 flight-53068 flight-53069 flight-53070 flight-53071 flight-53072 flight-53074 flight-53075 flight-53076 flight-53077 flight-53078 flight-53079 flight-53080 flight-53081 flight-53082 flight-53083 flight-53084 flight-53085 flight-53086 flight-53087 flight-53088 flight-53089 flight-53090 flight-53091 flight-53092 flight-53093 flight-53094 flight-53095 flight-53096 flight-53097 flight-53098 flight-53099 flight-53100 flight-53101 flight-53102 flight-53103 flight-53104 flight-53105 flight-53106 flight-53107 flight-53108 flight-53109 flight-53110 flight-53111 flight-53112 flight-53113 flight-53114 flight-53115 flight-53116 flight-53117 flight-53118 flight-53119 flight-53120 flight-53121 flight-53122 flight-53123 flight-53124 flight-53125 flight-53126 flight-53127 flight-53128 flight-53129 flight-53130 flight-53131 flight-53132 flight-53133 flight-53134 flight-53135 flight-53136 flight-53137 flight-53138 flight-53139 flight-53140 flight-53141 flight-53142 flight-53143 flight-53144 flight-53145 flight-53146 flight-53147 flight-53148 flight-53149 flight-53150 flight-53151 flight-53152 flight-53153 flight-53154 flight-53155 flight-53156 flight-53157 flight-53158 flight-53159 flight-53160 flight-53161 flight-53162 flight-53163 flight-53164 flight-53165 flight-53166 flight-53167 flight-53168 flight-53169 flight-53170 flight-53171 flight-53172 flight-53173 flight-53174 flight-53175 flight-53176 flight-53177 flight-53178 flight-53179 flight-53180 flight-53181 flight-53182 flight-53183 flight-53184 flight-53185 flight-53186 flight-53187 flight-53188 flight-53189 flight-53190 flight-53191 flight-53192 flight-53194 flight-53195 flight-53196 flight-53197 flight-53198 flight-53199 flight-53200 flight-53201 flight-53202 flight-53203 flight-53204 flight-53205 flight-53206 flight-53207 flight-53208 flight-53209 flight-53210 flight-53211 flight-53212 flight-53213 flight-53214 flight-53215 flight-53216 flight-53217 flight-53218 flight-53219 flight-53220 flight-53221 flight-53222 flight-53223 flight-53224 flight-53225 flight-53226 flight-53227 flight-53228 flight-53229 flight-53230 flight-53231 flight-53232 flight-53233 flight-53234 flight-53235 flight-53236 flight-53237 flight-53238 flight-53239 flight-53240 flight-53241 flight-53242 flight-53243 flight-53244 flight-53245 flight-53246 flight-53247 flight-53248 flight-53249 flight-53250 flight-53251 flight-53252 flight-53253 flight-53254 flight-53255 flight-53256 flight-53257 flight-53258 flight-53259 flight-53260 flight-53261 flight-53262 flight-53263 flight-53264 flight-53265 flight-53266 flight-53267 flight-53268 flight-53269 flight-53270 flight-53271 flight-53272 flight-53273 flight-53274 flight-53275 flight-53276 flight-53277 flight-53278 flight-53279 flight-53280 flight-53281 flight-53282 flight-53283 flight-53284 flight-53285 flight-53286 flight-53287 flight-53288 flight-53289 flight-53290 flight-53291 flight-53292 flight-53293 flight-53294 flight-53295 flight-53296 flight-53297 flight-53298 flight-53299 flight-53300 flight-53301 flight-53302 flight-53303 flight-53304 flight-53305 flight-53306 flight-53307 flight-53308 flight-53309 flight-53310 flight-53311 flight-53312 flight-53313 flight-53314 flight-53315 flight-53316 flight-53317 flight-53318 flight-53319 flight-53320 flight-53321 flight-53322 flight-53323 flight-53324 flight-53325 flight-53326 flight-53327 flight-53328 flight-53329 flight-53330 flight-53331 flight-53332 flight-53333 flight-53334 flight-53335 flight-53336 flight-53337 flight-53338 flight-53339 flight-53340 flight-53341 flight-53342 flight-53343 flight-53344 flight-53345 flight-53346 flight-53347 flight-53348 flight-53349 flight-53350 flight-53351 flight-53352 flight-53353 flight-53354 flight-53355 flight-53356 flight-53357 flight-53358 flight-53359 flight-53360 flight-53361 flight-53362 flight-53363 flight-53364 flight-53365 flight-53366 flight-53367 flight-53368 flight-53369 flight-53370 flight-53371 flight-53372 flight-53373 flight-53374 flight-53375 flight-53376 flight-53377 flight-53378 flight-53379 flight-53380 flight-53381 flight-53382 flight-53383 flight-53384 flight-53385 flight-53386 flight-53387 flight-53388 flight-53389 flight-53390 flight-53391 flight-53392 flight-53393 flight-53394 flight-53395 flight-53396 flight-53397 flight-53398 flight-53399 