Validate Docker v1.11 #23397

vishh · 2016-03-23T21:15:56Z

Docker v1.11 rcs are starting to show up. Its time to get started with the validation.

https://github.com/docker/docker/releases/tag/v1.11.0-rc1

cc @kubernetes/sig-node

EDIT(timstclair):

TASKS:

timstclair · 2016-03-23T23:30:40Z

xref #22765 - automate this process

luxas · 2016-03-24T17:30:35Z

The really big change here is that you need four binaries instead of one: docker, runc, containerd, containerd-shim if I recall correctly.
That may cause issues

vishh · 2016-03-24T17:45:27Z

The salt script uses packages. So multiple binaries should not be an issue
I think.

On Thu, Mar 24, 2016 at 10:31 AM, Lucas Käldström notifications@github.com
wrote:

The really big change here is that you need four binaries instead of one:
docker, runc, containerd, containerd-shim if I recall correctly.
That may cause issues

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#23397 (comment)

yujuhong · 2016-03-24T17:45:40Z

We may want to put both containerd and docker into one cgroup to account for the resource usage.

timstclair · 2016-03-25T00:33:35Z

kube-proxy fails to come up. From /var/log/kube-proxy.log

Flag --resource-container has been deprecated, This feature will be removed in a later release.
I0324 23:56:52.790786       6 iptables.go:177] Could not connect to D-Bus system bus: dial unix /var/run/dbus/system_bus_socket: no such file or directory
I0324 23:56:52.790882       6 server.go:154] setting OOM scores is unsupported in this build
I0324 23:56:52.855289       6 server.go:200] Using iptables Proxier.
F0324 23:56:52.855424       6 server.go:208] Unable to create proxier: can't set sysctl net/ipv4/conf/all/route_localnet: open /proc/sys/net/ipv4/conf/all/route_localnet: read-only file system
...

yujuhong · 2016-03-25T17:02:59Z

runc by default mounts /proc/sys as read-only (https://github.com/opencontainers/runc/blob/24950964ec70f18d18e767e9141d8d952efa416b/utils.go#L131)

The privileged container will need to unmount it to write to it. I added it to the manifest and kube-proxy seems to be working after that.

EDIT: I added umount /proc/sys as part of the command in the manifest file. The pod was up and running after that (i.e., not crashing anymore).

vishh · 2016-03-25T22:08:31Z

https://github.com/docker/docker/releases/tag/v1.11.0-rc2

timstclair · 2016-03-31T00:26:27Z

Results from running e2e tests on v1.11.0-rc2:

kubernetes-e2e-gce

Summarizing 6 Failures:

[Fail] [k8s.io] Addon update [BeforeEach] should propagate add-on file changes 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/addon_update.go:208

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] ReplicationController light [It] Should scale from 1 pod to 2 pods 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Services [It] should be able to up and down services 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/service.go:248

[Fail] [k8s.io] SSH [It] should SSH to all nodes and run commands 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/ssh.go:65

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] ReplicationController light [It] Should scale from 2 pods to 1 pod using HPA version v1 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Monitoring [It] should verify monitoring pods and all cluster nodes are available on influxdb using heapster. 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/util.go:1591

Ran 161 of 268 Specs in 5971.947 seconds
FAIL! -- 155 Passed | 6 Failed | 0 Pending | 107 Skipped --- FAIL: TestE2E (5972.33s)
FAIL

kubernetes-e2e-gce-serial

Summarizing 18 Failures:

[Fail] [k8s.io] Services [It] should work after restarting apiserver [Disruptive] 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/service.go:349

[Fail] [k8s.io] Nodes [Disruptive] [k8s.io] Network when a node becomes unreachable [It] All pods on the unreachable node should be marked as NotReady upon the node turn NotReady AND all pods should be mark back to Ready when the node get back to Ready before pod eviction timeout 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/util.go:3549

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] ReplicationController [It] Should scale from 5 pods to 3 pods and from 3 to 1 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Etcd failure [Disruptive] [It] should recover from network partition with master 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/etcd_failure.go:93

[Fail] [k8s.io] DaemonRestart [Disruptive] [It] Kubelet should not restart containers across restart 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/daemon_restart.go:62

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] Deployment [It] Should scale from 5 pods to 3 pods and from 3 to 1 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Etcd failure [Disruptive] [It] should recover from SIGKILL 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/etcd_failure.go:93

[Fail] [k8s.io] Services [It] should work after restarting kube-proxy [Disruptive] 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/service.go:307

