Installer: [OKD/Azure] 'openshift-install create cluster...' terminates at Blob creation (+ possible fix)

Created on 10 Aug 2020  路  4Comments  路  Source: openshift/installer

Version

4.5.0-0.okd

Platform:

Azure IPI

What happened?

The Installer (4.5 GA and later ones) teminates with the following log contents on cluster creation:

time="2020-07-24T12:08:44+02:00" level=error msg="Error: Error creating Blob \"rhcosq9903.vhd\" (Container \"vhd\" / Account \"clusterq9903\"): Error cop
y/waiting: Error copying: blobs.Client#Copy: Failure responding to request: StatusCode=400 -- Original Error: autorest/azure: Service returned an error.
Status=400 Code=\"InvalidHeaderValue\" Message=\"The value for one of the HTTP headers is not in the correct format.\\nRequestId:3e786611-501e-0001-17a2-
61a04b000000\\nTime:2020-07-24T10:07:48.3495807Z\""
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error msg="  on ../../../../tmp/openshift-install-856099604/main.tf line 168, in resource \"azurerm_storage_blob\"
 \"rhcos_image\":"
time="2020-07-24T12:08:44+02:00" level=error msg=" 168: resource \"azurerm_storage_blob\" \"rhcos_image\" {"
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error msg="Error: Error creating Blob \"bootstrap.ign\" (Container \"ignition\" / Account \"clusterq9903\"): Error
 PutPageBlob: blobs.Client#PutPageBlob: Invalid input: `input.BlobContentLengthBytes` must be aligned to a 512-byte boundary."
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error msg="  on ../../../../tmp/openshift-install-856099604/bootstrap/main.tf line 49, in resource \"azurerm_stora
ge_blob\" \"ignition\":"
time="2020-07-24T12:08:44+02:00" level=error msg="  49: resource \"azurerm_storage_blob\" \"ignition\" {"
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=error
time="2020-07-24T12:08:44+02:00" level=fatal msg="failed to fetch Cluster: failed to generate asset \"Cluster\": failed to create cluster: failed to appl
y Terraform: error(Timeout) from Infrastructure Provider: Copying the VHD to user environment was too slow, and timeout was reached for the success."

Anything else we need to know?

3891 should have fixed this, right?

I've done the following changes to make it work for me (again):

diff --git a/data/data/azure/bootstrap/main.tf b/data/data/azure/bootstrap/main.tf
index 9e7148a..9ca8633 100644
--- a/data/data/azure/bootstrap/main.tf
+++ b/data/data/azure/bootstrap/main.tf
@@ -51,7 +51,7 @@ resource "azurerm_storage_blob" "ignition" {
   source                 = local_file.ignition_bootstrap.filename
   storage_account_name   = var.storage_account.name
   storage_container_name = azurerm_storage_container.ignition.name
-  type                   = "Page"
+  type                   = "Block"
 }

 data "ignition_config" "redirect" {
diff --git a/data/data/azure/main.tf b/data/data/azure/main.tf
index 866841a..9017ed0 100644
--- a/data/data/azure/main.tf
+++ b/data/data/azure/main.tf
@@ -170,7 +170,7 @@ resource "azurerm_storage_blob" "rhcos_image" {
   storage_account_name   = azurerm_storage_account.cluster.name
   storage_container_name = azurerm_storage_container.vhd.name
   type                   = "Page"
-  source_uri             = var.azure_image_url
+  source                 = var.azure_image_url
   metadata               = map("source_uri", var.azure_image_url)
 }

References


PR #3891

Most helpful comment

All 4 comments

still, the installer for azure is broken after beta5, I have been testing this since last week and worked with @jomeier , see latest errors after fixing the main.tf bootstrap/main.tf as suggested by the above PR #3891

sponding to request: StatusCode=400 -- Original Error: autorest/azure: Service returned an error. Status=400 Code="InvalidHeaderValue" Message="The value forone of the HTTP headers is not in the correct format.\nRequestId:952de382-901e-004b-4734-721cad000000\nTime:2020-08-14T12:11:35.3266245Z"
ERROR
ERROR on ../../../tmp/openshift-install-404918262/main.tf line 168, in resource "azurerm_storage_blob" "rhcos_image":
ERROR 168: resource "azurerm_storage_blob" "rhcos_image" {
ERROR
ERRORERROR
ERROR Error: Error creating Blob "bootstrap.ign" (Container "ignition" / Account "clusterlgipx"): Error PutPageBlob: blobs.Client#PutPageBlob: Invalid input:input.BlobContentLengthBytes must be aligned to a 512-byte boundary.
ERROR
ERROR on ../../../tmp/openshift-install-404918262/bootstrap/main.tf line 49, in resource "azurerm_storage_blob" "ignition":
ERROR 49: resource "azurerm_storage_blob" "ignition" {
ERROR
ERROR
FATAL failed to fetch Cluster: failed to generate asset "Cluster": failed to create cluster: failed to apply Terraform: error(Timeout) from Infrastructure Provider: Copying the VHD to user environment was too slow, and timeout was reached for the success.

the #4054 against master, fixes the initial installation issues, however, after 1 1/2 hours still have issues with the monitoring operator and

BUG Still waiting for the cluster to initialize: Could not update imagestream "openshift/installer" (322 of 584): the server is down or not responding

DEBUG Still waiting for the cluster to initialize: Could not update imagestream "openshift/cli-artifacts" (321 of 584): the server is down or not responding
DEBUG Still waiting for the cluster to initialize: Working towards 4.5.0-0.okd-2020-08-13-174921: 85% complete
DEBUG Still waiting for the cluster to initialize: Working towards 4.5.0-0.okd-2020-08-13-174921: 86% complete
DEBUG Still waiting for the cluster to initialize: Could not update imagestream "openshift/oauth-proxy" (327 of 584): the server is down or not responding
DEBUG Still waiting for the cluster to initialize: Some cluster operators are still updating: monitoring, openshift-apiserver

it works with #4054
thx @jomeier and @walidshaari

Was this page helpful?
0 / 5 - 0 ratings