ICP 2.1.0.3 Vagrant Up - calico-node-xxxx pods CrashLoopBackOff #131
Comments
|
Reopened as the abovementioned problem STILL occurs with the latest Vagrantfile from last friday. @tpouyer I saw that you changed e.g. lxc.mount.auto settings ... any idea why I get this mount problem? |
|
Sorry missed this in my last commit. There still seems to be an issue with the prometheus node exporter pods in the two worker nodes, I'll have to look into that. |
|
👍 Many thanks for the update - 🔺 Please note: the original problem (calico-node-xxxxx in CrashLoopBackOff with |
jjasghar
pushed a commit
that referenced
this issue
Oct 23, 2023
updated NFS mount settings lines 63-66 Added 2.1.0.2 support, support for ICP docker package and printing of worker IPs. Support for ICP 2.1.0.3 add cfc-certs directory terraform vsphere HA Added info about Redhat install Updated with PR comments from edmondsw Additional PR comments More comment updates. Make IBM Cloud Private version 2.1.0.3 the default Issue #131 fix Update Vagrantfile Fixes for issue #136 Fixed variable not getting passed to the worker script changes to support SLES bootstrap Change parameter for kibana The parameter for installing kibana (as shown in https://www.ibm.com/support/knowledgecenter/en/SSBS6K_2.1.0.3/installing/config_yaml.html) is not install_kibana but kibana_install Fix ICP version check Update bootstrap.sh Added support for 3.1 of ICP Create camQuick310.sh this is for CAM 310 and ICP 310. utilizes helm chart install. Changes for ICP 3.1.0 Update vagrant doc fixup 3.1.0 install Reduce memory requirement Update docs - Reference new memory reqs - `vagrant halt` / `vagrant up` now works ;-) add remove workers (#159) Force confold etc for apt-get upgrade (#166) grub update (during `apt-get upgrade`) was trying to go interactive during `vagrant up` because of the modified grub config from an earlier step. change upgrade command to set options to use old config, and not go interactive. Updates for ICP 3.1.1 (#169) fix the missing ssh user variable issue (#168) OpenStack deployment network interface name change (#171) * Works on both new and old interface name scheme. * Fixed typo Update deploy-softlayer-ansible.md (#175) Consistent formatting for the commands. Added parameter on build to avoid apk to hang (#178) Made this small change on the playbook to include `--network host` parameter on docker build command to avoid apk to hang when trying to fetch Alpine linux image, as described on [docker-alpine GItHub issue 307](gliderlabs/docker-alpine#307). Check if resolvconf exists before trying to disable it (#179) Added an additional step to confirm if resolvconf package is installed before trying to disable it to avoid the playbook to abort, as reported on issue #160 install mcm controller and klusterlet (#185) install CAM and some cleanup (#188) Updated to use bento/ubuntu-16.04 v201808.24.0 (#161) * Updated to use bento/ubuntu-16.04 v201808.24.0 Updated to use bento/ubuntu-16.04 v201808.24.0 * Commit before merging * Update Vagrantfile Restored values after test with cache server as well changes for 3.1.2 default admin password (#190) Updated bento/ubuntu to latest version (#192) Allow SMT levels to be set for the master node (#195) management services changes (#196) Vagrant updates for ICP 3.1.2 (#197) * Updates for ICP 3.1.1 * ICP 3.1.2 updates - Changes for new requirement of 32 character password - Changes to allow setting default admin password - Update kublet settings (image-gc-low-threshold must be lower than high threshold, or kube fails to start) - Change kubectl and helm to be installed from ICP binaries - Add cloudctl install Fix x509 Error (#198) New Terraform template for PowerVC (#204) Nutanix (#205) * Initial code for deploying ICP on Nutanix hyperconverged clusters * Fix .gitignore * Fix relative directory on markup links * Fix teardown links in markup * Fix link to file in markup * Fix link to right sample file Update Openstack Terraform code structure and some cleanup (#206) Added new features to Openstack Terraform module (#207) * Added new features * Doc changes for new features common bootstrap and manage_etc_hosts to false (#209) Update deploy-vagrant.md (#211) Extra white spaces :sadpanda: icp 3.2.0 mcm installation (#218) Nutanix (#219) * Initial code for deploying ICP on Nutanix hyperconverged clusters * Fix .gitignore * Fix relative directory on markup links * Fix teardown links in markup * Fix link to file in markup * Fix link to right sample file * add support for ICP 3.1.2 and 3.2.0 fix mongodb and mysql dir permissions (#220)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
My Vagrantfile for an ICP 2.1.0.3 ran successfully, however, the final ensure_services_up step reports two calico-node-xxxxx pods on the worker nodes in a repeated CrashLoopBackOff loop.
The ICP Console shows a suspicious event for the failed calico-node-xxxxx pods right before the next restart attempt:
Warning | kubelet 192.168.27.101 | 2 | Failed | Error: failed to start container "calico-node": Error response from daemon: linux mounts: path /var/run/calico is mounted on /run but it is not a shared or slave mount
Has anyone observed this before and a clue what I can do about it?
The text was updated successfully, but these errors were encountered: