|
Re: Diego IAAS settings for vsphere
Eric Malm <emalm@...>
Hi, Daya,
I'd start by looking at the auctioneer logs at `/var/vcap/sys/log/auctioneer/auctioneer.stdout.log` on the brain VMs and the BBS logs at `/var/vcap/sys/log/bbs/bbs.stdout.log` on the database VMs. You could also get the state response from each cell by curling ` http://CELL_IP:1800/state` <http://CELL_IP:1800/state> to see the resources it advertises to the auctioneer. I think CC by default configures staging tasks to use 6G of disk, so it's possible that none of the cells in the deployment have the remaining capacity to accept the work, especially if they're already running other workloads. Are you able to stage or run a buildpack-based app against the Diego backend? Thanks, Eric On Tue, Jan 26, 2016 at 4:00 PM, Daya Shetty <daya.shetty(a)bnymellon.com> wrote: Hi Eric,
|
|
|
|
Re: Cloud Foundry being used for an EU social learning games platform
Amit Kumar Gupta
See these StackOverflow questions about driving CF interactions via Python
toggle quoted messageShow quoted text
from the last ~1 month: * http://stackoverflow.com/questions/34527475/automate-cloudfoundry-deployment-with-python * http://stackoverflow.com/questions/34585547/how-to-use-cloudfoundry-client-python-module * http://stackoverflow.com/questions/34651566/upload-app-bits-in-cloudfoundry-with-python * http://stackoverflow.com/questions/34717166/how-to-list-all-apps-in-cloudfoundry-using-python * http://stackoverflow.com/questions/34719957/how-to-upload-bits-for-cloudfoundry-app-with-python-requests * http://stackoverflow.com/questions/34918216/cf-login-with-python Best, Amit
On Wed, Jan 27, 2016 at 12:40 AM, Danny Rosen <drosen(a)pivotal.io> wrote:
@Amit: Could you embellish on the use cases you've encountered that have
|
|
|
|
Re: Cloud Foundry being used for an EU social learning games platform
Danny Rosen
@Amit: Could you embellish on the use cases you've encountered that have
toggle quoted messageShow quoted text
attempted to drive interactions with CF via a python library? @Juan: This is interesting and I'd like to obtain a better understanding of the use cases this library supports that drove you to publish it. Thanks in advance for the insight, Danny
On Wed, Jan 27, 2016 at 3:12 AM, Amit Gupta <agupta(a)pivotal.io> wrote:
Juan, good stuff! I've seen some people struggling recently to drive
|
|
|
|
Re: Cloud Foundry being used for an EU social learning games platform
Amit Kumar Gupta
Juan, good stuff! I've seen some people struggling recently to drive
interactions with CF via a Python library, it would be cool if there were well-maintained client libraries in a bunch of different languages. It's great you've written one for Node. I don't know how the cloudfoundry-community GH org is administered, I'd ask Dr Nic (cc'd). Cheers, Amit On Tue, Jan 26, 2016 at 8:12 AM, Juan Antonio Breña Moral < bren(a)juanantonio.info> wrote: Good afternoon community,
|
|
|
|
push a python django app to cloudfoundry
Lynn Lin
Hi , we have a python django web app . In this web app, we will call VMware ovf tool (https://my.vmware.com/group/vmware/details?productId=491&downloadGroup=OVFTOOL410) .This tool is needed to be installed (sudo bash VMware-ovftool-4.1.0-2459827-lin.x86_64.bundle),How can we support this in CF V2?
|
|
|
|
Re: Add a new job/vm instance using Bosh
Paul Bakare
Ah yeah! Thanks
toggle quoted messageShow quoted text
On Wed, Jan 27, 2016 at 6:44 AM, Amit Gupta <agupta(a)pivotal.io> wrote:
Depends on what your manifest looks like, but if we take the example in
|
|
|
|
Re: Add a new job/vm instance using Bosh
Amit Kumar Gupta
Depends on what your manifest looks like, but if we take the example in
toggle quoted messageShow quoted text
cf-release: https://github.com/cloudfoundry/cf-release/blob/b1c839f8682a61542e2ab4e65865b3426ae65455/example_manifests/minimal-aws.yml#L266 Just change that line to whatever number you want in your manifest file, and then run "bosh -d PATH_TO_MANIFEST_FILE deploy"
On Tue, Jan 26, 2016 at 8:39 PM, Kayode Odeyemi <dreyemi(a)gmail.com> wrote:
Hi,
|
|
|
|
Add a new job/vm instance using Bosh
Paul Bakare
Hi,
How can I increase DEA instance using Bosh? Appreciate your help.
