[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: OS 1.0.7 eating memory like candies after halloween



We've got a couple of folks looking at this right now.  Can you add
OPENSHIFT_PROFILE=heap before your openshift master start command, let
that run (until you see memory growth), then cleanly terminate the
process (SIGTERM)?  That'll leave a memory heap file, and we can look
at what's leaking.  I *think* your issue is fixed in 1.0.8 which I
plan to push today.

On Fri, Nov 6, 2015 at 10:48 AM, Philippe Lafoucrière
<philippe lafoucriere tech-angels com> wrote:
> We reinstalled 1.0.7 2 more times since, and each time we have different
> errors :(
> Yet, there's a common fact: openshift 1.0.7 is leaking memory for us.
> The last attempt has just one simple project, deploy is failing with:
>
> I1106 14:54:09.317330       1 sti.go:171] The value of ALLOWED_UIDS is [1-]
>
> I1106 14:54:09.323512       1 docker.go:213] Image
> groulot/golang-builder:latest available locally
>
> I1106 14:54:09.323652       1 sti.go:193] Creating a new S2I builder with
> build config: "Builder Name:\t\tGolang 1.5\nBuilder
> Image:\t\tgroulot/golang-builder\nSource:\t\t<tfile:///tmp/s2i-build072360653/upload/src\nOutput>
> Image
> Tag:\t172.30.173.71:5000/tech-angels-site/tech-angels-site:latest\nEnvironment:\t\tOPENSHIFT_BUILD_NAME=tech-angels-site-3,OPENSHIFT_BUILD_NAMESPACE=tech-angels-site,OPENSHIFT_BUILD_SOURCE=https://git.tech-angels.net/tech-angels/tech-angels-site.git\nIncremental
> Build:\tdisabled\nRemove Old Build:\tdisabled\nForce
> Pull:\t\tdisabled\nQuiet:\t\t\tdisabled\nLayered
> Build:\t\tdisabled\nWorkdir:\t\t/tmp/s2i-build072360653\nDocker
> NetworkMode:\tcontainer:9a2a0fb60d884f0da7041e5f0130a02973c3935b084c43a83245dde7e2e0cc5b\nDocker
> Endpoint:<tunix:///var/run/docker.sock\n>"
>
> I1106 14:54:09.327946       1 docker.go:213] Image
> groulot/golang-builder:latest available locally
>
> I1106 14:54:09.349093       1 sti.go:134] Preparing to build
> 172.30.173.71:5000/tech-angels-site/tech-angels-site:latest
>
> I1106 14:54:25.349721       1 cleanup.go:23] Removing temporary directory
> /tmp/s2i-build072360653
>
> I1106 14:54:25.349751       1 fs.go:99] Removing directory
> '/tmp/s2i-build072360653'
>
> F1106 14:54:25.350055       1 builder.go:55] Build error: dial tcp: i/o
> timeout
>
> We have monitored RAM since the cluster setup, and it's growing slowly. We
> are using 580MB, and still growing slowly...
>
> What info can we grab to help diagnose this?
>
> Ps : same setup with 1.0.6 is fine, openshift is using ~100MB with a lot of
> services.
>
> Thanks


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]