Closed
Bug 770378
Opened 13 years ago
Closed 13 years ago
New Elasticsearch index needed for a new project (x-tag-registry)
Categories
(Infrastructure & Operations Graveyard :: WebOps: Labs, task)
Tracking
(Not tracked)
RESOLVED
FIXED
People
(Reporter: aschaar, Unassigned)
Details
Dev Tools is creating a tag registry website for the x-tag project that is being spearheaded by Daniel Buchner.
We need a new Elasticsearch index so we can search through the tags and other metadata.
Currently the app is written in Nodejs and is hosted on Petri, so we need the ES index to be accessible from there.
Thanks,
Comment 1•13 years ago
|
||
(In reply to Arron Schaar [:arron] from comment #0)
> Dev Tools is creating a tag registry website for the x-tag project that is
> being spearheaded by Daniel Buchner.
>
> We need a new Elasticsearch index so we can search through the tags and
> other metadata.
>
> Currently the app is written in Nodejs and is hosted on Petri, so we need
> the ES index to be accessible from there.
>
> Thanks,
Hello :arron,
Since this is hosted on Petri, you'll need to see if :gozer currently has an ES cluster for Labs services or if he can provide you a VM to run ES in
Thanks!
Assignee: server-ops-webops → server-ops-labs
Component: Server Operations: Web Operations → Server Operations: Labs
QA Contact: cshields → zandr
Reporter | ||
Comment 2•13 years ago
|
||
Hello :gozer,
Can we get an ES index for a project on Petri?
Thanks
Comment 3•13 years ago
|
||
We have an app that is basically ready to go and getting an ES link to our project on Petri is our last blocker, what's the quickest way forward on this?
Comment 4•13 years ago
|
||
There is no shared ElasticSearch for Labs, nor is it offered as a cloudfoundry service.
But, you can deploy your own elasticsarch instance *in* petri yourself.
See the sample instructions here:
https://github.com/cloudfoundry-samples/rails-elastic-search#setting-up-elastic-search
Updated•13 years ago
|
Group: infra
Reporter | ||
Comment 5•13 years ago
|
||
I tried installing it from the example, but there isn't a vmc push --manifest='' command. I opened a stackoverflow question on it.
http://stackoverflow.com/questions/11319625/how-to-run-elasticsearch-on-cloudfoundry
Comment 6•13 years ago
|
||
I gave this a whirl myself, and you need vmc 0.3.18 to get --manifest, but the support for manifest and the support for the 'standalone' application type both appear to be newer than the CF code we're running on the server side.
Reporter | ||
Comment 7•13 years ago
|
||
Looks like they just updated vmc on July 3. I had 0.3.15. I'll give this another try.
Comment 8•13 years ago
|
||
(In reply to Arron Schaar [:arron] from comment #7)
> Looks like they just updated vmc on July 3. I had 0.3.15. I'll give this
> another try.
Don't upgrade your VMC, as it won't be compatible with our CF deployment yet, it will try to make API calls we don't support
I was only confirming the --manifest was in a newer version
Comment 9•13 years ago
|
||
Sooo, can we get our CF instance updated? We're ready to launch and this is a dead-in-the-water show stopper.
Comment 10•13 years ago
|
||
(In reply to Daniel Buchner [:dbuc] from comment #9)
> Sooo, can we get our CF instance updated?
It's a planned upgrade, but won't be hapenning anytime soon, it's a complex upgrade.
> We're ready to launch and this is
> a dead-in-the-water show stopper.
I will spin up a separate elastic search instance in labs that your CF app will be able to talk to. Hold on.
Comment 11•13 years ago
|
||
(In reply to Daniel Buchner [:dbuc] from comment #9)
> Sooo, can we get our CF instance updated? We're ready to launch and this is
> a dead-in-the-water show stopper.
For future reference, don't hesitate to bring me in early on projects like this and I could have helped avoid this last minute, hrm, excitement ;-)
Comment 12•13 years ago
|
||
Also -- cloudfoundry isn't really in a production environment -- you should check w/ gozer to make sure that labs is the environment you need.
Comment 13•13 years ago
|
||
We were given the go-ahead to use Petri for this project, all we need is this ES instance. This is not a user data sensitive app, or something we expect more than tens of thousands hits a month on, so I feel good about the choice.
:gozer what do you think the ETA is for the ES instance you are setting up?
Comment 14•13 years ago
|
||
(In reply to Daniel Buchner [:dbuc] from comment #13)
> We were given the go-ahead to use Petri for this project,
Just curious, where did that come from ?
> :gozer what do you think the ETA is for the ES instance you are setting up?
Minutes ?
Comment 15•13 years ago
|
||
:gozer, you are my hero buddy.
Comment 16•13 years ago
|
||
You should now be able to reach :
http://elasticsearch1.vm.labs.scl3.mozilla.com:9200/xtag
Please play nice with it, and limit yourself to the /xtag prefix, thank you.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
Updated•11 years ago
|
Product: mozilla.org → Infrastructure & Operations
Updated•8 years ago
|
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in
before you can comment on or make changes to this bug.
Description
•