Re: Dotfile boshrelease
Dr Nic Williams
I enjoy seeing more innovation in this area.
toggle quoted message
Show quoted text
As a bonus, there is also a `curl | bash` solution that I like to use as it requires zero setup at deploy time: https://github.com/cloudfoundry-community/bosh-root-env On Fri, Sep 18, 2015 at 1:03 AM, Ed King <ed(a)cloudcredo.com> wrote:
Awesome idea! Thanks for this. --
Dr Nic Williams Stark & Wayne LLC - consultancy for Cloud Foundry users http://drnicwilliams.com http://starkandwayne.com cell +1 (415) 860-2185 twitter @drnic |
|
Upload of admin ui new boshrelease to S3
Klevenz, Stephan <stephan.klevenz@...>
Hi,
The admin-ui boshrelease has created a new version (v6) which can be manual downloaded from here: [1] It would be nice to have this release also available in S3 as documented here [2]. I don't have credentials and want ask someone else for doing this. Thanks and Regards, Stephan [1] https://bosh.io/d/github.com/cloudfoundry-community/admin-ui-boshrelease [2] https://github.com/cloudfoundry-community/admin-ui-boshrelease#create-new-final-release |
|
Re: Dotfile boshrelease
Ed
Awesome idea! Thanks for this.
|
|
Re: Finding one vm from another in a single deployment
Craig Rubendall
Thanks for the information.
I'm assuming that I can use a "VIP" bosh network definition to solve this, then put the ip address in as the property. It was relatively easy to define this, but I can't seem to get the security groups correct on AWS to have my client VM see the server VM via the elastic ip address. Has anyone done this? It seems so straightforward, so I have no idea what I'm doing wrong. My deployment manifest looks like this: networks: - name: default type: dynamic cloud_properties: subnet: subnet-xxxxxxx - name: elasticip type: vip cloud_properties: {} I've also tried adding explicit security groups to the cloud_properties for the vip network, and i've tried adding the subnet. Then my job area has in it: networks: - name: elasticip static_ips: [52.xx.xx.xx] - name: default default: [dns, gateway] Any ideas on what I'm doing incorrectly? Or am I trying to do something that won't work? I've used the Terraform "tiny" script to set up my deployment of CF (if that in any way gives a clue of what I need to do). |
|
Getting "Task exited with status failed" while performing bosh upload release
Dermot Burke
Hi,
I am getting the following output and error when attempting a bosh upload release on AWS bosh upload release ~/cf-deployment/cf-release/releases/cf-217.yml Acting as user 'admin' on 'microbosh' You are using CLI >= 0.20 with director that doesn't support package matches. This will result in uploading all packages and jobs to your director. It is recommended to update your director or downgrade your CLI to 0.19.6 Verifying release... File exists and readable OK ... Monit file for 'gorouter' OK Release info ------------ Name: cf Version: 217 Packages - dea_logging_agent (5566742613be76062999af05a84e8bc06abd976f) ... - buildpack_java (82fb7b325de57573e9ebddadf004b9dac5b49185) - buildpack_python (f7f81defd518e5c9f29aefe9c8957cb2fbcd5028) Jobs - dea_logging_agent (726844edf31c1636e956ad1a270ec3f35b631615) ... - gorouter (6e6eb6e59d8a5c14c2b99b1c5327be716cbbe32c) License - license (861d82c0d9745784acfcd026ccf44e942579824a) You are using CLI >= 0.20 with director that doesn't support package matches. This will result in uploading all packages and jobs to your director. It is recommended to update your director or downgrade your CLI to 0.19.6 Checking if can repack release for faster upload... dea_logging_agent (5566742613be76062999af05a84e8bc06abd976f) UPLOAD ... Uploading the whole release Uploading release [WARNING] cannot access director, trying 4 more times...oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo | 1.8GB 32.0MB/s ETA: 00:00:17 [WARNING] cannot access director, trying 3 more times...oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo| 2.4GB 19.7MB/s ETA: 00:00:00 cf-217.tgz: 100% |ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo| 2.4GB 17.2MB/s Time: 00:02:21 Task exited with status failed There seems to be nothing helpful in any of the error logs. I am using BOSH 1.3071.0. Any help would be gratefully appreciated. D |
|
Dotfile boshrelease
Ronak Banka
Hello ,
We are open sourcing one of our bosh release : Dotfile bosh release https://github.com/rakutentech/dotfiles-boshrelease Inspired by cloudfoundry-community/root-env-boshrelease, dotfiles-boshrelease aims to make a better admin experience for bosh ssh, which is crucial for both dev & ops on bosh. https://github.com/rakutentech/dotfiles-boshrelease#what-is-done-by-default Thanks Ronak Banka Rakuten, Inc. -- View this message in context: http://cf-bosh.70367.x6.nabble.com/Dotfile-boshrelease-tp725.html Sent from the CF BOSH mailing list archive at Nabble.com. |
|
Re: Rendering template error Bosh-init
Dmitriy Kalinin
We currently have a problem rendering templates with default ruby installed - 1.9.x, because json gem changed some behavior. If you upgrade your ruby issue should go away.
toggle quoted message
Show quoted text
Sent from my iPhone On Sep 15, 2015, at 7:14 PM, Jamie OMeara <jomeara(a)pivotal.io> wrote: |
|
Rendering template error Bosh-init
Jamie OMeara
Trying to Bosh-init install and getting the following error rendering the
director.yml.erb.erb file. What is the best way to troubleshoot rendering issues? Error filling in template '/Users/jomeara/.bosh_init/installations/931295f1-e028-4bf2-59e4-74f7dc1565ca/tmp/bosh-init-release335269600/extracted_jobs/director/templates/director.yml.erb.erb' for director/0 (line 134: #<JSON::GeneratorError: only generation of JSON objects or arrays allowed>) (RuntimeError) *Jamie O'Meara* Field Engineer 303.898.5437 jomeara(a)pivotal.io www.pivotal.io |
|
Re: automatically registering gorouters with OpenStack LB
Koper, Dies <diesk@...>
Hi Ronak,
Thanks for your reply. I’m not sure what API we’re currently using and if not LBaaS, why not. Either way, it sounds like the collocated job is the easiest solution for now. Cheers, Dies Koper From: ronak banka [mailto:ronakbanka.cse(a)gmail.com] Sent: Tuesday, September 15, 2015 4:02 PM To: Discussions about the Cloud Foundry BOSH project. Subject: [cf-bosh] Re: Re: automatically registering gorouters with OpenStack LB Hello, I Agree that it will be helpful for Openstack deployments but having something like this in CPI will only be helpful when native Openstack LBaaS api are used . For custom ELB deployments it will be good to create separate Bosh release and make a elb-register job collocated along with router job which will add or remove router from ELB using api or else way. Regards, Ronak Banka Rakuten, Inc. On Tue, Sep 15, 2015 at 3:37 PM, Voelz, Marco <marco.voelz(a)sap.com<mailto:marco.voelz(a)sap.com>> wrote: Hi Dies, I agree that the registration at an external ELB would be very helpful to have – just like it is done for AWS. We currently use the HAProxy, so I guess this is a challenge which will hit us at a later point in time… I’d love to see a PR for this! :) Warm regards Marco On 15/09/15 01:54, "Koper, Dies" <diesk(a)fast.au.fujitsu.com<mailto:diesk(a)fast.au.fujitsu.com>> wrote: Hi, I’m aware of bosh aws cpi’s elbs option to list ELBs with gorouters. I’m looking for a similar option for OpenStack. How do people ensure that when gorouters are e.g. rolled during an upgrade, the new VMs are automatically registered with the load balancer? We use a LB we developed ourselves, which is hooked in to OpenStack using its pluggable LB infrastructure. Cheers, Dies Koper |
