Basic info

When managing big Proxmox VE clusters its hard to remember on which node LXC/KVM instance is located. Of course, you can just use a web-interface to find it, but I think it would be nice to have the ability to do it from the terminal.

Installation

Just put binary anywhere you need it and give execution permission to it (chmod +x). To make it available system-wide you can just put it in someplace from your $PATH (ex. /usr/bin or /usr/local/bin).

ATTENTION: You must run pve-search on the cluster node in which you want to perform a search.

Usage

To search instances across cluster simply run pve-search [GoRegexp for name], e.g. if you want to search cluster for some instance, that have "load-balancer" in its name, just run pve-search load-balancer, it will give you something like this:

+------+------+---------+-----------------+-------------+----------------+------------------------+-------------------------+---------------+---------------+
| VMID | TYPE |  STATE  |     VM NAME     |  NODE NAME  |      CPU       |      FREE MEMORY       |        FREE DISK        |    NET IN     |    NET OUT    |
+------+------+---------+-----------------+-------------+----------------+------------------------+-------------------------+---------------+---------------+
| 8104 | lxc  | running | load-balancer4  | node9       | 4.29% (24 CPU) | 12944MB/16384MB 79.00% | 7108MB/20030MB 35.48%   | 5828797.10 MB | 6431487.52 MB |
| 8103 | lxc  | running | load-balancer3  | node6       | 4.88% (24 CPU) | 12627MB/16000MB 78.92% | 7050MB/20030MB 35.20%   | 6237860.60 MB | 6855118.29 MB |
| 8105 | lxc  | running | load-balancer5  | node8       | 6.79% (16 CPU) | 12863MB/16384MB 78.51% | 7040MB/20030MB 35.14%   | 6250774.89 MB | 6868919.56 MB |
| 8109 | lxc  | stopped | load-balancer9  | node10      | 0.00% (8 CPU)  | 8192MB/8192MB 100.00%  | 20480MB/20480MB 100.00% | 0.00 MB       | 0.00 MB       |
| 8107 | lxc  | stopped | load-balancer7  | node7       | 0.00% (8 CPU)  | 8192MB/8192MB 100.00%  | 20480MB/20480MB 100.00% | 0.00 MB       | 0.00 MB       |
| 8110 | lxc  | stopped | load-balancer10 | node2       | 0.00% (8 CPU)  | 8192MB/8192MB 100.00%  | 20480MB/20480MB 100.00% | 0.00 MB       | 0.00 MB       |
| 8108 | lxc  | stopped | load-balancer8  | node5       | 0.00% (8 CPU)  | 8192MB/8192MB 100.00%  | 20480MB/20480MB 100.00% | 0.00 MB       | 0.00 MB       |
| 8102 | lxc  | running | load-balancer2  | node3       | 4.30% (24 CPU) | 16557MB/20000MB 82.78% | 7036MB/20030MB 35.13%   | 6243922.93 MB | 6867503.67 MB |
| 8101 | lxc  | running | load-balancer1  | node1       | 4.89% (24 CPU) | 12976MB/16384MB 79.20% | 6954MB/20030MB 34.72%   | 5757362.84 MB | 6688386.79 MB |
| 8106 | lxc  | running | load-balancer6  | node4       | 0.03% (8 CPU)  | 6517MB/8192MB 79.55%   | 13272MB/20030MB 66.26%  | 2481.95 MB    | 2145.17 MB    |
+------+------+---------+-----------------+-------------+----------------+------------------------+-------------------------+---------------+---------------+

You can use different sorting keys with --sort [key] flag. Valid keys are: cpu, mem, vmname, vmid, node. By default it sorts results in descending order, to sort by ascending use --asc flag. Also, you can output values in plain text, using --asc flag. You can also apply additional filters with --state --type and --vmid flag followed by goregexp. You can read about regexp syntax Here. Also, you can output results in space-separated values using --text flag. In this case, values will be in the same order, as in the table, but data values such as free memory or free disk will be in bytes instead of megabytes