Chirpstack v4 Joins Failing

Hey everyone,

I have setup a new Chirpstack v4 instance a while back, and I am getting a new batch of devices setup on it currently for the EU868 region. I seem to be having a weird issue with the devices sending the join attempts, but the accepts are not coming through, or they are being sent but the device is not actually joining. I am under the assumption there is a mismatch with the timing of the RX windows but I believe only at the time of the join. Once the device joins, it appears there aren’t any problems. This is a bigger batch of devices I am attempting to get setup, (250) so not sure if that has something to do with it. You can see the image below where the device is receiving the join accept in the join accept delay 2 window (6 seconds) but only appears to properly join when it properly receives it after 5 seconds.
Is there a latency problem that I am running into? Potentially duty cycling problem? Here are a few logs I see repeated in the gateway logs:

Fri Apr 18 21:19:16 2025 user.debug basicstation[1891]: [S2E:VERB] 101:101:101:101 diid=1024080672 [ant#0] 869.525MHz - no DC in band: txtime=21:19:21.790 free=21:19:22.842
Fri Apr 18 21:19:16 2025 user.debug basicstation[1891]: [S2E:VERB] 101:101:101:101 diid=1024080672 [ant#0] - class A has no more alternate TX time
Fri Apr 18 21:19:16 2025 user.warn basicstation[1891]: [S2E:WARN] 101:101:101:101 diid=1024080672 [ant#0] - unable to place frame

All of this seems to be happening often but kind of inconsistent. Thanks in advance!

I will state that I am using the LoRaMAC node project 1.0.3, been very stable and good for us (especially with US915). The receive windows appear to be setup as per the defaults in the 1.0.3 regional parameters. My Chirpstack Rx settings are the following.