|
|
|
|
Re: Diego IAAS settings for vsphere
Daya Shetty <daya.shetty@...>
Hi Eric,
Thanks for your helpI I could conjure up an iass-settings for vsphere and was able to deploy diego successfully but was having the same access_z1 VM failing problem but was able to bring up diego after incorporating your comments on this thread . https://lists.cloudfoundry.org/archives/list/cf-dev(a)lists.cloudfoundry.org/thread/JDD3SEP7CV6ONGED3DFAFYN2I54JQSIT/ Trying to push a docker app and I’m getting cf push my-tcp-receiver -o cloudfoundry/tcp-sample-receiver Creating app my-tcp-receiver in org BNYMellon / space testspace as admin... OK Creating route my-tcp-receiver.poc-apps.bnymellon.net... OK Binding my-tcp-receiver.poc-apps.bnymellon.net to my-tcp-receiver... OK Starting app my-tcp-receiver in org BNYMellon / space testspace as admin... FAILED InsufficientResources I checked the stager.stdout.log on the cc-bridge and it shows the following: "timestamp":"1453833835.995078564","source":"stager","message":"stager.starting","log_level":1,"data":{}} {"timestamp":"1453833835.995591879","source":"stager","message":"stager.Listening for staging requests!","log_level":1,"data":{}} {"timestamp":"1453850050.902931690","source":"stager","message":"stager.docker.build-recipe.staging-request","log_level":1,"data":{"app-id":"6992baee-ef29-4805-92ce-acac5268b78e","session":"2.1","staging-guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6"}} {"timestamp":"1453850050.903894424","source":"stager","message":"stager.staging-handler.staging-request.desiring-task","log_level":1,"data":{"callback_url":"http://stager.service.cf.internal:8888/v1/staging/6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6/completed","session":"3.1","staging-guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6","task_guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6"}} {"timestamp":"1453850051.046576500","source":"stager","message":"stager.completion-handler.task-complete-callback-received.posting-staging-complete","log_level":1,"data":{"guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6","payload":"eyJlcnJvciI6eyJpZCI6Ikluc3VmZmljaWVudFJlc291cmNlcyIsIm1lc3NhZ2UiOiJpbnN1ZmZpY2llbnQgcmVzb3VyY2VzIn19","session":"4.1"}} {"timestamp":"1453850051.047005653","source":"stager","message":"stager.completion-handler.task-complete-callback-received.cc-client.delivering-staging-response","log_level":1,"data":{"guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6","payload":"{\"error\":{\"id\":\"InsufficientResources\",\"message\":\"insufficient resources\"}}","session":"4.1.1"}} {"timestamp":"1453850051.143590689","source":"stager","message":"stager.completion-handler.task-complete-callback-received.cc-client.delivered-staging-response","log_level":1,"data":{"guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6","session":"4.1.1"}} {"timestamp":"1453850051.143915415","source":"stager","message":"stager.completion-handler.task-complete-callback-received.posted-staging-complete","log_level":1,"data":{"guid":"6992baee-ef29-4805-92ce-acac5268b78e-6e72ac5779384f68a04a245ade7184b6","session":"4.1”}} I have the following releases deployed: root(a)rsomtapae181 cf-release]# bosh releases Acting as user 'admin' on 'bosh2' +----------------------------+------------+-------------+ | Name | Versions | Commit Hash | +----------------------------+------------+-------------+ | cf | 222+dev.1* | e4eb9f4b+ | | diego | 0.1437.0* | 7a972628 | | etcd | 16* | 793d1c6b | | garden-linux | 0.308.0* | b5eced17 | +----------------------------+------------+-------------+ (*) Currently deployed (+) Uncommitted changes Any ideas on where I could look to debug this issue? Thanks Daya
|
|
|
|
Re: [abacus] MongoDB client
Jean-Sebastien Delfino
Hi Hristo,
toggle quoted messageShow quoted text
This looks really good! +1 to merging this as a new module in master, and continuing the work there, as then we can all work together on creating the abstraction layer that'll select between the Couch/Pouch/Mongo clients, and also on making the in-memory caches select the proper backend like I think you were suggesting the other day on our slack channel. Once your module is there in master, I can also help add a build config option to have some of the CI builds run the tests with it, if that helps. Thanks! - Jean-Sebastien - Jean-Sebastien
On Sun, Jan 24, 2016 at 1:33 AM, Hristo Iliev <hsiliev(a)gmail.com> wrote:
Hi,
|
|
|
|
Re: Cloud Foundry being used for an EU social learning games platform
Juan Antonio Breña Moral <bren at juanantonio.info...>
Good afternoon community,
after some months of, we have developed a stable client for cloud foundry: https://github.com/prosociallearnEU/cf-nodejs-client https://travis-ci.org/prosociallearnEU/cf-nodejs-client Docs: https://prosociallearneu.github.io/cf-nodejs-client/docs/v0.13.0/ I would like to explore the idea to share the development with the community and how to integrate the development on the repo: https://github.com/cloudfoundry-community/ Some useful developments are: https://github.com/prosociallearnEU/cf-nodejs-dashboard https://github.com/jthomas/cfbot Cheers
|
|
|
|
Re: *.cf.internal on diego windows
Matthew Horan
Edward -
toggle quoted messageShow quoted text
We used to detect the routable interface as you propose, but ran into issues with that approach. After testing, we determined that Windows will query to the first DNS server on every interface, regardless of whether or not that interface has a "routable" IP. We swapped out our approach as this was a more reliable way to configure DNS. In the future, Windows VMs will be managed by bosh for Windows, and this will be a non issue. For now, operators with more complicated setups can add the Consul DNS server to the appropriate interface manually, as you discovered. Amit - The Greenhouse team manually syncs the Bosh jobs with our compiled Windows services. We have a build in place which monitors the job scripts and fails when there are changes. I'm not sure if we're currently monitoring the consul_agent job, but perhaps we should be. Regardless, Consul does seem to be running properly at this time, it's just the DNS resolution interface config that gets out of sync.
