Skip to content
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

coredns: work on tcp requests concurrently #503

Merged
merged 2 commits into from
Sep 2, 2024

Conversation

Luap99
Copy link
Member

@Luap99 Luap99 commented Sep 2, 2024

Right now for a single network all requests where processed serial and
with tcp a caller is able to block us for a long time if it just opens
the connection but sends very little or no data. To avoid this always
spawn a new task if we accept a new tcp connection.

We could do the same for udp however my testing with contrib/perf/run.sh
has shown that it slows things down as the overhead of spawning a task
is greater than the few quick simple map lookups so we only spawn where
needed. We still have to spawn when forwarding external requests as this
can take a long time.

Fixes #500

Luap99 added 2 commits August 30, 2024 18:19
Add a 3s timeout to all tcp connections, we do not want clients to keep
the connections open forever. Also do not allow more than one message
per connection.

The API is a bit weird, we first get the message then have to poll again
where it return None otherwise the reply will not be send.

Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Right now for a single network all requests where processed serial and
with tcp a caller is able to block us for a long time if it just opens
the connection but sends very little or no data. To avoid this always
spawn a new task if we accept a new tcp connection.

We could do the same for udp however my testing with contrib/perf/run.sh
has shown that it slows things down as the overhead of spawning a task
is greater than the few quick simple map lookups so we only spawn where
needed. We still have to spawn when forwarding external requests as this
can take a long time.

Fixes containers#500

Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Copy link
Contributor

openshift-ci bot commented Sep 2, 2024

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: Luap99

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 /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@openshift-ci openshift-ci bot added the approved label Sep 2, 2024
@Luap99
Copy link
Member Author

Luap99 commented Sep 2, 2024

@mheon PTAL

@baude
Copy link
Member

baude commented Sep 2, 2024

yeah! LGTM

Copy link

Tests failed. @containers/packit-build please check.

@mheon
Copy link
Member

mheon commented Sep 2, 2024

/lgtm

@openshift-ci openshift-ci bot added the lgtm label Sep 2, 2024
@openshift-merge-bot openshift-merge-bot bot merged commit 07ece81 into containers:main Sep 2, 2024
39 of 47 checks passed
@Luap99 Luap99 deleted the tcp-timeout branch September 2, 2024 13:52
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

DNS Stops working after some time in bridge network
3 participants