[Fail] [k8s.io] Nodes [Disruptive] [k8s.io] Network when a node becomes unreachable [It] [replication controller] recreates pods scheduled on the unreachable node AND allows scheduling of pods on a node after it rejoins the cluster 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/util.go:3549

[Fail] [k8s.io] SchedulerPredicates [Serial] [It] validates MaxPods limit number of pods that are allowed to run [Slow] 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/util.go:2159

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] ReplicationController [It] Should scale from 1 pod to 3 pods and from 3 to 5 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] Deployment [It] Should scale from 1 pod to 3 pods and from 3 to 5 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] ReplicaSet [It] Should scale from 1 pod to 3 pods and from 3 to 5 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] Kubelet [Serial] [Slow] [k8s.io] regular resource usage tracking [It] for 100 pods per node over 20m0s 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/kubelet_perf.go:72

[Fail] [k8s.io] Kubelet [Serial] [Slow] [k8s.io] regular resource usage tracking [It] for 35 pods per node over 20m0s 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/kubelet_perf.go:143

[Fail] [k8s.io] DaemonRestart [Disruptive] [It] Controller Manager should not create/delete replicas across restart 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/daemon_restart.go:62

[Fail] [k8s.io] Horizontal pod autoscaling (scale resource: CPU) [k8s.io] [Serial] [Slow] ReplicaSet [It] Should scale from 5 pods to 3 pods and from 3 to 1 and verify decision stability 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/autoscaling_utils.go:325

[Fail] [k8s.io] DaemonRestart [Disruptive] [It] Scheduler should continue assigning pods to nodes across restart 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/daemon_restart.go:62

Ran 30 of 268 Specs in 9137.840 seconds
FAIL! -- 12 Passed | 18 Failed | 0 Pending | 238 Skipped --- FAIL: TestE2E (9138.18s)
FAIL

kubernetes-e2e-gce-slow

SUCCESS! -- 9 Passed | 0 Failed | 0 Pending | 259 Skipped PASS

Lots to look into here...

vishh · 2016-04-07T18:01:50Z

FYI: RC4 is out - Linux 64bit tgz: https://test.docker.com/builds/Linux/x86_64/docker-1.11.0-rc4.tgz

timstclair · 2016-04-13T17:48:06Z

Blocker: google/cadvisor#1206

yujuhong · 2016-04-13T18:31:12Z

FYI, docker-containerd should probably be moved to the same cgroup that docker daemon is in so that we can account the resource overhead of containerd-shim, a per-container process. The containerd-shim uses about 3-5MB of memory and may be rewritten in C (xref: moby/moby#21737)

timstclair · 2016-04-14T18:13:00Z

FYI, docker v1.11 is released

yujuhong · 2016-04-15T01:06:17Z

runc by default mounts /proc/sys as read-only (https://github.com/opencontainers/runc/blob/24950964ec70f18d18e767e9141d8d952efa416b/utils.go#L131)

The privileged container will need to unmount it to write to it. I added it to the manifest and kube-proxy seems to be working after that.

EDIT: I added umount /proc/sys as part of the command in the manifest file. The pod was up and running after that (i.e., not crashing anymore).

By the way, this should was already fixed by moby/moby#21751

timstclair · 2016-04-21T00:51:39Z

Almost all tests passing, yay! Remaining failures are:

[Fail] [k8s.io] Kubelet [Serial] [Slow] [k8s.io] regular resource usage tracking [It] resource tracking for 35 pods per node 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/kubelet_perf.go:150

[Fail] [k8s.io] SchedulerPredicates [Serial] [It] validates MaxPods limit number of pods that are allowed to run [Slow] 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/framework/util.go:2091

[Fail] [k8s.io] Kubelet [Serial] [Slow] [k8s.io] regular resource usage tracking [It] resource tracking for 100 pods per node 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/kubelet_perf.go:74

[Fail] [k8s.io] Kubelet [Serial] [Slow] [k8s.io] regular resource usage tracking [It] resource tracking for 0 pods per node 
/go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/kubelet_perf.go:150

3 of those are expected increases in memory usage (as described here). I think the other issue is just that my test cluster has insufficient disk quota (fails on master as well).

EDIT: Those results are with updated cAdvisor & go-dockerclient godeps, which still need to be merged.

yujuhong · 2016-04-21T18:55:00Z

Almost all tests passing, yay!

Awesome!

