-
Notifications
You must be signed in to change notification settings - Fork 5.8k
fix: multiple pool reads parallelize when possible #11537
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
fd5549f
to
c51b613
Compare
Mint Automation
11537-c51b613/mint-compress-encrypt-dist-erasure.sh.log:
Deleting image on docker hub |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
minio/cmd/erasure-server-pool.go
Line 253 in b3c56b5
for i, pool := range z.serverPools { |
@klauspost I didn't do that on purpose, since there ar 8000 e some top layers that have a dependency on the implementation such as bucket replication - will get to it if no regressions were found in the current "read" only part. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. We can always look at improvements later.
* fix: metacache should only rename entries during cleanup (minio#11503) To avoid large delays in metacache cleanup, use rename instead of recursive delete calls, renames are cheaper move the content to minioMetaTmpBucket and then cleanup this folder once in 24hrs instead. If the new cache can replace an existing one, we should let it replace since that is currently being saved anyways, this avoids pile up of 1000's of metacache entires for same listing calls that are not necessary to be stored on disk. * turn off http2 for TLS setups for now (minio#11523) due to lots of issues with x/net/http2, as well as the bundled h2_bundle.go in the go runtime should be avoided for now. golang/go#23559 golang/go#42534 golang/go#43989 golang/go#33425 golang/go#29246 With collection of such issues present, it make sense to remove HTTP2 support for now * fix: save ModTime properly in disk cache (minio#11522) fix minio#11414 * fix: osinfos incomplete in case of warnings (minio#11505) The function used for getting host information (host.SensorsTemperaturesWithContext) returns warnings in some cases. Returning with error in such cases means we miss out on the other useful information already fetched (os info). If the OS info has been succesfully fetched, it should always be included in the output irrespective of whether the other data (CPU sensors, users) could be fetched or not. * fix: avoid timed value for network calls (minio#11531) additionally simply timedValue to have RWMutex to avoid concurrent calls to DiskInfo() getting serialized, this has an effect on all calls that use GetDiskInfo() on the same disks. Such as getOnlineDisks, getOnlineDisksWithoutHealing * fix: support IAM policy handling for wildcard actions (minio#11530) This PR fixes - allow 's3:versionid` as a valid conditional for Get,Put,Tags,Object locking APIs - allow additional headers missing for object APIs - allow wildcard based action matching * fix: in MultiDelete API return MalformedXML upon empty input (minio#11532) To follow S3 spec * Update yaml files to latest version RELEASE.2021-02-14T04-01-33Z * fix: multiple pool reads parallelize when possible (minio#11537) * Add support for remote tier management With this change MinIO's ILM supports transitioning objects to a remote tier. This change includes support for Azure Blob Storage, AWS S3 and Google Cloud Storage as remote tier storage backends. Co-authored-by: Poorna Krishnamoorthy <poorna@minio.io> Co-authored-by: Krishna Srinivas <krishna@minio.io> Co-authored-by: Krishnan Parthasarathi <kp@minio.io> Co-authored-by: Harshavardhana <harsha@minio.io> Co-authored-by: Poorna Krishnamoorthy <poornas@users.noreply.github.com> Co-authored-by: Shireesh Anjal <355479+anjalshireesh@users.noreply.github.com> Co-authored-by: Anis Elleuch <vadmeste@users.noreply.github.com> Co-authored-by: Minio Trusted <trusted@minio.io> Co-authored-by: Krishna Srinivas <krishna.srinivas@gmail.com> Co-authored-by: Poorna Krishnamoorthy <poorna@minio.io> Co-authored-by: Krishna Srinivas <krishna@minio.io>
Description
fix: multiple pool reads parallelize when possible
Motivation and Context
improve performance when more than one pool
is present for lookups on small files.
How to test this PR?
Nothing special all functionality should work
as is.
Types of changes
Checklist:
commit-id
orPR #
here)