Wednesday, 2015-07-29

*** Qiming has quit IRC00:39
openstackgerritxu-haiwei proposed stackforge/senlin: Node module test case part2
lixinhuixuhaiwei, consult one thing:01:07
lixinhuiis the fucntions of Japan Phiten true on health keep or not01:08
*** Qiming has joined #senlin01:28
*** Yanyanhu has joined #senlin01:38
xuhaiweilixinhui, what Phiten?01:41
xuhaiweiok, I know that, I also bought it to my family before, not sure whether it works01:43
*** Yanyan has joined #senlin01:43
*** Yanyanhu has quit IRC01:46
lixinhuiwhich one is better titan or phiten01:49
lixinhuixuhaiwei, maybe not work or not, just reputation in Japanese01:50
xuhaiweiI really don't know, few people beside me have it01:51
lixinhuiokay, thanks, all the same01:52
lixinhuimaybe just a legend outside Jaoan...01:52
*** Zhenqi has joined #senlin02:01
*** Tennyson has joined #senlin02:14
*** Yanyanhu has joined #senlin02:29
*** Yanyan has quit IRC02:31
*** mathspanda has joined #senlin02:32
*** Zhenqi has quit IRC02:47
*** Zhenqi has joined #senlin02:47
*** xuhaiwei has quit IRC03:21
Yanyanhuhi, Qiming, does the gnocchi-aggregate alarm means an aggregation method(e.g. max, min, avg) will be used to calculate the alarm result based on the metric value in specified number of evaluation periods(with given granularity)?03:21
YanyanhuAlso found 'alarm-gnocchi-aggregation-by-metrics-threshold-create' doesn't accept 'query' paramter. So this kind of alarm can only use global metric result to do evaluation?03:30
*** mathspanda has quit IRC03:34
*** mathspanda has joined #senlin03:56
*** xuhaiwei has joined #senlin05:04
QimingYanyanhu, I'm not quite sure06:01
QimingI was only following ceilometer client/server APIs, especially the parameters acceptable for rule definitions06:01
QimingI have to say, it is a mess06:01
Yanyanhuseems we can only try it by ourself...06:01
Qimingyes, we have to install a gnocchi engine to play with it06:02
Qimingthe aggregation, according to the docs, can be done on either resources or metrics06:02
Yanyanhuor you can just start from the basic threshold alarm to build the framework06:03
Yanyanhuand then we can try other alarm types06:03
Qimingyou can aggregate metrics from several resources, or values from several (seemingly irrelevant) metrics06:03
Qimingit doesn't matter that much06:04
Tennysonis this a bug about cluster resize07:19
QimingTennyson, can you find out the location and help fix it?07:20
Qimingwhen issuing a command using senlin command line07:21
Qimingyou can use option -d to print some detailed call trace07:22
Qimingthe server side log can be checked from the senlin-engine and senlin-api console respectively07:22
Tennysonthe bug may occurs in the client side07:23
Tennysonpython-senlin client07:24
Qimingyes, looks like a client side problem07:24
*** Yanyan has joined #senlin07:29
*** Yanyanhu has quit IRC07:32
QimingTennyson, any progress? questions?07:51
Qimingfeel free to ask questions if you need help08:05
Tennysonno adjustment_type defined in min_step, min&max_size08:25
Tennysonbut it need an adjustment_type in invoke_method08:25
*** Zhenqi is now known as Peiyu08:29
Tennysonwe then add a default value (None) to both adjustment_type and number08:30
Peiyuand request accepted.08:31
QimingTennyson, just took a look at the command08:59
Qimingyou were asking for an adjustment of the max_size only, right?08:59
Qimingthat is a .... weird request, but valid09:01
Tennysonif we want to assign -m, we must also set -c or percentage ?09:04
Qimingoh no09:14
Qimingit looks like a utility function error09:14
Qimingwhen invoking action methods, there have been some filtering and parsing logic09:15
Qimingwe have to ensure the action params exist09:15
Qimingso ... there is a quick fix09:15
openstackgerritYanyan Hu proposed stackforge/senlin: Add test case for LoadBalancingPolicy
openstackgerritYanyan Hu proposed stackforge/senlin: Add test case for LoadBalancingPolicy(2)
openstackgerritQiming Teng proposed stackforge/python-senlinclient: Fix cluster-resize operation param checking
QimingTennyson, see the patch above ^09:20
Qimingplease review09:20
Peiyuthis two line code cannot fix the problem09:25
Peiyuif this, cluster -c -m will lead another problem09:26
Peiyucluster-show will find the capacity is 009:26
*** mathspanda has quit IRC09:28
*** Peiyu has quit IRC09:31
*** Tennyson has quit IRC09:34
*** Qiming has quit IRC09:56
openstackgerritYanyan Hu proposed stackforge/senlin: Add test case for LoadBalancingPolicy
openstackgerritYanyan Hu proposed stackforge/senlin: Add test case for LoadBalancingPolicy(2)
*** Yanyan has quit IRC10:12
*** Qiming has joined #senlin10:43
openstackgerritMerged stackforge/senlin: Updated from global requirements
openstackgerritMerged stackforge/senlin: Node module test case part2
*** jruano has joined #senlin13:21
openstackgerritMerged stackforge/senlin: Fix trust listing
*** jruano has quit IRC14:27
openstackgerritQiming Teng proposed stackforge/senlin: Unit test case for ceilometer v2 driver
*** jruano has joined #senlin15:21
*** Qiming has quit IRC15:52
*** openstackgerrit has quit IRC18:31
*** openstackgerrit has joined #senlin18:32
*** openstackgerrit has quit IRC20:16
*** openstackgerrit has joined #senlin20:17
*** jruano has quit IRC20:40
openstackgerritJulio Ruano proposed stackforge/senlin: Add test case for Trust Middleware

Generated by 2.14.0 by Marius Gedminas - find it at!