Tony Mack [Mon, 16 Jul 2012 16:52:43 +0000 (12:52 -0400)]
revert changes to plc_configure()
Tony Mack [Mon, 16 Jul 2012 16:47:30 +0000 (12:47 -0400)]
use sfa-config-tty to configure sfa
Thierry Parmentelat [Mon, 9 Jul 2012 10:18:47 +0000 (12:18 +0200)]
Setting tag tests-5.1-7
improved (now cross_)check_tcp to run more combos or server/client, and use 2 nodes whenever possible
new step sfa-renew-slice
new convenience step debug-nodemanager
run export to report only about 1st plc
new step sfa-fsclean to clean up filesystem
Thierry Parmentelat [Sun, 8 Jul 2012 23:01:10 +0000 (01:01 +0200)]
bugfix
Thierry Parmentelat [Thu, 5 Jul 2012 16:14:39 +0000 (18:14 +0200)]
new step sfa_fsclean, and added to the 'sfa' macro step
Thierry Parmentelat [Thu, 5 Jul 2012 13:15:04 +0000 (15:15 +0200)]
return True for export on secondary plcs
Thierry Parmentelat [Thu, 5 Jul 2012 10:13:06 +0000 (12:13 +0200)]
revert to using sfaadmin reg import_registry for lxc16 builds that
don't run the latest sfa
Thierry Parmentelat [Wed, 4 Jul 2012 15:30:34 +0000 (17:30 +0200)]
try sfaadmin reg import
Thierry Parmentelat [Wed, 4 Jul 2012 07:10:16 +0000 (09:10 +0200)]
tweak run export for multi-plcs, only report about the first one
Thierry Parmentelat [Fri, 29 Jun 2012 07:57:30 +0000 (09:57 +0200)]
tweak check_tcp again
Thierry Parmentelat [Wed, 27 Jun 2012 18:47:20 +0000 (20:47 +0200)]
try the tcp client twice
Thierry Parmentelat [Wed, 27 Jun 2012 18:42:12 +0000 (20:42 +0200)]
increase timeout of tcp server
Thierry Parmentelat [Sun, 24 Jun 2012 22:23:07 +0000 (00:23 +0200)]
add sfa_renew_slice to the sfa macro
Thierry Parmentelat [Sun, 24 Jun 2012 22:03:40 +0000 (00:03 +0200)]
bugfix
Thierry Parmentelat [Thu, 21 Jun 2012 14:58:37 +0000 (16:58 +0200)]
new sfa_renew_slice step
Thierry Parmentelat [Thu, 21 Jun 2012 11:14:20 +0000 (13:14 +0200)]
improve check_tcp a lot
Thierry Parmentelat [Thu, 21 Jun 2012 08:38:46 +0000 (10:38 +0200)]
minor fix
Thierry Parmentelat [Tue, 19 Jun 2012 13:14:26 +0000 (15:14 +0200)]
new step debug_nodemanager
Thierry Parmentelat [Fri, 15 Jun 2012 05:01:36 +0000 (08:01 +0300)]
sfi_slices out of sfa_standalone, again
Thierry Parmentelat [Tue, 12 Jun 2012 10:49:56 +0000 (12:49 +0200)]
kvm64-{4,5} often end up with nodes spending more than 20 minutes
getting started, let's lower their burden
Thierry Parmentelat [Tue, 12 Jun 2012 09:24:55 +0000 (11:24 +0200)]
oops
Thierry Parmentelat [Tue, 12 Jun 2012 09:22:39 +0000 (11:22 +0200)]
checking system slice will try netflow and drl
Thierry Parmentelat [Tue, 12 Jun 2012 09:05:43 +0000 (11:05 +0200)]
sfi_slices requires an aggregate, so won't work in standalone mode
Thierry Parmentelat [Mon, 11 Jun 2012 09:50:08 +0000 (11:50 +0200)]
typo
Thierry Parmentelat [Sun, 3 Jun 2012 13:29:32 +0000 (15:29 +0200)]
Setting tag tests-5.1-6
missing import
Thierry Parmentelat [Sun, 3 Jun 2012 12:10:07 +0000 (14:10 +0200)]
missing import
Thierry Parmentelat [Sat, 2 Jun 2012 12:25:59 +0000 (14:25 +0200)]
Setting tag tests-5.1-5
refactored the sfa area