3 of those are expected increases in memory usage (as described here). I think the other issue is just that my test cluster has insufficient disk quota (fails on master as well).

Just wanted to double-check: Is containerd in the same cgroup with docker? If so, we can check the memory increase and bump the limits in the test accordingly.

timothysc · 2016-05-03T19:02:26Z

/cc @eparis - opinions?

timstclair · 2016-05-03T22:59:25Z

Performed manual restart tests, no issues found.

Tested the following with no apparent issues:

Kill docker daemon (sudo pkill -9 -f /usr/bin/dock[er]) while running steady state with lots of pause pods, wait and restart (sudo /etc/init.d/docker start)
Kill docker daemon while scaling pause (gcr.io/google_containers/pause) replication controller to 0 replicas. kubectl reports pods as terminated. Restart docker daemon, containers are quickly terminated.
Kill docker daemon while scaling pause rc to 200 replicas (3 node + 1 master cluster). Pods eventually report error status. Restart docker daemon, pods eventually report running.
All of the above on the master node
All of the above killing docker-containerd (it restarts automatically, with some backoff)

timstclair · 2016-05-05T01:32:38Z

v1.9 results copied from this comment

Introduction
I benchmarked docker v1.11 using the docker micro benchmark tool.

Benchmark Environment

Cloud Provider: GCE
VM Instance: n1-standard-2 (2 vCPUs, 7.5 GB memory)
OS: Debian GNU/Linux 8.3 (jessie)
Docker version:

 Version:      1.11.1
 API version:  1.23
 Go version:   go1.5.4
 Git commit:   5604cbe
 Built:        Tue Apr 26 23:11:07 2016
 OS/Arch:      linux/amd64

1. Benchmark list/inspect with varying number of containers

Using benchmark.sh -c

Latency
- Docker 1.9.1
- Docker 1.11.1
CPU Usage
- Docker 1.9.1
- Docker 1.11.1

2. Benchmark list/inspect with varying operation intervals

Using benchmark.sh -i

ps [all=true] Latency
- Docker 1.9.1
- Docker 1.11.1
ps [all=false] Latency
- Docker 1.9.1
- Docker 1.11.1
inspect Latency
- Docker 1.9.1
- Docker 1.11.1
CPU Usage
- Docker 1.9.1
- Docker 1.11.1

3. Benchmark list/inspect with varying number of goroutines

Using benchmark.sh -r

ps [all=true] Latency
- Docker 1.9.1
- Docker 1.11.1
inspect Latency
- Docker 1.9.1
- Docker 1.11.1
CPU Usage
- Docker 1.9.1
- Docker 1.11.1

Conclusion
Latency is significantly lower for list and inspect operations under docker 1.11.1, but CPU is much spikier (smoothed usage may be similar to 1.9.1).

yujuhong · 2016-05-05T17:12:59Z

Latency is significantly lower for list and inspect operations under docker 1.11.1, but CPU is much spikier (smoothed usage may be similar to 1.9.1). I will follow up with results including docker-containerd usage.

If it helps, I remember the docker-containerd cpu usage to be quite low that it's insignificant.

timstclair · 2016-05-05T21:31:26Z

Yeah, that's consistent with my findings. I've updated the graphs to include containerd.

Random-Liu · 2016-05-05T21:52:19Z

Maybe the CPU spike is because docker processes request much faster than before, the benchmark sends more request in the same period.

vishh · 2016-05-20T21:45:13Z

We might want to bump /proc/sys/kernel/keys/root_maxbytes too. Seems to
be set to 20000 bytes by default.

On Fri, May 20, 2016 at 2:38 PM, Yu-Ju Hong notifications@github.com
wrote:

I also reproduced this with @dchen1107 https://github.com/dchen1107's
command on my node.
Bumping /proc/sys/kernel/keys/root_maxkeys worked for me, until the ip
addresses ran out.

—
You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub
#23397 (comment)

timstclair · 2016-05-20T21:49:41Z

I was able to reproduce this on a container VM GCE instance, but not on an ubuntu trusty instance (that I installed docker v1.11 on using curl -fsSL https://get.docker.com/ | sh)

Random-Liu · 2016-05-21T01:47:27Z

It's wried that docker micro benchmark #23397 (comment) didn't catch the error #23397 (comment).

