You are here
I have the latest LAPP (12) installed on Proxmox 2.1 as an OpenVZ. I proceeded to upgrade the Pg version from 8.4 to 9.1 without incident.
Today I attempted to change the IP of the turnkey from 192.168.0.61 to 192.168.1.61 within Proxmox. I couldn't change the venet, so I added another with the new address. It connected, so I removed the old venet. Then all hell broke loose.
First, I couldn't start the PostgreSQL server anymore (I'm sorry, I don't have the error message). I then tried to use turnkey-init to start over the internals, but this failed (cron error); I then remembered confconsole. When I started it, I first noticed all the text was white, for some reason. But more than that, every address was set to loopback (127.0.0.1)... If I change the IP in Proxmox, I'm able to log into to Webmin and do everything as such, but everything Pg stops working completely.
By this point, if I put everything back to the original IP address, it works — that is, I can access the database server. The confconsole still reports everything as loopback and looks strange, however.
I feel like I broke something in the delicate balance of Turnkey when I removed the original venet interface. As much as I've learned about Linux in the last six months, this is simply out of my league and I'm unsure how to begin fixing (and otherwise troubleshooting) this. Please help.
Thanks!
Confconsole doesn't work with venet
For some reason it can't see venet, probably something to do with the fact that it's not a true virtual NIC. And due to the fact that it is set on the host (ie outside of the container), even if it could see it, it couldn't configure it anyway.
Why PostgreSQL can't use the other IP is quite possibly to do with the config of that itself. Probably the most reliable way to go would be to backup/dump your DB and start again with a fresh container, setting your correct venet IP from the start. If you wish to use TKL with an upgraded version of PostgreSQL then you could create a custom template. Before you put any data in or run turnkey-init (but after updating PostgreSQL) archive the root filesystem. I don't recall the full command nor the folder paths OTTOMH but if you search the PVE forums/wiki/mailing list you will find the instructions there I'm sure.
Can you elaborate, please?
I had both a venet and veth on the CT. I left the bridged connection and swapped out the veth with the appropriate addresses (should've done this within the TK). I need the venet to be able to download the appropriate files, as I've never been able to connect the TK to the internet without bridging.
Thanks for the idea! I'll hunt around to see if I can find how to make my own OpenVZ template for Pg 9.1. I'm doing array heavy work, so I really need the amended array_agg function and array optimization done from 9.0 on.
Thanks for your help!
Add new comment