use new sfi interface (options instead of xml file)
Thierry Parmentelat [Sat, 2 Jun 2012 11:25:34 +0000 (13:25 +0200)]
reviewed imports between TestAuthSfa and TestSliceSfa
Thierry Parmentelat [Sat, 2 Jun 2012 11:08:06 +0000 (13:08 +0200)]
enable regular sfa user with sfi update
Thierry Parmentelat [Fri, 1 Jun 2012 21:03:41 +0000 (23:03 +0200)]
a few more tweaks
Thierry Parmentelat [Fri, 1 Jun 2012 19:11:36 +0000 (21:11 +0200)]
missing import
Thierry Parmentelat [Fri, 1 Jun 2012 18:00:42 +0000 (20:00 +0200)]
now that we create an SFA auth to hold the sfa entities involved:
refactored tests to split TestAuthSfa and TestSliceSfa
cleaned up the config side a lot as well as a bonus
Thierry Parmentelat [Thu, 31 May 2012 16:22:06 +0000 (18:22 +0200)]
use current sfi with --extra
Thierry Parmentelat [Wed, 30 May 2012 13:56:47 +0000 (15:56 +0200)]
reviewed the ssh keys system
use 5 different keys for pladmins, plpis, plusers, sfapis and sfausers
kind of intrusive so probably not quite thorough
Thierry Parmentelat [Wed, 30 May 2012 11:01:00 +0000 (13:01 +0200)]
use 4 different keys for (pl,sfa) x (pi,user)
invoke sfi instead of sfi.py
Thierry Parmentelat [Wed, 30 May 2012 10:58:10 +0000 (12:58 +0200)]
need to cleanup sfi work dirs
Thierry Parmentelat [Wed, 30 May 2012 10:03:34 +0000 (12:03 +0200)]
user creation is done with new sfi options instead of xml-based
Thierry Parmentelat [Wed, 30 May 2012 10:02:55 +0000 (12:02 +0200)]
step sfa-create needs site and pi creation as well
Thierry Parmentelat [Fri, 25 May 2012 08:53:59 +0000 (10:53 +0200)]
provide a bit more info when check-sys-slice fails
Thierry Parmentelat [Thu, 24 May 2012 08:17:07 +0000 (10:17 +0200)]
Merge branch 'master' of ssh://git.onelab.eu/git/tests
Thierry Parmentelat [Thu, 24 May 2012 08:16:52 +0000 (10:16 +0200)]
builds that don't provide SFA need to skip steps in 'sfi' as well
Thierry Parmentelat [Wed, 23 May 2012 13:51:59 +0000 (15:51 +0200)]
Setting tag tests-5.1-4
check for slice to be brought down
new speed-up-slices step
reviewed for standalone SFA
Thierry Parmentelat [Wed, 23 May 2012 10:36:47 +0000 (12:36 +0200)]
comment
Thierry Parmentelat [Wed, 23 May 2012 10:16:02 +0000 (12:16 +0200)]
sfa_import only relies on sfaadmin
sfa_standalone does invoke sfa-import to create the toplevel auth record
Thierry Parmentelat [Wed, 23 May 2012 10:04:36 +0000 (12:04 +0200)]
improve formatting of run -lv for nested macros
Thierry Parmentelat [Wed, 23 May 2012 09:41:28 +0000 (11:41 +0200)]
renamed sfa_* into sfi_* for steps list/show/slices
Thierry Parmentelat [Wed, 23 May 2012 09:35:56 +0000 (11:35 +0200)]
fix old bug about doc strings for mapped methods
trim TestPlc accordingly, have the doc where it belongs
split sfa_view into sfa_list sfa_show and sfa_slices
Thierry Parmentelat [Wed, 23 May 2012 08:19:28 +0000 (10:19 +0200)]
attempt to workaround the odd issue with yum failing to install sfa-client
[('installing package sfa-client-2.1-7.onelab.2012.05.23.i686 needs 68KB on the / filesystem', (9, '/', 69632L))]