I think there are 2 possible reasons:

docker micro benchmark starts at most 200 running containers in the test, which didn't hit the limit.
old docker micro benchmark didn't handle operation error properly, the errors are not reported.

kubernetes-retired/contrib#581 will add error rate statistics. I'll run new docker micro benchmark with 1.11 again.

yujuhong · 2016-05-23T18:04:43Z

@Random-Liu, 200 running containers won't hit this case. In addition to that, one needs to run the benchmark using a containerVM instance to hit this case (or it depends on the settings on your desktop). We should get your PR in to at least surface the errors when running the microbenchmark

dchen1107 · 2016-05-23T18:38:55Z

@Random-Liu and I talked about this last week, and we are extending the test suite for this.

Random-Liu · 2016-05-23T18:39:51Z

@dchen1107 ACK. Will increase the max container number in benchmark test.

vishh · 2016-05-25T20:33:11Z

@timstclair #23397 (comment) is still unresolved AFAIK.

timstclair · 2016-05-25T20:36:06Z

Ah, I missed that. Thanks.

@vishh

Automatic merge from submit-queue Attempt 2: Bump GCE containerVM to container-v1-3-v20160517 (Docker 1.11.1) again. Workaround the issue of small root_maxkeys on the debian based container-vm image, and bump our image to the new alpha version for docker 1.11.1 validation. ref: #23397 #25893 cc/ @vishh @timstclair

@vishh

Automatic merge from submit-queue Move containerd process into docker cgroup for versions >= v1.11 Addresses #23397 (comment) /cc @vishh @kubernetes/sig-node

yujuhong · 2016-06-02T18:14:35Z

Docker 1.11.2 is out: https://github.com/docker/docker/releases/tag/v1.11.2

Fixes that may be relevant:

Fix a possible deadlock on image deletion and container attach
Fix an issue preventing container to be restarted after daemon restart
Fix an issue causing docker ps to hang on CentOS when using devicemapper
Fix a bug preventing to docker exec into a container when using devicemapper

dchen1107 · 2016-07-02T01:53:32Z

@thockin reported https://github.com/docker/containerd/issues/275

bgrant0607 · 2016-08-31T02:34:27Z

@timstclair Is there anything left to do here?

timstclair · 2016-08-31T16:52:49Z

Nope, we've been defaulting to v1.11 on GCE / GKE since 1.3.

dchen1107 · 2016-12-02T17:13:09Z

Known issue of docker 1.11.X: #33882 (comment) by @mtaufen

This is theorized to be the fix in Docker: moby/moby#25523
That fix goes in with Docker 1.13.0.

vishh added priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. sig/node Categorizes an issue or PR as relevant to SIG Node. labels Mar 23, 2016

vishh added this to the next-candidate milestone Mar 23, 2016

dchen1107 assigned timstclair Mar 23, 2016

dchen1107 added the area/docker label Mar 24, 2016

timstclair mentioned this issue Mar 31, 2016

[docker v1.11.0-rc2] container info missing metadata google/cadvisor#1185

Closed

yujuhong mentioned this issue Apr 12, 2016

Validate Docker 1.10 #19720

Closed

thaJeztah mentioned this issue Apr 26, 2016

Kubernetes fails on docker 1.11 rancher/rancher#4428

Closed

bacongobbler mentioned this issue Apr 28, 2016

Bad parameters and missing X-Registry-Auth: EOF deis/controller#643

Closed

dchen1107 mentioned this issue May 20, 2016

Attempt 2: Bump GCE containerVM to container-v1-3-v20160517 (Docker 1.11.1) again. #26001

Merged

timstclair mentioned this issue May 27, 2016

Move containerd process into docker cgroup for versions >= v1.11 #26391

Merged

dchen1107 mentioned this issue May 27, 2016

Add docker performance tests for image pulling #26464

Closed

k8s-github-robot pushed a commit that referenced this issue May 28, 2016

Merge pull request #26391 from timstclair/containerd

f32b258

Automatic merge from submit-queue Move containerd process into docker cgroup for versions >= v1.11 Addresses #23397 (comment) /cc @vishh @kubernetes/sig-node