flight-53400 flight-53401 flight-53402 flight-53403 flight-53404 flight-53405 flight-53406 flight-53407 flight-53408 flight-53409 flight-53410 flight-53411 flight-53412 flight-53413 flight-53414 flight-53415 flight-53416 flight-53417 flight-53418 flight-53419 flight-53420 flight-53421 flight-53422 flight-53423 flight-53424 flight-53425 flight-53426 flight-53427 flight-53428 flight-53429 flight-53430 flight-53431 flight-53432 flight-53433 flight-53434 flight-53435 flight-53436 flight-53437 flight-53438 flight-53439 flight-53440 flight-53441 flight-53442 flight-53443 flight-53444 flight-53445 flight-53446 flight-53447 flight-53448 flight-53449 flight-53450 flight-53451 flight-53452 flight-53453 flight-53454 flight-53455 flight-53456 flight-53457 flight-53458 flight-53459 flight-53460 flight-53461 flight-53462 flight-53463 flight-53464 flight-53465 flight-53466 flight-53467 flight-53468 flight-53469 flight-53470 flight-53471 flight-53472 flight-53473 flight-53474 flight-53475 flight-53476 flight-53477 flight-53478 flight-53479 flight-53480 flight-53481 flight-53482 flight-53483 flight-53484 flight-53485 flight-53486 flight-53487 flight-53488 flight-53489 flight-53490 flight-53491 flight-53492 flight-53493 flight-53494 flight-53495 flight-53496 flight-53497 flight-53498 flight-53499 flight-53500 flight-53501 flight-53502 flight-53503 flight-53504 flight-53505 flight-53506 flight-53507 flight-53508 flight-53509 flight-53510 flight-53511 flight-53512 flight-53513 flight-53514 flight-53515 flight-53516 flight-53517 flight-53518 flight-53519 flight-53520 flight-53521 flight-53522 flight-53523 flight-53524 flight-53525 flight-53526 flight-53527 flight-53529 flight-53530 flight-53531 flight-53532 flight-53533 flight-53534 flight-53535 flight-53536 flight-53537 flight-53538 flight-53539 flight-53540 flight-53541 flight-53542 flight-53543 flight-53544 flight-53545 flight-53546 flight-53547 flight-53548 flight-53549 flight-53550 flight-53551 flight-53552 flight-53553 flight-53554 flight-53555 flight-53556 flight-53557 flight-53558 flight-53559 flight-53560 flight-53561 flight-53562 flight-53563 flight-53564 flight-53565 flight-53566 flight-53567 flight-53568 flight-53569 flight-53570 flight-53571 flight-53572 flight-53573 flight-53574 flight-53575 flight-53576 flight-53577 flight-53578 flight-53579 flight-53580 flight-53581 flight-53582 flight-53583 flight-53584 flight-53585 flight-53586 flight-53587 flight-53588 flight-53589 flight-53590 flight-53591 flight-53592 flight-53593 flight-53594 flight-53595 flight-53596 flight-53597 flight-53598 flight-53599 flight-53600 flight-53601 flight-53602 flight-53603 flight-53604 flight-53605 flight-53606 flight-53607 flight-53608 flight-53609 flight-53610 flight-53611 flight-53612 flight-53613 flight-53614 flight-53615 flight-53616 flight-53617 flight-53618 flight-53619 flight-53620 flight-53621 flight-53622 flight-53623 flight-53624 flight-53625 flight-53626 flight-53627 flight-53628 flight-53629 flight-53630 flight-53631 flight-53632 flight-53633 flight-53634 flight-53635 flight-53636 flight-53637 flight-53638 flight-53639 flight-53640 flight-53641 flight-53642 flight-53643 flight-53644 flight-53645 flight-53646 flight-53647 flight-53648 flight-53649 flight-53650 flight-53651 flight-53652 flight-53653 flight-53654 flight-53655 flight-53656 flight-53657 flight-53658 flight-53659 flight-53660 flight-53661 flight-53662 flight-53663 flight-53664 flight-53665 flight-53666 flight-53667 flight-53668 flight-53669 flight-53670 flight-53671 flight-53672 flight-53673 flight-53674 flight-53675 flight-53676 flight-53677 flight-53678 flight-53679 flight-53680 flight-53681 flight-53682 flight-53683 flight-53684 flight-53685 flight-53686 flight-53687 flight-53688 flight-53689 flight-53690 flight-53691 flight-53692 flight-53693 flight-53694 flight-53695 flight-53696 flight-53697 flight-53698 flight-53699 flight-53700 flight-53701 flight-53702 flight-53703 flight-53704 flight-53705 flight-53706 flight-53707 flight-53708 flight-53709 flight-53710 flight-53711 flight-53712 flight-53713 flight-53714 flight-53715 flight-53716 flight-53717
commit | commitdiff | tree
Roger Pau Monne [Thu, 9 Apr 2015 15:11:13 +0000 (17:11 +0200)]
FreeBSD: Cleanups relating to guest images and ts-freebsd-install script