Thierry Parmentelat [Tue, 22 May 2012 14:44:36 +0000 (16:44 +0200)]
using flavour=void, the sfa-standalone step can check more stuff
Thierry Parmentelat [Tue, 22 May 2012 14:44:08 +0000 (16:44 +0200)]
new step speed-up-slices
finer-grained steps to remove nodes from slice
check that sliver is *not* reachable once deleted
various fixes
Thierry Parmentelat [Tue, 22 May 2012 14:41:55 +0000 (16:41 +0200)]
continued
Thierry Parmentelat [Tue, 22 May 2012 14:39:18 +0000 (16:39 +0200)]
minor fix for handling _ and - in step names
Thierry Parmentelat [Tue, 22 May 2012 14:38:46 +0000 (16:38 +0200)]
-c sfavoid allows to test with flavour=void
Thierry Parmentelat [Mon, 21 May 2012 14:46:02 +0000 (16:46 +0200)]
dry_run might be undefined in options
Thierry Parmentelat [Mon, 21 May 2012 14:44:31 +0000 (16:44 +0200)]
missing import
Thierry Parmentelat [Mon, 21 May 2012 14:43:34 +0000 (16:43 +0200)]
avoid infinite recursion
Thierry Parmentelat [Tue, 15 May 2012 13:05:02 +0000 (15:05 +0200)]
use simpler names for sfa entities
Thierry Parmentelat [Tue, 15 May 2012 12:49:33 +0000 (14:49 +0200)]
stop sfa before cleaning up - might lock the db otherwise
Thierry Parmentelat [Tue, 15 May 2012 08:15:26 +0000 (10:15 +0200)]
cosmetic
Thierry Parmentelat [Sun, 13 May 2012 21:24:41 +0000 (23:24 +0200)]
Merge branch 'master' of ssh://git.onelab.eu/git/tests
Thierry Parmentelat [Sun, 13 May 2012 21:24:30 +0000 (23:24 +0200)]
fix typo
Thierry Parmentelat [Sat, 12 May 2012 16:06:40 +0000 (18:06 +0200)]
Setting tag tests-5.1-3
the sfa_slice scenario now uses sfaadmin to bootstrap a specific site
where the sfa-related entries are confined
it should thus in principle be possible to test sfa without installing myplc
also added a new key (key3) but that currently is not used ...
Thierry Parmentelat [Fri, 11 May 2012 18:22:31 +0000 (20:22 +0200)]
Merge branch 'master' of ssh://git.onelab.eu/git/tests
Thierry Parmentelat [Fri, 11 May 2012 18:22:19 +0000 (20:22 +0200)]
first draft at a scenario where SFA stuff gets created independently
from myplc, ussing sfaadmin to bootstrap the site & pi
Thierry Parmentelat [Mon, 7 May 2012 10:30:01 +0000 (12:30 +0200)]
Setting tag tests-5.1-2
knowns about the new interface API - and got rid of lxc-wait
Mohamed Larabi [Fri, 4 May 2012 12:52:51 +0000 (14:52 +0200)]
remove lxc-wait calls
Thierry Parmentelat [Thu, 26 Apr 2012 07:25:41 +0000 (09:25 +0200)]
fix the way we determine if the PLCAPI is newinterface (AddIpAddress) or not
Thierry Parmentelat [Tue, 24 Apr 2012 09:32:49 +0000 (11:32 +0200)]
clearer comments on how the newint stuff is used
Thierry Parmentelat [Tue, 24 Apr 2012 09:09:51 +0000 (11:09 +0200)]
some code uses list() with no verbose arguement
Thierry Parmentelat [Tue, 24 Apr 2012 07:29:48 +0000 (09:29 +0200)]
LocalSubstrate.py by default also lists the test_box (only running tests)
Can get the full list of tests using -tv
Thierry Parmentelat [Mon, 23 Apr 2012 16:50:23 +0000 (18:50 +0200)]
knows about the new plcapi interface about ipaddresses and routes
