Install Issue

[robs007]

OS Version: Debian 12 Node.js version: node -v Yarn version: yarn -v

Server specs 4 porc, 8 gig ram amount of vCPUs and RAM on the machine where you attempt to install

Issue I have been playing with xcp-ng for a few weeks now. I have used you script multiple times to create XO installs with no issues. Last Friday, I went to do another install on a fresh Debian 12 VM, as I have been doing, and I can't get it to work. When I run the install it tells me the "Looks like Xo-server is already running". If I select yes, it fails because it can't shut down the xo-server process because it doesn't exist. If I choose upgrade it tells me the node isn't installed.

I finally had to use the xo virtual machine on the server to get thing running. I have deleted the Deb 12 VM and created new ones and still the same thing. Strange because I have spun and killed a bunch of these up in the past few weeks doing testing and learning he process and now I'm just stuck. Any thoughts? Thanks Installation logfile Attach complete log as a file.

[ronivay]

What does pgrep -f xo-server command return on this server where you're seeing the issue? That is the command the script uses to determine if xo-server is already running. To stop it, it assumes the installation has originally been completed with a script and tries to use xo-server systemd service. So this is expected behaviour (that it fails) if eq. one installs XO by hand and then attempts to use the script, hence why it is suggested to use a fresh installation of any of the supported OS's.

[robs007]

replies 529

---

From: Roni Väyrynen @.> Sent: Thursday, June 27, 2024 7:52 AM To: ronivay/XenOrchestraInstallerUpdater @.> Cc: Rob Szabo @.>; Author @.> Subject: Re: [ronivay/XenOrchestraInstallerUpdater] Install Issue (Issue #246)

What does pgrep -f xo-server command return on this server where you're seeing the issue?

— Reply to this email directly, view it on GitHubhttps://github.com/ronivay/XenOrchestraInstallerUpdater/issues/246#issuecomment-2194478728, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AAHHGT7FSB2LFZS4KJ5YSMLZJP4I7AVCNFSM6AAAAABJ7VXXWCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJUGQ3TQNZSHA. You are receiving this because you authored the thread.Message ID: @.***>

[robs007]

Should I be loading the xen guest tools on the xo install? I noticed the pre-made VM one did not have them.

[ronivay]

Which VM are you talking about? Prebuilt VM in this repo should have the tools. Those don't have anything to do with your issue though.

[robs007]

Yeah, sorry, I see that now. I assumed you should be loading the tools when building from the script.

---

From: Roni Väyrynen @.> Sent: Friday, June 28, 2024 10:06 AM To: ronivay/XenOrchestraInstallerUpdater @.> Cc: Rob Szabo @.>; Author @.> Subject: Re: [ronivay/XenOrchestraInstallerUpdater] Install Issue (Issue #246)

Which VM are you talking about? Prebuilt VM in this repo should have the tools. Those don't have anything to do with your issue though.

— Reply to this email directly, view it on GitHubhttps://github.com/ronivay/XenOrchestraInstallerUpdater/issues/246#issuecomment-2197013253, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AAHHGT5ERKBYNKFPLBR33A3ZJVUWDAVCNFSM6AAAAABJ7VXXWCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJXGAYTGMRVGM. You are receiving this because you authored the thread.Message ID: @.***>

[robs007]

Added a new host to the pool and tried to create another XO on the new host, new VM. Same issue, this was working fines, I don't know what changed.

[ronivay]

Can you please share logs (after answering yes and when choosing update option) and the pgrep command output i mentioned earlier. Not much i can do with information so far, definitely can't reproduce and nightly test installs work flawlessly. What VM are you using as base? Do you install it from scratch using ISO or are you using some readymade image?

[robs007]

pgrep -f xo-server returns 490 Using this ISO debian-12.6.0-amd64-netinst.iso and creating a new VM using the Debian Bookworm 12 (preview) xcp-ng template. Thinking about it I was using the Debian 11 template before because there wasn't a 12. Don't think that should make a difference, but I can try and build another on with the Deb 11 template. Thanks log.txt

[ronivay]

Okay. So clearly the pgrep command finds a process that matches xo-server. If there's no node pkg installed from repository and no systemd service as those steps fail for you too, it almost feels like you're using some image that already contains XO installation but not installed by the script in this repo.

Could you run ps -fp $(pgrep -f xo-server) too and paste the output here.

[robs007]

robszabo@xo-server-2:~$ ps -p $(pgrep -f xo-server) PID TTY TIME CMD 499 ? 00:00:00 avahi-daemon

[ronivay]

robszabo@xo-server-2:~$ ps -p $(pgrep -f xo-server)

PID TTY          TIME CMD

499 ?        00:00:00 avahi-daemon

can you add f to the command, so ps -fp... This is interesting.

Looks like this is a bug. It is because of your VM's hostname containing xo-server and the pgrep then matches the avahi-daemon process which it shouldn't. I'll look into implementing a fix for this. For now you can fix this by changing your VM hostname to something else, reboot and re-run the script.

[robs007]

UID PID PPID C STIME TTY TIME CMD avahi 499 1 0 07:53 ? 00:00:00 avahi-daemon: running [xo-server-2.local]

[robs007]

So I noticed that the XO's that I had installed that worked where on Deb12.5. I tried to install from the Deb 12.5 ISO, but it upgraded to 12.6. That;'s the only difference I can see.

[ronivay]

So I noticed that the XO's that I had installed that worked where on Deb12.5. I tried to install from the Deb 12.5 ISO, but it upgraded to 12.6. That;'s the only difference I can see.

As mentioned above. This is a bug in the script and happens because of the hostname you gave to the server. Pushing a fix shortly.

[ronivay]

Fix is now pushed into place. Please re-run the install, script should auto update itself with a new fixed version.

[robs007]

Yes, running now, thanks.

[ronivay]

Great. I’ll close this as solved.