Remove some unused variables from ts-freebsd-install script. Also make the
third parameter of target_put_guest_image optional and fix both callers of
this function.

Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
10 years agoFreeBSD: Increase guest disk size
commit | commitdiff | tree
Roger Pau Monne [Tue, 14 Apr 2015 10:53:21 +0000 (12:53 +0200)]
FreeBSD: Increase guest disk size

New 10.1 images are larger than the previous 10.0 images, so change
the size of the LVM volume to accommodate them, in preparation.
Increase the size to 24000 in case of future increases upstream.

Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoFreeBSD: Update FreeBSD guests to 10.1
commit | commitdiff | tree
Roger Pau Monne [Thu, 9 Apr 2015 15:11:12 +0000 (17:11 +0200)]
FreeBSD: Update FreeBSD guests to 10.1

Update FreeBSD guests in OSSTest to FreeBSD 10.1. The following images
should be placed in the osstest images folder:

ftp://ftp.freebsd.org/pub/FreeBSD/releases/VM-IMAGES/10.1-RELEASE/amd64/Latest/FreeBSD-10.1-RELEASE-amd64.raw.xz
ftp://ftp.freebsd.org/pub/FreeBSD/releases/VM-IMAGES/10.1-RELEASE/i386/Latest/FreeBSD-10.1-RELEASE-i386.raw.xz

Since new images are in raw format rather than qcow2 remove the runes to
convert from qcow2 to raw.

Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoReduce guest ssh check up timeout to 10s flight-50331 flight-50355 flight-50356 flight-50357 flight-50358 flight-50359 flight-50360 flight-50361 flight-50362 flight-50363 flight-50364 flight-50365 flight-50366 flight-50367 flight-50368 flight-50369 flight-50370 flight-50371 flight-50372 flight-50373 flight-50374 flight-50375 flight-50376 flight-50380 flight-50382 flight-50383 flight-50384 flight-50385 flight-50386 flight-50388 flight-50389 flight-50390 flight-50391 flight-50392 flight-50393 flight-50394 flight-50395 flight-50396 flight-50398 flight-50399 flight-50400 flight-50401 flight-50402 flight-50403 flight-50404 flight-50405 flight-50406 flight-50407 flight-50408 flight-50409 flight-50410 flight-50411 flight-50412 flight-50413 flight-50414 flight-50416 flight-50417 flight-50418 flight-50419 flight-50420 flight-50422 flight-50424 flight-50425 flight-50427 flight-50428 flight-50429 flight-50430 flight-50431 flight-50432 flight-50433 flight-50435 flight-50436 flight-50437 flight-50438 flight-50439 flight-50441 flight-50442 flight-50443 flight-50445 flight-50446 flight-50447 flight-50449 flight-50450 flight-50451 flight-50453 flight-50454 flight-50455 flight-50456 flight-50457 flight-50459 flight-50460 flight-50461 flight-50462 flight-50463 flight-50464 flight-50465 flight-50466 flight-50467 flight-50468 flight-50469 flight-50470 flight-50473 flight-50474 flight-50475 flight-50476 flight-50478 flight-50479 flight-50480 flight-50481 flight-50482 flight-50483 flight-50484 flight-50485 flight-50486 flight-50488 flight-50489 flight-50490 flight-50491 flight-50492 flight-50493 flight-50494 flight-50495 flight-50496 flight-50497 flight-50498 flight-50500 flight-50502 flight-50503 flight-50504 flight-50505 flight-50506 flight-50507
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 14:58:00 +0000 (14:58 +0000)]
Reduce guest ssh check up timeout to 10s

The default for target_cmd_root is 30s but 10s should be plenty.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoUse ssh -v when ssh'ing to guests to check they are up
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 14:55:20 +0000 (14:55 +0000)]
Use ssh -v when ssh'ing to guests to check they are up

* Make tcmd take an optional $extrasshopts argument
* Make target_cmd{,_root} pass this through
* Supply this with value [qw(-v)]

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agocambridge: Set HostProp_DhcpWatchMethod back to woking
commit | commitdiff | tree
Ian Campbell [Thu, 2 Apr 2015 10:40:09 +0000 (11:40 +0100)]
cambridge: Set HostProp_DhcpWatchMethod back to woking

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agots-host-ping-check: Do not run if host is being reused
commit | commitdiff | tree
Ian Campbell [Wed, 1 Apr 2015 13:07:09 +0000 (14:07 +0100)]
ts-host-ping-check: Do not run if host is being reused

Assume it was done at install time.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agostandalone: Add --dry-run option for run-job.
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 15:06:47 +0000 (16:06 +0100)]
standalone: Add --dry-run option for run-job.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agotcl: Handle environment variables which are unset.
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 15:06:46 +0000 (16:06 +0100)]
tcl: Handle environment variables which are unset.

This allows wrappers such as the standalone wrapper to do
     OSSTEST_SIMULATE=$foo ./sg-run-job
and not worry if $foo is unset.

Do likewise for OSSTEST_TCL_JOBDB_DEBUG.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoMerge branch 'production' of osstest.test-lab:/home/osstest/testing into wip
commit | commitdiff | tree
Ian Jackson [Wed, 1 Apr 2015 10:12:14 +0000 (11:12 +0100)]
Merge branch 'production' of osstest.test-lab:/home/osstest/testing into wip

10 years agoproduction-config: Adjust LogsMinSpaceMby to not be half of the VM's disk flight-50306 flight-50307 flight-50308 flight-50309 flight-50310 flight-50311 flight-50312 flight-50313 flight-50314 flight-50315 flight-50316 flight-50317 flight-50318 flight-50319 flight-50320 flight-50321 flight-50322 flight-50323 flight-50324 flight-50325 flight-50326 flight-50327 flight-50328 flight-50329 flight-50330 flight-50332 flight-50333 flight-50334 flight-50335 flight-50336 flight-50337 flight-50339 flight-50341
commit | commitdiff | tree
Ian Jackson [Wed, 1 Apr 2015 10:10:58 +0000 (11:10 +0100)]
production-config: Adjust LogsMinSpaceMby to not be half of the VM's disk

10 years agoMerge branch 'colo.2015-03-31' of xenbits.xen.org:/home/iwj/ext/osstest into queue flight-50284
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 14:49:15 +0000 (15:49 +0100)]
Merge branch 'colo.2015-03-31' of xenbits.xen.org:/home/iwj/ext/osstest into queue

10 years agoproduction-config*: Set HostnameSortSwapWords as appropriate
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 14:39:19 +0000 (15:39 +0100)]
production-config*: Set HostnameSortSwapWords as appropriate

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years ago(Re)Add production config for Cambridge instance
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 13:25:34 +0000 (14:25 +0100)]
(Re)Add production config for Cambridge instance

Although it is no longer the production instance I'd like to be able
to use it for ad-hoc testing.

Created with:
 $ cp production-config production-config-cambridge
 $ git diff origin/production origin-cam/incoming -- production-config | patch production-config-cambridge

Where origin is the new production colo, origin-cam is the old
instance and:
$ git rev-parse origin/production
dce745682f1e326a04519908f07c0d589bd0fc62
$ git rev-parse origin-cam/incoming
a623769cbac25ae716ff9d8251116f95e2ee2531

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agoOsstest.pm: Add "use IO::File"
commit | commitdiff | tree
Ian Campbell [Tue, 31 Mar 2015 14:38:52 +0000 (15:38 +0100)]
Osstest.pm: Add "use IO::File"

This is needed under Squeeze (but apparently not Wheezy)

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
10 years agots-host-install: honour arch-specific boot append properties
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:51:42 +0000 (13:51 +0000)]
ts-host-install: honour arch-specific boot append properties

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agots-host-ping-check: New ubiquitous test step
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:50:43 +0000 (13:50 +0000)]
ts-host-ping-check: New ubiquitous test step

Check that packet loss is within acceptable levels.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agosg-report-flight: Produce better output for running jobs.
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:49:57 +0000 (13:49 +0000)]
sg-report-flight: Produce better output for running jobs.

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agosg-report-flight: Sort email output by results, not job name
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:49:42 +0000 (13:49 +0000)]
sg-report-flight: Sort email output by results, not job name

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agocs-adjust-flight: Support negation of foo-specs
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:48:15 +0000 (13:48 +0000)]
cs-adjust-flight: Support negation of foo-specs

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoconfig: turn off HostnameSortSwapWords by default
commit | commitdiff | tree
Ian Jackson [Tue, 31 Mar 2015 13:47:26 +0000 (13:47 +0000)]
config: turn off HostnameSortSwapWords by default

Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agostart/stop the guest 10 times in the standard test jobs
commit | commitdiff | tree
Ian Campbell [Wed, 11 Mar 2015 10:08:57 +0000 (10:08 +0000)]
start/stop the guest 10 times in the standard test jobs

The recent libvirt failures seemed to happen after a couple of
stop/start pairs, so arrange to start/stop each guest 10 times.

By cribbing from the existing use of repeat-ts I hope I've arranged
for a new test with testid ts-guest-start/debian.repeat.

We already arrange by more adhoc means to repeat a localhost migration
10 times.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
10 years agoREADME.dev: Document steps after restarting ms-ownerdaemon
commit | commitdiff | tree
Ian Campbell [Fri, 13 Mar 2015 16:00:20 +0000 (16:00 +0000)]
README.dev: Document steps after restarting ms-ownerdaemon

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
next
Xen Test Framework and Suite, used for Open Source Xen CI push gate