Warning: Can't synchronize with repository "(default)" (/home/git/ome.git does not appear to be a Git repository.). Look in the Trac log for more information.
Notice: In order to edit this ticket you need to be either: a Product Owner, The owner or the reporter of the ticket, or, in case of a Task not yet assigned, a team_member"

Task #12203 (closed)

Opened 10 years ago

Closed 10 years ago

Bug: IIS deployment scenario for release

Reported by: jamoore Owned by: kgillen
Priority: blocker Milestone: 5.0.2
Component: Deployment Version: 5.0.1
Keywords: n.a. Cc: web-team@…, cxallan, pwalczysko, sbesson
Resources: n.a. Referenced By: n.a.
References: n.a. Remaining Time: n.a.
Sprint: n.a.

Description

For each of the last couple of releases (5.0.1, 4.4.11, possibly 5.0.0) there have been some issues about whether and how we test OMERO.web on IIS. Before releasing 5.0.2, we should have a clear location, and a baseline for testing. If any bugs arise while getting the baseline set up, those should also be added to 5.0.2 as blockers so we can release with confidence.

Change History (10)

comment:1 Changed 10 years ago by bpindelski

Just a thing on the side (FYI): http://pytools.codeplex.com/wikipage?title=wfastcgi. I don't even know if it's useful. It might be even harder to set-up than our current way of doing things with IIS. But it also gives the user flexibility of configuring the URL prefix (currently it's hardcoded to /omero in components/tools/OmeroPy/src/omero_web_iis.py).

comment:2 follow-up: Changed 10 years ago by pwalczysko

@jamoore josh: considering this we said this morning with Kenny that

  • we will stick just to having the installation checked in 5.0.2 (in the sense of the "Release scenario" ).
  • We will go for similar multi-port setup just like we have on gretzky
  • the data and users preparation will be done in 5.0.3 and the pre-release testing on 5.0.3 will be partially done on this new Windows server
  • is it what was meant to happen ? hopefully we are on the same page here.

`cc @kgillen

Last edited 10 years ago by pwalczysko (previous) (diff)

comment:3 in reply to: ↑ 2 Changed 10 years ago by khgillen

@pwalczysko, @jamoore: I'd agree - we now have a Windows Server 2008 x86 running IIS 7, and this will suffice for the 'Download and Install it' testing. @pwalczysko suggests for 5.0.3 running other, full scenarios on this server (and I'd suggest a peer x64 Server 2008R2 for this as well).


Replying to pwalczysko:

@jamoore josh: considering this we said this morning with Kenny that

  • we will stick just to having the installation checked in 5.0.2 (in the sense of the "Release scenario" ).
  • We will go for similar multi-port setup just like we have on gretzky
  • the data and users preparation will be done in 5.0.3 and the pre-release testing on 5.0.3 will be partially done on this new Windows server
  • is it what was meant to happen ? hopefully we are on the same page here.

`cc @kgillen

comment:4 Changed 10 years ago by pwalczysko

@kgillen: Sorry, I found one workflow which should be checked for 5.0.2. It is https://github.com/openmicroscopy/openmicroscopy/pull/2386 - the PR has been merged already unfortunately. It is about a DropBox? on Windows.
/cc @cblackburn

Last edited 10 years ago by pwalczysko (previous) (diff)

comment:5 Changed 10 years ago by jamoore

Possibly https://github.com/openmicroscopy/openmicroscopy/pull/2448 (root-owned template paths) as well.

comment:6 Changed 10 years ago by khgillen

@pwalczysko @cblackburn - will have to consult @rleigh wrt how to test it. The Windows server with IIS is -not- in the Python 2.7 and Ice 3.5, as such it may require this setup exactly to test PR 2386

comment:7 Changed 10 years ago by rleigh

PR 2386 was about fixing some incompatibilities with Python 2.7 on Windows. Without it, the server won't start, so given a Windows system with Python 2.7 and Ice 3.5, testing should involve making sure the sever can start successfully. It should of course continue to work with Python 2.6 and Ice 3.4. And the rest of the Python code needs extensive testing on Windows: web, scripts, OmeroPy, the lot. We found the most obvious incompatibility when the server wouldn't start, but we do need to determine if there are additional problems as well which we didn't find when originally testing the PR--testing was only for the single issue identified, and was not comprehensive.

comment:8 Changed 10 years ago by khgillen

@rleigh, since https://github.com/openmicroscopy/openmicroscopy/pull/2386 was tested Initially by @bpindelski, we can assume it can be retested with his infrastructure (and is not IIS relevant). Is not https://github.com/openmicroscopy/openmicroscopy/pull/2448 also non-relevant to IIS?

If @pwalczysko or @jamoore lets me know what's needed to test https://github.com/openmicroscopy/openmicroscopy/pull/2448 (root-owned template paths) on the Windows Server at Ice33, then this can be done.

Since we've now got an IIS deployment ready for at least Installation testing, can we consider this ticket done?

comment:9 Changed 10 years ago by jamoore

@khgillen: sorry, 2448 is for the more general Windows testing as opposed to IIS. Don't mind me.

comment:10 Changed 10 years ago by jamoore

  • Resolution set to fixed
  • Status changed from new to closed

Deployments done; just needs testing now.

Note: See TracTickets for help on using tickets. You may also have a look at Agilo extensions to the ticket.

1.3.13-PRO © 2008-2011 Agilo Software all rights reserved (this page was served in: 0.66376 sec.)

We're Hiring!