Thierry Parmentelat [Mon, 16 Apr 2012 14:33:29 +0000 (16:33 +0200)]
Setting tag tests-5.1-1
extra -X option to use an lxc-based vplc host box
extra step to check for system slices creation
Thierry Parmentelat [Fri, 13 Apr 2012 13:06:08 +0000 (15:06 +0200)]
check system slice using 'drl' for now
Thierry Parmentelat [Thu, 12 Apr 2012 05:20:54 +0000 (07:20 +0200)]
gather plc's /root/sfi into testlogs
Thierry Parmentelat [Wed, 11 Apr 2012 16:35:35 +0000 (18:35 +0200)]
cosmetic
Mohamed Larabi [Fri, 6 Apr 2012 08:58:58 +0000 (10:58 +0200)]
PlcLxcBox sensing
Mohamed Larabi [Mon, 2 Apr 2012 22:14:42 +0000 (00:14 +0200)]
put lxcname.timestamp file inside /var/lib/lxc/lxcname/
Mohamed Larabi [Mon, 2 Apr 2012 22:08:25 +0000 (00:08 +0200)]
put lxcname.timestamp file inside /var/lib/lxc/lxcname/
Mohamed Larabi [Mon, 2 Apr 2012 22:08:25 +0000 (00:08 +0200)]
put .timestamp file inside /var/lib/lxc//
Thierry Parmentelat [Mon, 2 Apr 2012 15:39:26 +0000 (17:39 +0200)]
Merge branch 'master' of ssh://git.onelab.eu/git/tests
Thierry Parmentelat [Mon, 2 Apr 2012 15:37:59 +0000 (17:37 +0200)]
review output of run export for vs or lxc
Thierry Parmentelat [Mon, 2 Apr 2012 15:37:38 +0000 (17:37 +0200)]
cormetic
Mohamed Larabi [Sun, 1 Apr 2012 22:25:17 +0000 (00:25 +0200)]
daemonize the container at start time
Thierry Parmentelat [Sun, 1 Apr 2012 20:05:47 +0000 (22:05 +0200)]
fix hos to join guest from lxc host
Thierry Parmentelat [Sun, 1 Apr 2012 14:03:27 +0000 (16:03 +0200)]
containers need to be stopped before they can be destroyed
Thierry Parmentelat [Sat, 31 Mar 2012 15:20:29 +0000 (17:20 +0200)]
fix syntax error
Thierry Parmentelat [Sat, 31 Mar 2012 10:48:24 +0000 (12:48 +0200)]
improved handling vm timestamp for lxc
Thierry Parmentelat [Sat, 31 Mar 2012 10:29:49 +0000 (12:29 +0200)]
first blotch for managing lxc-hosted vplcs
Thierry Parmentelat [Fri, 30 Mar 2012 05:57:26 +0000 (07:57 +0200)]
oops
Thierry Parmentelat [Thu, 29 Mar 2012 14:40:13 +0000 (16:40 +0200)]
again
Thierry Parmentelat [Thu, 29 Mar 2012 14:39:06 +0000 (16:39 +0200)]
again
Thierry Parmentelat [Thu, 29 Mar 2012 14:38:36 +0000 (16:38 +0200)]
less verbose logs
Thierry Parmentelat [Thu, 29 Mar 2012 14:36:14 +0000 (16:36 +0200)]
check_netflow should hopefully be usable
Thierry Parmentelat [Thu, 29 Mar 2012 14:06:17 +0000 (16:06 +0200)]
bump version
Thierry Parmentelat [Thu, 29 Mar 2012 14:06:08 +0000 (16:06 +0200)]
turn off check_netflow for now - need to wait for it to show up
Thierry Parmentelat [Thu, 29 Mar 2012 11:17:11 +0000 (13:17 +0200)]
add check_netflow so we probe for at least one alive system slice
Thierry Parmentelat [Wed, 28 Mar 2012 21:07:30 +0000 (23:07 +0200)]
bootstrapfs vs nodeimage
Thierry Parmentelat [Wed, 28 Mar 2012 09:30:47 +0000 (11:30 +0200)]
cosmetic
Thierry Parmentelat [Wed, 28 Mar 2012 08:04:43 +0000 (10:04 +0200)]
manage displays both vs and lxc boxes
Thierry Parmentelat [Wed, 28 Mar 2012 07:58:51 +0000 (09:58 +0200)]
cosmetic - sense uses 2chars to describe the type of box