UPDATE:

The solution I've found for this problem is to run the following command inside the NodeBB installation folder

git cherry-pick 1f3fa3a90f5fc02b3b2dc950b3a340c361206cef

Doing so automatically merges the fixing commit without any further problems, making the Admin panel usable instantly after a NodeBB restart

I don't use MongoDB (my forum is on Redis), but NodeBB/NodeBB#1f3fa3a does indeed seem to solve the problem.

Is there any way for me to fix this problem while still running 1.0.3, or am I forced to wait for a 1.0.4 release?

This appears to be the exact same problem as -> https://community.nodebb.org/topic/6338/forum-crashes-when-going-to-admin-users-registration-queue

Seems like the issue NodeBB/NodeBB#3465 is back somehow

The log is exactly the same as well.

25/6 23:51 [19313] - error: TypeError: Cannot read property 'frequency' of undefined
    at Request._callback (/root/nodebb/src/user/approval.js:183:41)
    at Request.self.callback (/root/nodebb/node_modules/request/request.js:200:22)
    at emitTwo (events.js:87:13)
    at Request.emit (events.js:172:7)
    at Request.<anonymous> (/root/nodebb/node_modules/request/request.js:1067:10)
    at emitOne (events.js:82:20)
    at Request.emit (events.js:169:7)
    at IncomingMessage.<anonymous> (/root/nodebb/node_modules/request/request.js:988:12)
    at emitNone (events.js:72:20)
    at IncomingMessage.emit (events.js:166:7)
    at endReadableNT (_stream_readable.js:913:12)
    at nextTickCallbackWith2Args (node.js:442:9)
    at process._tickCallback (node.js:356:17)
TypeError: Cannot read property 'frequency' of undefined
    at Request._callback (/root/nodebb/src/user/approval.js:183:41)
    at Request.self.callback (/root/nodebb/node_modules/request/request.js:200:22)
    at emitTwo (events.js:87:13)
    at Request.emit (events.js:172:7)
    at Request.<anonymous> (/root/nodebb/node_modules/request/request.js:1067:10)
    at emitOne (events.js:82:20)
    at Request.emit (events.js:169:7)
    at IncomingMessage.<anonymous> (/root/nodebb/node_modules/request/request.js:988:12)
    at emitNone (events.js:72:20)
    at IncomingMessage.emit (events.js:166:7)
    at endReadableNT (_stream_readable.js:913:12)
    at nextTickCallbackWith2Args (node.js:442:9)
    at process._tickCallback (node.js:356:17)
25/6 23:51 [19313] - info: [app] Shutdown (SIGTERM/SIGINT) Initialised.
25/6 23:51 [19313] - info: [app] Database connection closed.
25/6 23:51 [19313] - info: [app] Web server closed to connections.
25/6 23:51 [19313] - info: [app] Shutdown complete.
[cluster] Child Process (19313) has exited (code: 1, signal: null)
[cluster] Spinning up another process...

Afterwards, the forum keeps restarting

Problem was solved by leaving blank the Cookie Domain field in Admin -> Settings -> Advanced

I'm experiencing the same problem, yet I'm not using Apache or any other reverse proxy.
The forum is hosted on a remote VPS, and accessed directly from an IP address as it currently does not have a domain.
It didn't start right away, it worked normally in the beginning. Reloads and restarts don't help.

I have NodeBB 1.0.3 as well