Fix transient CSE exit 9 #3230
Fix transient CSE exit 9 #3230
Conversation
ghost
assigned
parts/k8s/kubernetescustomscript.sh
Outdated
| # See https://github.com/kubernetes/kubernetes/blob/master/build/debian-hyperkube-base/Dockerfile#L25-L44 | ||
| apt_get_install 20 30 180 apt-transport-https ca-certificates iptables iproute2 socat util-linux mount ebtables ethtool init-system-helpers nfs-common ceph-common conntrack glusterfs-client ipset jq || exit $ERR_APT_INSTALL_TIMEOUT | ||
| apt_get_install 20 30 300 apt-transport-https ca-certificates iptables iproute2 socat util-linux mount ebtables ethtool init-system-helpers nfs-common ceph-common conntrack glusterfs-client ipset jq || exit $ERR_APT_INSTALL_TIMEOUT |
There was a problem hiding this comment.
@jackfrancis how do we feel about increasing the timeout here?
CecileRobertMichon
changed the title
parts/k8s/kubernetescustomscript.sh
Outdated
| echo `date`,`hostname`, apt-get_update_begin>>/opt/m | ||
| apt_get_update || exit $ERR_APT_INSTALL_TIMEOUT | ||
| echo `date`,`hostname`, apt-get_update_end>>/opt/m | ||
| # make sure walinuxagent doesn't get updated in the middle of running this script |
There was a problem hiding this comment.
nit: whitespace indent variance
|
This makes the cluster provisioning wait until cloud-init configuration has completed prior to updating the cache. A part of the cloud-init configuration includes adding apt source (templating), and it appears as though there was a race condition causing transient failures (for more information, please see #3204). |
Codecov Report
@@ Coverage Diff @@
## master #3230 +/- ##
==========================================
- Coverage 52.37% 52.31% -0.07%
==========================================
Files 103 103
Lines 15430 15458 +28
==========================================
+ Hits 8082 8087 +5
- Misses 6621 6643 +22
- Partials 727 728 +1 |
|
@jackfrancis added as PR description |
|
/approve |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: CecileRobertMichon, jackfrancis The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
* increase timeout for apt-get install * add apt_get_update * move apt_get update block * fix indents # Conflicts: # parts/k8s/kubernetescustomscript.sh
What this PR does / why we need it: This makes the cluster provisioning wait until cloud-init configuration has completed prior to updating the cache. A part of the cloud-init configuration includes adding apt source (templating), and it appears as though there was a race condition causing transient failures. (creds: @trstringer)
Which issue this PR fixes (optional, in
fixes #<issue number>(, fixes #<issue_number>, ...)format, will close that issue when PR gets merged): fixes #3204Special notes for your reviewer:
If applicable:
Release note: