|
Re: App failed to startup or crashed
Jason Huang
Sorry my previous email was not meant for the group. It was for Phong only.
:-). But since I am on it, let me gave more details on what we have seen before. What we encountered before was that the upgrade of our own applications in CF took very long time due to insufficient entropy in the client vm where our upgrade tool was run. We later installed Haveged package in the client VM to solve the problem. Jason On Tue, Dec 13, 2016 at 7:49 AM, Jason Huang <jasonxs.huang(a)gmail.com> wrote: I see a reply saying that it is likely because there is not enough |
|
|
|
Re: How to create user provided service step by step
DHR
A user-provided service in CF is simply a pointer, with credentials, to some infrastructure that is itself managed outside of CF.
toggle quoted message
Show quoted text
This pointer can be created using the "CF cups" command. If you're having trouble running it, you could run "cf cups --help" which prints the required command arguments. Or of this still does not make sense, Perhaps you could describe your use case & we will try to help that way On 13 Dec 2016, at 16:09, john J <johnjerrby(a)yahoo.com> wrote: |
|
|
|
Re: How to create user provided service step by step
john J
Btw this guide is too abstract and doesnt help much....
https://docs.cloudfoundry.org/devguide/services/user-provided.html |
|
|
|
How to create user provided service step by step
john J
HI,
I need some guide how to create user provided service step by step, there is someting like this? Thanks John |
|
|
|
Re: App failed to startup or crashed
Jason Huang
I see a reply saying that it is likely because there is not enough entropy. We had the problem before.
toggle quoted message
Show quoted text
Likely the problem is not enough entropy in /dev/random. You can try to use /dev/urandom instead to check if that helps: https://github.com/cloudfoundry/java-buildpack/blob/master/docs/container-tomcat.md (at the bottom of the page). Sent from my iPhone On Dec 13, 2016, at 4:35 AM, Michal Kuratczyk <mkuratczyk(a)pivotal.io> wrote: |
|
|
|
Re: App failed to startup or crashed
Michal Kuratczyk
Hi Phong,
toggle quoted message
Show quoted text
Likely the problem is not enough entropy in /dev/random. You can try to use /dev/urandom instead to check if that helps: https://github.com/cloudfoundry/java-buildpack/blob/master/docs/container-tomcat.md (at the bottom of the page). Best regards, On Tue, Dec 13, 2016 at 3:41 AM, Phong Tran <phong.tran(a)emc.com> wrote:
Recently we just just consolidated several Cloud Foundry (CF) instances --
Michal Kuratczyk |
|
|
|
Re: CF CAB call next week Wednesday December 14th @ 8a PST --- last one for 2016
Thanks Max for the reminder.
I prepared some questions into the Cab agenda google doc [1]. As there might be too many of them, or some of them being not suited for the Cab (too specific ?), participants could choose to prepare some alternative questions, or possibly vote (with "+1") for questions they'd like to be answered 1st ? [1] https://docs.google.com/document/d/1SCOlAquyUmNM-AQnekCOXiwhLs6gveTxAcduvDcW_xI/edit# Guillaume. On Mon, Dec 12, 2016 at 11:14 PM, Michael Maximilien <maxim(a)us.ibm.com> wrote: Final reminder of the last CAB call for 2016 is this coming Wednesday at |
|
|
|
App failed to startup or crashed
Phong Tran
Recently we just just consolidated several Cloud Foundry (CF) instances into one shared CF instance for resource sharing and we are now occasionally running into a problem with starting up applications . Basically when trying to start an application, we noticed one DEA is selected to run the app, but app failed to startup, so another DEA is selected. This DEA selection process would continue until the app is successfully started up on the seelcted DEA or app startup is timed out.
