Hi, Andrew. >I get the same problem if I install on a VM running on my machine. This in particular doesn't seem right. You should be able to install it at least in local vm. What kind of virtualization product do you use? You also didn't specify your linux distro on that local vm. Did you make sure it meets our [minimum requirements](https://gluu.org/docs/ce/3.0.2/installation-guide/#minimum-requirements)?

Hi, Thanks for getting back to me so quickly. I'm using Virtual Box running on Ubuntu 16.04. The VM has 8gigs of RAM and is running on a vanilla Ubuntu 16.04 that I followed the installation instructions for the server. I made a gist at https://gist.github.com/andybeak/84e1b1a57543c1cbd7fb970a9926f145 with some of the files. I think oxauth looks a bit dodgy, perhaps there are some Java setup steps missing? I increased the memory size to 16GB because I noticed there wasn't a lot free, but that hasn't helped.

>The VM has 8gigs of RAM and is running on a vanilla Ubuntu 16.04 Did you also make sure it has at least 2 CPU cores? Those error messages in logs are not good at all. Please try to remove Gluu from the local vm, then do `# apt-get clean`. After that restart the vm, and re-try installation again. If re-installation won't help, please also provide output of next commands (in the container, from the re-installed instance): 1. `# hostname` 2. `# cat /etc/hosts` 3. `# cat /etc/resolv.conf` 4. `# netstat -nlpt` 5. `# ps -aux | grep -i java` 6. `# ps -aux | grep -i slapd` 7. `# ifconfig` Next are run outside of the container on Gluu host: 1. `# dpkg -l|grep -i gluu-server` 2. `# cat /etc/*release` 3. `# free -m` 4. `# df -H` 5. `# iptables -L -n -v`

Hi, Yes - the first thing I did was check the machine config because the help files are pretty specific about the requirements. It has two processors (and stepping that up to four didn't help). I reinstalled from my "clean Ubuntu" snapshot (which I took after running the config steps on the manual page about setting up the VM). It works perfectly after the reinstallation. I'm sorry to have wasted your time on this, I really should have tried the "turn it off and back on again" solution before bothering you. I thought from looking at the errors that I had missed a Java installation step or something. I'll try run the AWS installation again too. I was running it on a t2.large (2 CPU, 8Gig RAM). Thanks a lot!

Good, at least it's not that convoluted as it seemed first. >I'll try run the AWS installation again too. I was running it on a t2.large (2 CPU, 8Gig RAM). I would also recommend to rebuild/restart this AWS vm before trying again.

I produced the error again. I tried to run setup.py to install the Asimba SAML proxy and got an HTTP 503 error message "service not available" (not sure which service though). When I ran the configuration script (setup.py) again to revert to default config I get the redirect bug. It's not the inclusion of the Asimba package because if I include it from a brand new fresh installation it works fine. It's only when I start with a default configuration and then rerun setup.py that the problem occurs. I think the problem is somewhere in running setup.py more than once. The workaround to purge the package seems okay at the moment, but what happens if we need to change the configuration later? I have no idea why this doesn't work on the Amazon image, but I'll just install manually instead of using the marketplace image.

>I produced the error again. I tried to run setup.py to install the Asimba SAML proxy and got an HTTP 503 error message "service not available" (not sure which service though). >When I ran the configuration script (setup.py) again to revert to default config I get the redirect bug. Do you mean you re-run `setup.py` to add Asimba? Doing this will most likely break a healthy instance, you also can't add components after initial install at the moment, at least not with running `setup.py` again.

Yes, I didn't realize that running setup.py would break it. That's what is happening here, thanks for the help.