On Mon, Jan 25, 2016 at 8:11 PM, Amit Gupta <agupta(a)pivotal.io> wrote:
Not too sure about how the Windows Cell gets provisioned, but the
|
|
|
|
Re: Cannot start access_z1
Stanley Shen <meteorping@...>
Thanks Eric for detail explanation and it works for me.
|
|
|
|
Re: Cannot start access_z1
Eric Malm <emalm@...>
Hi, Stanley,
toggle quoted messageShow quoted text
I think you're missing the 'ssh_proxy.uaa_secret' value in your manifest. This should be the same value as the secret in the 'ssh-proxy' client that's registered with UAA. The Diego team just finished a story ( https://www.pivotaltracker.com/story/show/109642196) to make this configuration more transparent, and we'll be doing one shortly ( https://www.pivotaltracker.com/story/show/107970964) to update the SSH configuration docs to explain the UAA configuration that's required. For now, it should suffice for you to add a client like the one at https://github.com/cloudfoundry/cf-release/blob/master/bosh-lite/stubs/enable-diego-ssh.yml#L11-L17 to the list of UAA clients in your CF manifest, if you haven't done so already. Thanks, Eric, CF Runtime Diego PM
On Mon, Jan 25, 2016 at 11:24 PM, Stanley Shen <meteorping(a)gmail.com> wrote:
Hello, all
|
|
|
|
Cannot start access_z1
Stanley Shen <meteorping@...>
Hello, all
I am trying to deploy cf+diego to AWS based on the minimal-aws.yml provided in cf-release project. And I add some jobs for diego. Right now the access_z1 is not started up and here are the errors: ==================================== head ssh_proxy.stderr.log panic: uaaTokenURL is required for Cloud Foundry authentication goroutine 1 [running]: github.com/pivotal-golang/lager.(*logger).Fatal(0xc20805c120, 0xa162b0, 0x10, 0x7f5b697e8b40, 0xc20802aab0, 0x0, 0x0, 0x0) /var/vcap/packages/ssh_proxy/src/github.com/pivotal-golang/lager/logger.go:152 +0x5d0 main.configureProxy(0x7f5b697ed770, 0xc20805c120, 0xc2080a5ef0, 0x0, 0x0) /var/vcap/packages/ssh_proxy/src/github.com/cloudfoundry-incubator/diego-ssh/cmd/ssh-proxy/main.go:205 +0x9b7 main.main() /var/vcap/packages/ssh_proxy/src/github.com/cloudfoundry-incubator/diego-ssh/cmd/ssh-proxy/main.go:137 +0x1ab ============================= head ssh_proxy.stdout.log {"timestamp":"1453790308.941943645","source":"ssh-proxy","message":"ssh-proxy.uaa-url-required","log_level":3,"data":{"error":"uaaTokenURL is required for Cloud Foundry authentication","trace":"goroutine 1 [running]:\ngithub.com/pivotal-golang/lager.(*logger).Fatal(0xc20805c120, 0xa162b0, 0x10, 0x7f5b697e8b40, 0xc20802aab0, 0x0, 0x0, 0x0)\n\t/var/vcap/packages/ssh_proxy/src/github.com/pivotal-golang/lager/logger.go:131 +0xc8\nmain.configureProxy(0x7f5b697ed770, 0xc20805c120, 0xc2080a5ef0, 0x0, 0x0)\n\t/var/vcap/packages/ssh_proxy/src/github.com/cloudfoundry-incubator/diego-ssh/cmd/ssh-proxy/main.go:205 +0x9b7\nmain.main()\n\t/var/vcap/packages/ssh_proxy/src/github.com/cloudfoundry-incubator/diego-ssh/cmd/ssh-proxy/main.go:137 +0x1ab\n"}} From the error message it's about the uaa_token_url is not provided. But I did provided it in deployment like this: ===================================== ssh_proxy: bbs: api_location: bbs.service.cf.internal:8889 ca_cert: |+ -----BEGIN CERTIFICATE----- -----END CERTIFICATE----- client_cert: |+ -----END CERTIFICATE----- client_key: |+ -----BEGIN RSA PRIVATE KEY----- -----END RSA PRIVATE KEY----- client_session_cache_size: null max_idle_conns_per_host: null require_ssl: false diego_credentials: null enable_cf_auth: true enable_diego_auth: false host_key: |+ -----BEGIN RSA PRIVATE KEY----- -----END RSA PRIVATE KEY----- servers: - 10.0.16.106 