asterixdb-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ian Maxon <ima...@uci.edu>
Subject Re: AsterixDB command line access
Date Tue, 08 Sep 2015 20:04:06 GMT
I'm certain more than one person has created a command-line utility at
some point (I think Chris had one at some point? There's one lurking
in asterix-tools, not sure if it's that one, or that is yet another
one). The HTTP API though isn't so complex (hopefully?) as to really
merit a special client in most cases.

If one just has a shell account, and port forwarding is available,
Till's suggestion is definitely a good one, and usually what I do.
However if for some reason that isn't tenable, 'curl' should also do
the trick. For example:

curl --data 'use dataverse issues;
{
"projects" : [ {
"name" : "asterixdb",
"key": "ASTERIXDB",
"description": "AsterixDB is a semi-structured, parallel data
management platform.",
"components": ["AsterixDB","Hyracks"],
"issues": for $y in dataset jiraIssues return $y
} ]
}' -H "Accept: application/x-adm" http://localhost:19002/aql >
/files/adb_jira_issues.json

Just replace the --data part with your query, and the host with the
proper one that's behind the firewall. I'm piping the output to a file
here, but stdout is fine too.

-Ian

On Tue, Sep 8, 2015 at 2:28 AM,  <schultze@informatik.hu-berlin.de> wrote:
> Hi Till,
>
> I probably have to do so. It would have just been easier to later have
> direct command line access for automatic execution of some queries I want
> to run to compare the two systems.
>
> @Mike I am directly translating the AQL Queries to Flink's Scala API.
>
> Regards, Max
>
>> Hi Max,
>>
>> If you have ssh-access to the cluster you could also tunnel the web
>> interface port (usually 19001) on the cluster machine that hosts
>> AsterixDB's cluster controller to your machine using ssh.
>> Of course that's only an option if the web interface works for you
>> otherwise.
>>
>> Cheers,
>> Till
>>
>> On 7 Sep 2015, at 16:09, Mike Carey wrote:
>>
>>> We have an HTTP-based API you could use - with curl - not a great
>>> answer - broadening the discussion to see if anyone on the dev list
>>> can offer you what they use.  (We need to add a nicer command line
>>> interface at some point.)
>>>
>>> Random side comment:  Hopefully you are doing this by translating from
>>> Algebricks physical plans to the Flink runtime?  (That would be the
>>> cleanest place to redirect queries, I would think, to be against Flink
>>> instead of against Hyracks.)
>>>
>>> Cheers,
>>> Mike
>>>
>>>
>>> On 9/7/15 2:03 AM, schultze@informatik.hu-berlin.de wrote:
>>>> Hello AsterixDB group,
>>>>
>>>> I am a student currently working on a diploma thesis to execute AQL
>>>> on
>>>> Apache Flink. Therefor I did setup AsterixDB on a cluster.
>>>>
>>>> My problem is, that I am not in the same network as the cluster and
>>>> only
>>>> have remote command line access, so I cannot reach Asterix's web
>>>> interface. I am sure it is possible to run queries via command line,
>>>> too,
>>>> but I could not find anything on that topic on your official
>>>> documentation
>>>> page. Do you have any document you can link me to, that provides a
>>>> basic
>>>> explanation how to work with AsterixDB without the web interface?
>>>>
>>>> Best Regards,
>>>> Max Schultze
>>>>
>>>>
>>>>
>>>>
>>
>
>

Mime
View raw message