Xen Orchestra - Community Edition - Installing with Yarn

noelweston

Thanks for the various replies. I did also try going back to the working snapshot and running the update script with "-n lts" - no different.

I have now tried removing the node_modules directory and rebuilding as in the link from Dan - again, no different.

Given that XO is the only thing on this machine, node must have been installed originally with Xen Orchestra. It could have been updated with regular OS updates, I guess, but Xen Orchestra is working at the snapshot point and only fails when I try to run the updater script.

System reports 7.6, and I was apparently able to change to lts. Running the updater again then - same results. So I'm back at my snapshot now, with XO running but unable to update...

Thanks again.

DustinB3403

@noelweston so the simple answer is export your configuration file from your current Orchestra installation build a new instance on a brand new VM and then import your configuration file.

DustinB3403

XO is designed to be disposalable

noelweston

OK, hadn't really thought of it that way.
I've built a fresh Ubuntu 18.04 VM and reinstalled, and all seems happy enough... just running my backup now to check.

Thanks for all the assistance.

Noel

momurda

I have tried this installer script at least half a dozen times over the last 2 years. Not once has it ever worked.
Today i try again and it doesnt work still.

bnrstnr

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

I have tried this installer script at least half a dozen times over the last 2 years. Not once has it ever worked.
Today i try again and it doesnt work still.

On Ubuntu 18.04?

DustinB3403

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

I have tried this installer script at least half a dozen times over the last 2 years. Not once has it ever worked.
Today i try again and it doesnt work still.

You're the first person to say that this hasn't worked for them, especially over such a length of time.

What distro are you trying this on?

Danp

@momurda "Doesn't work" isn't sufficient feedback for us to determine why you are having such issues with this script.

Perhaps you could show us the commands you used to perform the installation and the output from the SSH session.

momurda

I typed in the command on the github page. It says it completed successfully, but no xo.
Ill just use the virtual appliance since it works.

scottalanmiller

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

I typed in the command on the github page. It says it completed successfully, but no xo.
Ill just use the virtual appliance since it works.

What OS, though?

momurda

3 separate versions of ubuntu and even different minor versions of them. 16, 17 and 18 over the last two years. Never works, theres alwasys some package not found, some configuration not supported, something that keeps it from working, yet the output still says it works.

DustinB3403

@momurda are you using the mini iso by chance?

black3dynamite

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

3 separate versions of ubuntu and even different minor versions of them. 16, 17 and 18 over the last two years. Never works, theres alwasys some package not found, some configuration not supported, something that keeps it from working, yet the output still says it works.

For Ubuntu 18.04, are using the new installer or the old version?
If you are using KVM, are you deploying Ubuntu using virt-builder?

Danp

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

I typed in the command on the github page. It says it completed successfully, but no xo.

You need to dig deeper to figure out why it isn't working correctly in your environment.

For instance, from an SSH session, you could issue the command sudo systemctrl status xo-server to check and see if the service successfully started and is actvie.

To view the logs, you can enter the command sudo journalctl -u xo-server -f -n 50

momurda

@danp I know how to check error logs. The point is the installer keeps on going after errors are encountered, then says everything is ok when it is done.
Like just now, "no package redis-server available" well that seems pretty important, why not err out there and tell me? Instead it goes through and says "Your xo is now ready, go to ip address and start using it"

bnrstnr

@momurda said in Xen Orchestra - Community Edition - Installing with Yarn:

3 separate versions of ubuntu and even different minor versions of them. 16, 17 and 18 over the last two years. Never works, theres alwasys some package not found, some configuration not supported, something that keeps it from working, yet the output still says it works.

I've deployed this at least a dozen times on all of these releases and I don't recall a time that it hasn't worked :man_shrugging:

It looks like a few of us are willing to help if you have the time to post your logs...

Danp

@momurda Understood. I'm sure the error checking could be improved. However, you haven't given any concrete examples to show where it is failing. If we can't duplicate the issue, then it is difficult for us to fix.

I've never seen the issue with redis-server. Can you provide a link to the ISO you are using to perform the OS install?

bnrstnr

Sanity check here, just installed on the regular 18.04 iso

Also, for shits and giggles, just did it on the mini.iso since I've never done it before. Selected "basic server" option with nothing but apt install openssh-server and ufw allow 80/tcp

@momurda Something funky is definitely going on. Any proxy or firewall stuff that could be blocking some of the packages? I seem to remember somebody having that problem in the past, too...

momurda

It is the 1804 server iso
The first step of the script fails and cant find redis-server, it then continues on through and installs everything else without a db.
0_1532984581348_bcab9e6c-28e0-4cbb-9a83-054bd2adca7b-image.png
https://mirror.pnl.gov/releases/bionic/ubuntu-18.04.1-live-server-amd64.iso

Danp

While the script should detect the error and abort, you can't hold it responsible for something on your end that is preventing the package from being found. This is what I get when I run the same command --

sudo apt search redis-server
Sorting... Done
Full Text Search... Done
golang-github-stvp-tempredis-dev/bionic,bionic 0.0~git20160122.0.83f7aae-1 all
  Go package to start and stop temporary redis-server processes

libtest-redisserver-perl/bionic,bionic 0.21-1 all
  redis-server runner for tests

python-hiredis/bionic 0.2.0-3 amd64
  redis protocol reader for Python 2.X using hiredis

python3-hiredis/bionic 0.2.0-3 amd64
  redis protocol reader for Python using hiredis

redis/bionic,bionic 5:4.0.9-1 all
  Persistent key-value database with network interface (metapackage)

redis-server/bionic,now 5:4.0.9-1 amd64 [installed]
  Persistent key-value database with network interface