Random-Liu mentioned this issue Jun 2, 2016

Broken test run: kubelet-gce-e2e-ci - 3662 [7 failures] #26569

Closed

dchen1107 mentioned this issue Jun 8, 2016

Known issues for Kubernetes 1.3 release #27076

Closed

This was referenced Jun 17, 2016

Work around KubeProxy panic that makes Node's network not working #25543

Closed

Kernel panic when having a privileged container with docker >= 1.10 #27885

Closed

krancour mentioned this issue Jul 8, 2016

Not compatible with Docker v1.11.x deis/registry#60

Closed

timstclair closed this as completed Aug 31, 2016

Random-Liu mentioned this issue Sep 3, 2016

Validate docker v1.12 #28698

Closed

aronchick mentioned this issue Sep 24, 2016

Automate Docker Validation Process kubernetes/enhancements#57

Closed

16 tasks

dchen1107 mentioned this issue Sep 24, 2016

draft release notes for kubernetes v1.4.0 #33410

Merged

dchen1107 mentioned this issue Nov 19, 2016

[k8s.io] Reboot [Disruptive] [Feature:Reboot] each node by ordering clean reboot and ensure they function upon restart {Kubernetes e2e suite} #33874

Closed

This was referenced Feb 9, 2017

Docker release tracking supereagle/experiences#5

Open

Performance comparation between Docker v1.13 and v1.10.3 supereagle/experiences#10

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Validate Docker v1.11 #23397

Validate Docker v1.11 #23397

vishh commented Mar 23, 2016 •

edited by timstclair

Loading

timstclair commented Mar 23, 2016

luxas commented Mar 24, 2016

vishh commented Mar 24, 2016

yujuhong commented Mar 24, 2016

timstclair commented Mar 25, 2016

yujuhong commented Mar 25, 2016

vishh commented Mar 25, 2016

timstclair commented Mar 31, 2016

vishh commented Apr 7, 2016

timstclair commented Apr 13, 2016

yujuhong commented Apr 13, 2016

timstclair commented Apr 14, 2016

yujuhong commented Apr 15, 2016

timstclair commented Apr 21, 2016 •

edited

Loading

yujuhong commented Apr 21, 2016

timothysc commented May 3, 2016

timstclair commented May 3, 2016

timstclair commented May 5, 2016 •

edited

Loading

yujuhong commented May 5, 2016

timstclair commented May 5, 2016

Random-Liu commented May 5, 2016

vishh commented May 20, 2016

timstclair commented May 20, 2016

Random-Liu commented May 21, 2016 •

edited

Loading

yujuhong commented May 23, 2016

dchen1107 commented May 23, 2016

Random-Liu commented May 23, 2016

vishh commented May 25, 2016

timstclair commented May 25, 2016

yujuhong commented Jun 2, 2016 •

edited

Loading

dchen1107 commented Jul 2, 2016

bgrant0607 commented Aug 31, 2016

timstclair commented Aug 31, 2016

dchen1107 commented Dec 2, 2016

Validate Docker v1.11 #23397

Validate Docker v1.11 #23397

Comments

vishh commented Mar 23, 2016 • edited by timstclair Loading

timstclair commented Mar 23, 2016

luxas commented Mar 24, 2016

vishh commented Mar 24, 2016

yujuhong commented Mar 24, 2016

timstclair commented Mar 25, 2016

yujuhong commented Mar 25, 2016

vishh commented Mar 25, 2016

timstclair commented Mar 31, 2016

vishh commented Apr 7, 2016

timstclair commented Apr 13, 2016

yujuhong commented Apr 13, 2016

timstclair commented Apr 14, 2016

yujuhong commented Apr 15, 2016

timstclair commented Apr 21, 2016 • edited Loading

yujuhong commented Apr 21, 2016

timothysc commented May 3, 2016

timstclair commented May 3, 2016

timstclair commented May 5, 2016 • edited Loading

1. Benchmark list/inspect with varying number of containers

2. Benchmark list/inspect with varying operation intervals

3. Benchmark list/inspect with varying number of goroutines

yujuhong commented May 5, 2016

timstclair commented May 5, 2016

Random-Liu commented May 5, 2016

vishh commented May 20, 2016

timstclair commented May 20, 2016

Random-Liu commented May 21, 2016 • edited Loading

yujuhong commented May 23, 2016

dchen1107 commented May 23, 2016

Random-Liu commented May 23, 2016

vishh commented May 25, 2016

timstclair commented May 25, 2016

yujuhong commented Jun 2, 2016 • edited Loading

dchen1107 commented Jul 2, 2016

bgrant0607 commented Aug 31, 2016

timstclair commented Aug 31, 2016

dchen1107 commented Dec 2, 2016

vishh commented Mar 23, 2016 •

edited by timstclair

Loading

timstclair commented Apr 21, 2016 •

edited

Loading

timstclair commented May 5, 2016 •

edited

Loading

Random-Liu commented May 21, 2016 •

edited

Loading

yujuhong commented Jun 2, 2016 •

edited

Loading