Bug 1523354 - [starter-us-west-1] AWS API invocations from storage contributing to API call limit
Summary: [starter-us-west-1] AWS API invocations from storage contributing to API call...
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Storage
Version: 3.7.0
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: 3.9.0
Assignee: Hemant Kumar
QA Contact: Chao Yang
URL:
Whiteboard:
Depends On:
Blocks: 1523359
TreeView+ depends on / blocked
 
Reported: 2017-12-07 19:43 UTC by Justin Pierce
Modified: 2018-03-28 14:15 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
: 1523359 (view as bug list)
Environment:
Last Closed: 2018-03-28 14:14:24 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2018:0489 0 None None None 2018-03-28 14:14:59 UTC

Description Justin Pierce 2017-12-07 19:43:14 UTC
Description of problem:
Over the course of about 4 hours, the following API invocation were recorded. DescribeVolumes and AttachVolumes are major contributors to hitting the AWS API limits for our clusters.

cloud_provider:
    DescribeSubnets   1904
    DescribeInstances   110747
    CreateTags   247
    CreateVolume   2765
    DescribeVolumes   137380    <--- 
    DetachVolume   3459
    AttachVolume   159765    <---
    DeleteVolume   274
    DescribeLoadBalancers   16622

(~4700 PV's on the cluster. ~4800 PVCs)

When these limits are hit, Ops cannot perform vital management operations on the clusters. 

Version-Release number of selected component (if applicable):
v3.7.9

How reproducible:
100%

Steps to Reproduce:
1. Monitor AWS API usage on large cluster


Actual results:
AWS API limits being hit and interfering with critical Operations to manage the clusters. Amazon will not raise these limits.

Expected results:
AWS API invocations must be well under the API limits imposed by Amazon.

Comment 3 Hemant Kumar 2018-01-04 16:25:42 UTC
We have fix majority of DescribeVolume and AttachVolume calls in https://github.com/openshift/ose/pull/981 PR.

Specifically - https://github.com/openshift/ose/pull/981/commits/13570756be7d2d9119760f70465198716e6b83a5 commit ensures that we do not make calls for AttachVolume if volume is not in attachable state.

Also because starter-us-east-2 is running old version of 3.6, it does not have multiattach fix we made (https://github.com/openshift/ose/pull/877).

In a nutshell https://github.com/openshift/ose/pull/877 and https://github.com/openshift/ose/pull/981 will reduce frequency of these calls within a very reasonable limit.

Comment 4 Hemant Kumar 2018-01-18 22:56:38 UTC
This should be fixed in 3.9

Comment 6 Chao Yang 2018-01-29 09:43:15 UTC
It is passed on 
oc v3.9.0-0.31.0
kubernetes v1.9.1+a0ce1bc657
features: Basic-Auth GSSAPI Kerberos SPNEGO

Server https://ip-172-18-1-24.ec2.internal:443
openshift v3.9.0-0.31.0
kubernetes v1.9.1+a0ce1bc657

The describevolumes and attachvolumes calls is reduced in V3.9

Comment 9 errata-xmlrpc 2018-03-28 14:14:24 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2018:0489


Note You need to log in before you can comment on or make changes to this bug.