Starting app spring-music in org CI / space ngisspace as admin... 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 down 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 0 of 1 instances running, 1 starting 1 of 1 instances running App started OK cf logs spring-music ... 2016-12-12T15:06:50.54-0800 [DEA/10] OUT Starting app instance (index 0) with guid ccc3299d-758e-42a3-9210-8c8264d6d9a7 2016-12-12T15:08:43.91-0800 [DEA/8] OUT Starting app instance (index 0) with guid ccc3299d-758e-42a3-9210-8c8264d6d9a7 2016-12-12T15:09:24.68-0800 [App/0] OUT 2016-12-12T15:09:24.68-0800 [App/0] OUT . ____ _ __ _ _ 2016-12-12T15:09:24.68-0800 [App/0] OUT /\\ / ___'_ __ _ _(_)_ __ __ _ \ \ \ \ 2016-12-12T15:09:24.68-0800 [App/0] OUT ( ( )\___ | '_ | '_| | '_ \/ _` | \ \ \ \ 2016-12-12T15:09:24.68-0800 [App/0] OUT \\/ ___)| |_)| | | | | || (_| | ) ) ) ) 2016-12-12T15:09:24.68-0800 [App/0] OUT ' |____| .__|_| |_|_| |_\__, | / / / / 2016-12-12T15:09:24.68-0800 [App/0] OUT =========|_|==============|___/=/_/_/_/ 2016-12-12T15:09:24.68-0800 [App/0] OUT :: Spring Boot :: (v1.3.5.RELEASE) 2016-12-12T15:09:24.83-0800 [App/0] OUT 2016-12-12 23:09:24.833 INFO 29 --- [ main] pertySourceApplicationContextInitializer : Adding 'cloud' PropertySource to ApplicationContext 2016-12-12T15:09:24.90-0800 [App/0] OUT 2016-12-12 23:09:24.908 INFO 29 --- [ main] nfigurationApplicationContextInitializer : Adding cloud service auto-reconfiguration to ApplicationContext 2016-12-12T15:09:25.19-0800 [App/0] OUT 2016-12-12 23:09:25.197 INFO 29 --- [ main] .m.c.SpringApplicationContextInitializer : Found serviceInfos: 2016-12-12T15:09:25.20-0800 [App/0] OUT 2016-12-12 23:09:25.202 INFO 29 --- [ main] o.c.samples.music.Application : Starting Application on 1a37tgli287 with PID 29 (/home/vcap/app started by vcap in /home/vcap/app) ... Initially we though it could be related to the disk/memory constraint on DEAs, so we have increased more capacity, so no help here. We thought the problem be be related to our applications, so we tried to use the Spring Music sample app (https://github.com/cloudfoundry-samples/spring-music) and we still have the same startup problem (app crash) as indicated above. When looked at the dea_next.log, we could not find any obviously wrong according to our knowledge, however obvious errors except these errors: "error":"Container state must be one of 'active', current state is 'stopped'","backtrace":["/var/vcap/packages/dea_next/vendor/cache/warden-b72b8c5a532d/em-warden-client/lib/em/warden/client/connection.rb:25:in `get'","/var/vcap/packages/dea_next/vendor/cache/warden-b72b8c5a532d/em-warden-client/lib/em/warden/client.rb:46:in `call'","/var/vcap/packages/dea_next/lib/container/container.rb:193:in `call'","/var/vcap/packages/dea_next/lib/container/container.rb:133:in `block in setup_egress_rules'","/var/vcap/packages/dea_next/lib/container/container.rb:132:in `each'","/var/vcap/packages/dea_next/lib/container/container.rb:132:in `setup_egress_rules'","/var/vcap/packages/dea_next/lib/container/container.rb:126:in `block in create_container'","/var/vcap/packages/dea_next/lib/container/container.rb:235:in `call'","/var/vcap/packages/dea_next/lib/container/container.rb:235:in `with_em'","/var/vcap/packages/dea_next/lib/container/container.rb:119:in `create_container'","/var/vcap/packages/de a_next/lib/dea/starting/instance.rb:552:in `block in promise_container'","/var/vcap/packages/dea_next/lib/dea/promise.rb:92:in `call'","/var/vcap/packages/dea_next/lib/dea/promise.rb:92:in `block in run'"]},"thread_id":46964431542680,"fiber_id":46964755305680,"process_id":5695,"file":"/var/vcap/packages/dea_next/lib/dea/task.rb","lineno":105,"method":"block in resolve_and_log"} It looked like DEA failed to create a container for the app when setting up egress rules for the container (race/deadlock?), but I could be wrong here. The dea_next.log is avaialable in the following pastebin URI. dea_next.log: http://pastebin.com/PBm1RFtq We're using Cloud Foundry Release #237. Please let me me know you if you need further information for troubleshooting the problem. Thanks in advance for your help. Thanks, Phong |
|
|
|
CF CAB call next week Wednesday December 14th @ 8a PST --- last one for 2016
Michael Maximilien
Final reminder of the last CAB call for 2016 is this coming Wednesday at 8a Pacific time. All info below.
