In some conditions, regulator.exit may not send a signal to blocked
regulator.enter.
Let's assume we are in the critical section of regulator.exit and r.available
is equal to 0. And there are three more gorotines. One goroutine also executes
regulator.exit and waits for the lock. Rest run regulator.enter and wait for
the signal.
We send the signal, and after releasing the lock, there will be lock
contention:
1. Wait from regulator.enter
2. Lock from regulator.exit
If the winner is Lock from regulator.exit, we will not send another signal to
unlock the second Wait.
Signed-off-by: Oleg Bulatov <obulatov@redhat.com>
* Add instructions to remove also proxy_set_header Host
Add instructions to remove also proxy_set_header Host when using ELB.
In my case I only had commented out X-Real-IP, X-Forwarded-For, X-Forwarded-Proto, but not Host, and I was getting lots of retrys in Docker. Commenting the proxy_set_header Host fixed the issue, as recommended in https://github.com/moby/moby/issues/16949
Add build args. Defaults to Linux/x64 so no change to existing image, but can build for other platforms - e.g.
```
docker build --build-arg GOOS=windows -t distribution-builder:windows .
```
Signed-off-by: Elton Stoneman <elton@sixeyed.com>
Updates resumable hash implementation to Go 1.8 equivalent. This should
be a major speedup, since it includes a number of optimizations from Go
1.7.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
* Update fedora.md
add warning class to blogquote
* Update linux-postinstall.md
add warning class to blogquote
* Update ubuntu.md
add warning class to blogquote
* Update https.md
add warning class to blogquote
* Update swarm_manager_locking.md
add warning class to blogquote
* Update dockerlinks.md
add warning class to blogquote
* Update deploying.md
add warning class to blogquote
* Update deploying.md
add warning class to blogquote
* Update insecure.md
add warning class to blogquote
* Update discovery.md
add warning class to blogquote
* Update dockerd.yaml
add warning class to blogquote
* Update docker_secret_rm.yaml
add warning class to blogquote
* Update docker_service_rm.yaml
add warning class to blogquote
* Update docker_secret_rm.yaml
add warning class to blogquote
* Update scale-your-cluster.md
add warning class to blogquote
* Update resource_constraints.md
add warning class to blogquote
* Update binaries.md
add warning class to blogquote
* Update content_trust.md
add warning class to blogquote
* Update secrets.md
add warning class to blogquote
* Update index.md
add warning class to blogquote
* Update install-sandbox-2.md
add warning class to blogquote
* Update docker-toolbox.md
add warning class to blogquote
* Update index.md
add warning class to blogquote
* Update centos.md
add warning class to blogquote
* Update debian.md
add warning class to blogquote
* Update faqs.md
add linebreak after Looking for popular FAQs on Docker for Windows?
* Update install.md
add linebreake after **Already have Docker for Windows?**
* Revert "Update dockerd.yaml"
This reverts commit 3a98eb86f700ade8941483546c33f69a9dab8ac3.
* Revert "Update docker_secret_rm.yaml"
This reverts commit 5dc1e75f37033932486c11287052b7d64bf83e55.
* Revert "Update docker_service_rm.yaml"
This reverts commit a983380a5625b471f1a03f8ed2301ead72f98f1b.
* Revert "Update docker_secret_rm.yaml"
This reverts commit 4c454b883c300e26fbb056b954bb49ec2933b172.
This updates the grammar to allow more than one digest algorithm
separator, matching the regular expression and intended grammar.
Signed-off-by: Stephen J Day <stephen.day@docker.com>
It turns out that origin/master may not be up to date in CircleCI
checkouts. Fetch origin so that diffing for the vendor check can be done
correctly.
Note that this doesn't help the case of PRs against branches other than
master, but the worst case is that those will waste a few seconds doing
unnecessary vendor validation. Since those PRs are relatively rare, that
seems fine.
Signed-off-by: Aaron Lehmann <aaron.lehmann@docker.com>