|
Re: BOSH/CF package compilation time
Dmitriy Kalinin
We have fixed an issue with director->nginx communication on Director VM. (
https://github.com/cloudfoundry/bosh/commit/3e82ae48555c0f8c4f01ce339ca864d0cd85894f ) It's available in bosh release 202+ / stemcell 3067+. On Tue, Sep 15, 2015 at 1:51 PM, Satya Thokachichu <tsnraju(a)yahoo.com> wrote: While doing bosh export,I kept on getting 502 after copying packages |
|
Re: BOSH/CF package compilation time
Satya Thokachichu
While doing bosh export,I kept on getting 502 after copying packages ..Systemlevel resources looks good..Tried to locate compiled packages under /var/vcap/data and /var/vcap/store..Once it fails,I cant find any data..Did anybody encounter this issue ? Any help would be greatly appreciated
ubuntu(a)ip-10-0-0-146:~$bosh export release cf/214 ubuntu-trusty/3048 Director task 49 Started unknown Started unknown > Binding deployment. Done (00:00:00) ...... Started preparing package compilation > Finding packages to compile. Done (00:00:00) Started 2015-09-15 16:07:02 UTC Finished 2015-09-15 16:09:26 UTC Duration 00:02:24 release-cf-214-on-ubuntu-trusty-stemcell-... downloading...Cannot download resource `96c943ac-3891-4b32-9fc3-68a7db34eac5': HTTP status 502 |
|
Re: How to upgrade my bosh?
Alex Lomoff <lomov.as@...>
Thank you for correct me.
toggle quoted message
Show quoted text
Shame on me, I meant bosh-init instead of bosh-lite. On Sep 15, 2015, at 5:09 PM, Amit Gupta <agupta(a)pivotal.io> wrote: |
|
Re: How to upgrade my bosh?
Amit Kumar Gupta
Hi Alex,
toggle quoted message
Show quoted text
That's not correct. bosh-init is an executable binary, written in golang, used to deploy single-VM deployments of anything you can colocate onto one VM. Using it to deploy BOSH itself is the most common use case, so you could call that Micro BOSH. *bosh-init* is different from *BOSH-Lite*. BOSH-Lite is a project for producing a vagrant VM running the BOSH director along with garden, which creates containers which simulates VMs (the BOSH director in the BOSH-Lite vagrant box talks to garden via the warden-cpi to create "VMs", which in this care are just garden containers). daxiezhi, If you are using BOSH-Lite, I believe there is no upgrade path, as it is only meant for development purposes. From the "bosh-lite" directory, you will need to: vagrant box update vagrant destroy vagrant up See http://stackoverflow.com/questions/25914383/update-a-vagrant-box. You will then need to re-provision everything, i.e. reupload the stemcell, release, and do a "bosh deploy" with your cf manifest. Best, Amit On Tue, Sep 15, 2015 at 9:56 AM, Alex Lomoff <lomov.as(a)gmail.com> wrote:
Sorry, I guess some misunderstanding took place. |
|
Re: How to upgrade my bosh?
Alex Lomoff <lomov.as@...>
@daxiezhi in case you are talking about BOSH-director update you should described the way you deployed BOSH. Did you used `bosh micro` command to deploy MicroBOSH? Or did you used bosh-lite?
toggle quoted message
Show quoted text
If you used the first option then here a way to update you MicroBOSH instance. MicroBOSH is built-in stemcell, originally stemcell version is a version of MicroBOSH release that it contains. This means first you'll need to do is upload newer version of stemcell. After it I recommend to make a snapshot of the MicroBOSH instance with its persistent disk. Then you can run update process with the following command: bosh -n micro deploy bosh-stemcell-xxxx-openstack-kvm-ubuntu-trusty-go_agent.tgz --update Best Wishes, Alex L. On Sep 15, 2015, at 4:23 PM, Amit Gupta <agupta(a)pivotal.io> wrote: |
|
Re: How to upgrade my bosh?