toggle quoted message
Show quoted text
Talk to you then. Best, dr.max ibm cloud labs sillicon valley, ca Sent from my iPhone Begin forwarded message: From: "Michael Maximilien" <maxim(a)us.ibm.com> |
|
|
|
Re: Using a deployment based on a Buildpack, it is possible to define multiple cores for that app?
Chip Childers <cchilders@...>
That's odd. It works for me. Did the URL get mangled on your side?
On Mon, Dec 12, 2016 at 2:02 PM Juan Antonio Breña Moral < bren(a)juanantonio.info> wrote: Hi Daniel, the link return:-- Chip Childers CTO, Cloud Foundry Foundation 1.267.250.0815 |
|
|
|
Re: Using a deployment based on a Buildpack, it is possible to define multiple cores for that app?
Juan Antonio Breña Moral <bren at juanantonio.info...>
Hi Daniel, the link return:
The requested URL /archives/list/cf-dev＠lists.cloudfoundry.org/message/B76YGMZ5DQS2WGJZLAI45LBV6ZRKVVRI/ was not found on this server. Another address? Juan Antonio |
|
|
|
Re: Using a deployment based on a Buildpack, it is possible to define multiple cores for that app?
Daniel Mikusa
No. See this email for an explanation.
https://lists.cloudfoundry.org/archives/list/cf-dev(a)lists.cloudfoundry.org/message/B76YGMZ5DQS2WGJZLAI45LBV6ZRKVVRI/ Dan On Sat, Dec 10, 2016 at 11:46 AM, Juan Antonio Breña Moral < bren(a)juanantonio.info> wrote: Hi, |
|
|
|
Re: CVE-2016-8218: Unauthenticated JWT signing algorithm in routing
Shannon Coen
Clarification:
The routing-api job is not deployed with cf-release. It is deployed with the routing-release [1], which provides support for TCP routing to apps on Diego. If you have not deployed the routing-release specifically, then chances are you do not have the routing-api job deployed and no action is required. To confirm, look at your BOSH deployments. $ bosh deployments For those who have deployed the routing-release, we recommend upgrading now to version 0.142.0 which contains a fix for the vulnerability: https://github.com/cloudfoundry-incubator/routing-release/releases/tag/0.142.0 [1] https://github.com/cloudfoundry-incubator/routing-release Best, Shannon Coen Product Manager, Cloud Foundry Pivotal, Inc. On Fri, Dec 9, 2016 at 11:00 PM, Molly Crowther <mcrowther(a)cloudfoundry.org> wrote: A new cf-release has been cut to mitigate the following issue. This notice |
|
|
|
Using a deployment based on a Buildpack, it is possible to define multiple cores for that app?
Juan Antonio Breña Moral <bren at juanantonio.info...>
Hi,
I would like to know if using CF with Buildpacks, it is possible to define the number of cores to be used by the app. With Docker, it is possible to define the number of cores with the following option: ``` bash docker run ubuntu /bin/echo 'Hello world --cpuset-cpus="0-3" ```` Juan Antonio |
|
|
|
CVE-2016-8218: Unauthenticated JWT signing algorithm in routing
Molly Crowther
A new cf-release has been cut to mitigate the following issue. This notice
can also be found at https://www.cloudfoundry.org/cve-2016-8218/. An RSS feed of Cloud Foundry vulnerabilities is available at https://www.cloudfoundry.org/category/security/feed/. CVE-2016-8218: Unauthenticated JWT signing algorithm in routing Severity Critical Vendor Cloud Foundry Foundation Versions Affected - cf-release versions prior to 237, if: - any versions of routing release have been uploaded, AND - the routing API is enabled through a manifest property on cf-release - cf-release versions 237 and later, prior to v249 Description Incomplete validation logic in JSON Web Token (JWT) libraries can allow unprivileged attackers to impersonate other users to the routing API. Mitigation OSS users of affected versions are strongly encouraged to: - Upgrade to Cloud Foundry v249 [1] or later Credit The issue was responsibly reported by a Pivotal team member. References - [1] https://github.com/cloudfoundry/cf-release/releases History2016-12-09: Initial vulnerability report published |
|
|
|
Proposal: Private Docker repository support
Will Pragnell <wpragnell@...>
Hi all,
We've recently come up with a proposal for adding support for private Docker repositories to Cloud Foundry [1]. The runtime product team spent some time considering various solutions to the problem (details of which are linked to from within the doc) and we feel that the proposed solution is the best next step, given the relative ease of implementation and the simplicity of the UX. We'd love to hear any feedback from the community. Please leave input and feedback in the comments on the document. We're hoping to begin work on this with at least some of the teams involved before the end of the year. We intend to start at the Garden / GrootFS level, as the changes needed there should be essentially the same regardless of any changes we might make to the whole feature. Thanks, Will - GrootFS project lead [1] https://goo.gl/xw2Z6V |
|
|
|
Custom health checks
John Feminella <jxf@...>
hi all,
Is there a custom health checks proposal that is being worked on or has already been accepted? If not, what is the state of the CF roadmap with respect to custom health checks? best,~ jf--John Feminella Advisory Platform Architect ✉ ·jxf(a)pivotal.io t · @jxxf |
|
|
|
Re: USN-3151-2: Linux kernel (Xenial HWE) vulnerability
Molly Crowther
Hi Marco,
toggle quoted message
Show quoted text
Thanks for the feedback - I do usually include the Ubuntu link in the postings but overlooked it this time (and just added it on the site). I agree it's annoying to have to copy-paste out the link. :) https://www.ubuntu.com/usn/usn-3151-2/ Thanks, Molly On Thu, Dec 8, 2016 at 11:35 PM, Voelz, Marco <marco.voelz(a)sap.com> wrote:
Dear Molly, |
|
|
|
Re: Memory/Disk usage information -MB vs MiB
Nicholas Calugar
Hi Ponraj,
toggle quoted message
Show quoted text
Apologies for the delay, I promise you I have been considering this query, just having trouble finding the correct response. I think I actually do need more of an example of how this is failing you, it’s not clear why presenting [1] Org Memory Usage in MB is inherently wrong when [2] App Memory is also MB. For example: cf curl "/v2/apps?q=space_guid:d2357fe4-a9f2-4c76-99d9-321b7668fde4" | grep memory "memory": 256, "memory": 1000, cf curl /v2/organizations/$(cf org system --guid)/memory_usage { "memory_usage_in_mb": 1256 } Do you think you could provide a more end-to-end example of when this goes wrong? Thanks, Nick [1] http://apidocs.cloudfoundry.org/234/organizations/retrieving_organization_memory_usage.html [2] http://apidocs.cloudfoundry.org/234/apps/creating_an_app.html -- Nicholas Calugar On December 7, 2016 at 11:16:53 AM, Ponraj E (ponraj.e(a)gmail.com) wrote:
Hi Nicholas, Anything that you expect from us? Are we missing something here? --- Ponraj |
|
|
|
Re: USN-3151-2: Linux kernel (Xenial HWE) vulnerability
Graham Bleach
On 9 December 2016 at 08:13, Guillaume Berche <bercheg(a)gmail.com> wrote:
+1thanks for pushing these security notifications to the cloudfoundrysite and this mailing list! We're currently doing this by running an app [1] in our CF that scrapes the Pivotal security page once an hour & creates events in Datadog corresponding to new CVEs. Datadog then raises a ticket in our support system. It's perhaps a bit over-engineered, but we wanted to have the events in Datadog so we could put the alerts in a dashboard etc. if we wanted. We only used the Pivotal security page because the Cloud Foundry one didn't exist then. Ideally there would be a stable machine-readable list of all CVEs relevant to the open source product. If it is RSS/atom we might be able to do what we want without writing code, using something like https://ifttt.com/ Cheers, Graham [1] https://github.com/alphagov/paas-cve-notifier |
|
|