(Just joined the list and had this forwarded to me - please forgive if
threading is broken)
Quick Cloudmonkey question (CS 4.3.1, Cloudmonkey 5.3.0) - I noticed in
the archives there are questions regarding cloudmonkey and embedded
control characters. We’re still getting them and can’t get rid of
t;
>Hope this helps.
>
>> On 05-Dec-2014, at 1:06 am, Ian Forde wrote:
>>
>> (Just joined the list and had this forwarded to me - please forgive if
>> threading is broken)
>>
>> Quick Cloudmonkey question (CS 4.3.1, Cloudmonkey 5.3.0) - I noticed in
>>
ease
>build/install from latest master and confirm if it¹s fixed? Thanks.
>
>CloudMonkey 5.3.0 does not output colored text when run on bash or in
>scripts, so if you¹re using in those scripts you don¹t need to set color
>off.
>
>> On 05-Dec-2014, at 1:44 am, Ian Forde wrote
I’ve got a bash script performing an API call (in my case, via cloudmonkey) as
follows:
cloudmonkey api listLdapUsers listtype=new display=default
filter=username,domain
(note that we’re using Active Directory)
There’s at least one user in AD defined with a capital letter in front.
Example:
01:48, "ilya musayev" wrote:
>
>> Hi Ian,
>>
>> This is probably better suited for Dev list, Ian Duffy has done alot of
>> work on LDAP and CloudStack, he is cc'd to this thread. Perhaps he can
>>shed
>> some light.
>>
>> Regards
>&g
If I want to see what instances in a project are in the destroyed state via
Cloudmonkey, I should be able to use both the API call and the native
Cloudmonkey call.
[root@sjmpc-mgr1 cloudmanager_utils]# cloudmonkey -d json api
listVirtualMachines domainid=c4fe811a-eeb5-4826-85a6-67c8d3f26234
p
Following up on thisŠ
It¹s via the UI. We¹re using LDAP authentication with Active Directory as
the backend, where AD allows Œ<Œ and Œ>¹ but Cloudstack apparently
doesn¹t. We¹ve disabled connection security on LDAP and used tcpdump to
verify that CS is mistakenly encoding those characters before
(Seems like I have character encoding issues of my own.)
The characters that AD allows but CS doesn’t are the greater than (>) and
less than (<) characters. Hope the previous message wasn’t too garbled
for decipherment…
-I
On 12/10/14, 2:37 PM, "Ian Forde" wrote:
>Fo
Is anyone actively working on the foreman-cloudstack compute resource for
Foreman? There’s a repo on github
(https://github.com/ke4qqq/foreman-cloudstack), but it hasn’t been updated in 3
months. Googling yields references to other repos (foreman/foreman-cloudstack,
citrix/foreman-cloudstack)
on, table) will follow same filtering
>algorithm.
>
>Can you confirm testing against latest master?
>https://github.com/apache/cloudstack-cloudmonkey
>
>On Thursday 11 December 2014 02:28 AM, Ian Forde wrote:
>> If I want to see what instances in a project are in the destroyed st
One thing you could try would be to just setup Apache on the management
server with mod_rewrite, sending your site request down one path, and CS
down the other. This way you could keep similar URLs and not require your
users to remember port numbers (yuck). Hope that helps..
-Ian
On 12/16/14,
Hi all -
I’ve just detected some more strange behavior in Cloudmonkey. If I have a
cluster named, “cluster1”, in a zone named, “pod1” inside a zone named,
“zone1”, I should be able to list it correctly, yes?
(local) 🐵 > list clusters
count = 1
cluster:
name = cluster1
id = d2e79c24-88e4-
).
>
>If you run raw API in say browser, with and without the name arg do you
>get the same result?
>
>> On 31-Dec-2014, at 5:28 am, Ian Forde wrote:
>>
>> Hi all -
>>
>>
>> I’ve just detected some more strange behavior in Cloudmonkey. If I
>>have a
rches for resource names
>matching a passed substring.
>
>> On 31-Dec-2014, at 3:04 pm, Ian Forde wrote:
>>
>> Note that “ster1” is a proper substring match with “cluster1”. I tested
>> it again on both 4.3.1 (RPMs on RHEL) and 4.4.2 (RPMs on CentOS) and got
>&g
(Note for anyone remotely interested in this later, it’s actually an array
returned, not a hash. My mistake.)
On 12/31/14, 2:28 AM, "Ian Forde" wrote:
>Hi Rohit,
>
>From a bash perspective (since I started down this path from cloudmonkey),
>I think that’s a very strange
es
>matching a passed substring.
>
>> On 31-Dec-2014, at 3:04 pm, Ian Forde wrote:
>>
>> Note that “ster1” is a proper substring match with “cluster1”. I tested
>> it again on both 4.3.1 (RPMs on RHEL) and 4.4.2 (RPMs on CentOS) and got
>> the same results.
>&g
in the APIs but since I
>did not implement those APIs I won't know any technical reason behind
>those decisions.
>
>On Friday 02 January 2015 10:54 AM, Ian Forde wrote:
>> Note - thinking about this some more, you may want to note that it
>>behaves
>> as you say f
Agreed. Very good article. I have a question though - is it really
appropriate to describe this as “self-healing”, when there are manual
steps to be performed in case of failure of the Pool Master?
Typically, I use this term to refer to situations where the master fails,
a different host gets el
As a followup to this (just thinking about it) I’m wondering if this is
the same issue that I reported a while back, where it’s not really
cloudmonkey that has the issue - but rather the API. In my situation, any
password with ‘<‘ or ‘>’ in it got url-encoded, causing the password check
to fail…
Question - Currently, options set in Cloudmonkey seem to persist into
~/.cloudmonkey/config. For example, if one uses cloudmonkey
interactively, and changes the display setting from json to table (without
putting the setting back to json when finished), any cron jobs that depend
on the output set
|
>Twitter:@cloudstackguru<https://twitter.com/#!/cloudstackguru>
>
>ShapeBlue Ltd, 53 Chandos Place, Covent Garden, London, WC2N
>4HS
>
>
>On 27 Feb 2015, at 23:20, Ian Forde
>mailto:ifo...@marketo.com>> wrote:
>
>Question - Currently, options set in Cloud
21 matches
Mail list logo