Alex Lomoff <lomov.as@...>
Sorry, I guess some misunderstanding took place.
toggle quoted message
Show quoted text
Approximately the same number you’ll get if you run `gem list bosh_cli` because bosh-director has the same version numbering as a bosh_cli, as far they are parts of the same project of cloudfoundry/bosh repo. So I thought @daxiezhi probably wants to update CLI through gem. As far as I know bosh-lite is a go utility to deploy MicroBOSH. It aims to replace BOSH CLI plugin “micro", the outdated way to deploy MicrBOSH. Do you mean that at this moment MicroBOSH deployed with bosh-lite should be called BOSH-Lite? On Sep 15, 2015, at 4:23 PM, Amit Gupta <agupta(a)pivotal.io> wrote: |
|
Re: How to upgrade my bosh?
Amit Kumar Gupta
For me, `bosh -t lite status` shows "Version 1.3056.0 (00000000)". I
toggle quoted message
Show quoted text
believe that's the version of the bosh-director gem: https://rubygems.org/gems/bosh-director/versions/ On Tue, Sep 15, 2015 at 4:34 AM, Alex Lomoff <lomov.as(a)gmail.com> wrote:
Hi, all. |
|
Re: How to upgrade my bosh?
Alex Lomoff <lomov.as@...>
Hi, all.
toggle quoted message
Show quoted text
Amit, I guess it is not bosh-lite version, because bosh-lite versions are over 9000 [1]. daxiezhi, if you mean by “local bosh” your version of bosh_cli gem, then everything you need to do is to run: “gem update bosh_cli” (possibly with sudo). If you mean you need to update BOSH director, then everything depend on the you deployed the BOSH. In this case could you tell what commands did you used to deploy BOSH? [1] https://github.com/cloudfoundry/bosh-lite/releases <https://github.com/cloudfoundry/bosh-lite/releases> Thank you, Alex L. On Sep 14, 2015, at 7:00 PM, Amit Gupta <agupta(a)pivotal.io> wrote: |
|
Re: automatically registering gorouters with OpenStack LB
Ronak Banka
Hello,
toggle quoted message
Show quoted text
I Agree that it will be helpful for Openstack deployments but having something like this in CPI will only be helpful when native Openstack LBaaS api are used . For custom ELB deployments it will be good to create separate Bosh release and make a elb-register job collocated along with router job which will add or remove router from ELB using api or else way. Regards, Ronak Banka Rakuten, Inc. On Tue, Sep 15, 2015 at 3:37 PM, Voelz, Marco <marco.voelz(a)sap.com> wrote:
Hi Dies, |
|
Re: automatically registering gorouters with OpenStack LB
Marco Voelz
Hi Dies,
toggle quoted message
Show quoted text
I agree that the registration at an external ELB would be very helpful to have – just like it is done for AWS. We currently use the HAProxy, so I guess this is a challenge which will hit us at a later point in time… I’d love to see a PR for this! :) Warm regards Marco On 15/09/15 01:54, "Koper, Dies" <diesk(a)fast.au.fujitsu.com<mailto:diesk(a)fast.au.fujitsu.com>> wrote:
Hi, I’m aware of bosh aws cpi’s elbs option to list ELBs with gorouters. I’m looking for a similar option for OpenStack. How do people ensure that when gorouters are e.g. rolled during an upgrade, the new VMs are automatically registered with the load balancer? We use a LB we developed ourselves, which is hooked in to OpenStack using its pluggable LB infrastructure. Cheers, Dies Koper |
|
automatically registering gorouters with OpenStack LB
Koper, Dies <diesk@...>
Hi,
I'm aware of bosh aws cpi's elbs option to list ELBs with gorouters. I'm looking for a similar option for OpenStack. How do people ensure that when gorouters are e.g. rolled during an upgrade, the new VMs are automatically registered with the load balancer? We use a LB we developed ourselves, which is hooked in to OpenStack using its pluggable LB infrastructure. Cheers, Dies Koper |
|