uaa_secret: null uaa_token_url: https://uaa.MyDomain/oauth/token ssl: skip_cert_verify: true And my env is: ================================ +--------------+-----------------+-------------+ | Name | Versions | Commit Hash | +--------------+-----------------+-------------+ | cf | 226+dev.1* | 5de34b6a | | diego | 0.1442.0+dev.1* | 375c6064 | | etcd | 18* | 52f3a004+ | | garden-linux | 0.328.0* | ce51a708 | +--------------+-----------------+-------------+ +-----------------------------------------+---------------+---------+--------------------+ | Name | OS | Version | CID | +-----------------------------------------+---------------+---------+--------------------+ | bosh-aws-xen-hvm-ubuntu-trusty-go_agent | ubuntu-trusty | 3177* | ami-7eafb41f light | +-----------------------------------------+---------------+---------+--------------------+ Can someone help on it, do I missed any configuration to make access running up?
|
|
|
|
Re: *.cf.internal on diego windows
Amit Kumar Gupta
Not too sure about how the Windows Cell gets provisioned, but the
toggle quoted messageShow quoted text
consul_agent job in consul-release has been undergoing a lot of changes recently in terms of how the startup scripts work. How does the Windows Cell project keep up to date with changes to https://github.com/cloudfoundry-incubator/consul-release/blob/master/jobs/consul_agent/templates/agent_ctl.sh.erb? And is it possible that there's some logic that is now out of sync?
On Mon, Jan 25, 2016 at 2:28 PM, Edward Hill <hill_edd(a)yahoo.co.uk> wrote:
Think the test should be:
|
|
|
|
Re: *.cf.internal on diego windows
Edward Hill
Think the test should be:
TestScript = { [array]$routeable_interfaces = Get-WmiObject Win32_NetworkAdapterConfiguration | Where { $_.IpAddress -AND ($_.IpAddress | Where { $addr = [Net.IPAddress] $_; $addr.AddressFamily -eq "InterNetwork" -AND ($addr.address -BAND ([Net.IPAddress] "255.255.0.0").address) -ne ([Net.IPAddress] "169.254.0.0").address }) } foreach($routeable_interface in $routeable_interfaces) { $interface = (Get-WmiObject Win32_NetworkAdapter | Where { $_.DeviceID -eq $routeable_interface.Index }).netconnectionid if(-Not (Get-DnsClientServerAddress -InterfaceAlias $interface -AddressFamily ipv4 -ErrorAction Stop).ServerAddresses[0] -eq "127.0.0.1") { Write-Verbose -Message "DNS Servers not yet correct." return $false } } Write-Verbose -Message "DNS Servers are set correctly." return $true }
|
|
|
|
Re: *.cf.internal on diego windows
Edward Hill
I had run the setup.ps1 but I had a lot of indecision about networking and reduced the number of adapters later which might have caused it
I notice the setup.ps1 selects the first interface matched: $ifindex = $routeable_interfaces[0].Index
|
|
|
|
Re: *.cf.internal on diego windows
Matthew Horan
That's it. localhost needs to be the first DNS server in the list for any
toggle quoted messageShow quoted text
interface on the host. Otherwise, name resolution routines won't be able to lookup names in Consul. I'd be interested to hear why the DNS settings got cleared out in your environment. Or, had you not run the setup.ps1 script in the first case? Thanks!
On Sun, Jan 24, 2016 at 4:17 PM, Edward Hill <hill_edd(a)yahoo.co.uk> wrote:
setting had gone
|
|
|
|
Re: HA Docker cf services for MongoDB, Redis, &etc?
Johannes Hiemer <jvhiemer@...>
For all except commercial it is. :-)
toggle quoted messageShow quoted text
On Mon, Jan 25, 2016 at 2:05 PM, Samuel Marks <samuelmarks(a)gmail.com> wrote:
Oh, I should be clearer. I mean open-source for research, commercial, --
Mit freundlichen Grüßen Johannes Hiemer
|
|
|