Stash Clone via SSH on windows 2008

When we clone via SSH on our Windows 2008 server we get slow speeds around 100-120KiB/s. From where we are testing. If we put Cygwin on that same windows server, we'll get ~600KiB/s transfer speeds. We've tried installling Stash on several bare-bones servers nowm and we still get these slow speeds, but only going to Stash via SSH. The minute we use Cygwin as the SSH server, we get much better times.

As another test we installed Stash on a 2012 server, in the same subnet and it is fast. So does anyone know what about 2008 that woud cause SSH on Stash to run slow, but not on Cygwin?

6 answers

We experienced similar behaviour when using the same repo with HTTP versus SSH. There is a significant speed difference.

Is this just a known issue then? It could lead us to use another prodcut if so.

We experienced a "speed limit" for SSH transfers of 1MB/s with Stash 2.10.1 and now 10MB/s with 2.11.3. It is a suspiciously round number, seems like there is a setting somewhere we need to find.

Unfortunately, we are also running on 2.11.3 and that hasn't changed anything.

Hi Balazs and Scott,

We have a known issue with slowness on Windows environments and we have documented it:

I was wondering if that could be the case for you too after seeing your posts.

As described on our Supported Platforms document, we do not support Cygwin.

Best regards,

Thiago Bomfim

Atlassian Support - DevTools

So unfortunately, it does not solve our problems. We've made a local user with non-admin rights and start Stash with this user. There are are still slow downs. We've tried multiple bare-bones Microsoft Windows 2008 servers. We've provisioned one from the Azure cloud and the same results exists, Stash is slow, and Cygwin is fast. Run it on 2012 and Stash is fast again. Unfortuantely we cannot run it on 2012 in our production environment.

It seems somehow latency sensitive. If we run it on the local LAN, we get ~19MB of throughput, with very little other traffic. If we make a single hop over a dedicated 150Mb PIP line with 2-4ms latency, it drops down to 4MB. Then we go over our MPLS link from datacenter to datacenter with about 30ms latency and the connection drops down to the 100KB range. All the while Cygwin is exceptionally.

Suggest an answer

Log in or Join to answer
Community showcase
Piotr Plewa
Published Dec 27, 2017 in Bitbucket

Recipe: Deploying AWS Lambda functions with Bitbucket Pipelines

Bitbucket Pipelines helps me manage and automate a number of serverless deployments to AWS Lambda and this is how I do it. I'm building Node.js Lambda functions using node-lambda&nbsp...

631 views 0 4
Read article

Atlassian User Groups

Connect with like-minded Atlassian users at free events near you!

Find a group

Connect with like-minded Atlassian users at free events near you!

Find my local user group

Unfortunately there are no AUG chapters near you at the moment.

Start an AUG

You're one step closer to meeting fellow Atlassian users at your local meet up. Learn more about AUGs

Groups near you
Atlassian Team Tour

Join us on the Team Tour

We're bringing product updates and pro tips on teamwork to ten cities around the world.

Save your spot