Having a lot of DISQUEUE errors since upgrading from 1.1.0 to 1.3.0

[GuyCarmy]

Hi, we have upgraded NSQ from 1.1.0 to 1.3.0 and started having more cases of disk getting full. looks like there are many more .bad files being generated regularly compared to before.
in the logs we see:

[nsqd] 2024/06/13 10:26:43.184774 ERROR: DISKQUEUE(topic:channel) reading at 104806550 of /data/topic:channel.diskqueue.000060.dat - EOF
[nsqd] 2024/06/13 10:26:43.184787 WARNING: DISKQUEUE(topic:channel) jump to next file and saving bad file as /data/topic:channel.diskqueue.000060.dat.bad

I should add the flags we use while running nsqd:

          --max-body-size=10485760 \
          --max-msg-size=10485760 \
          --max-req-timeout=3h0m0s \
          --max-msg-timeout=1h0m0s \
          --lookupd-tcp-address=<address> \
          --data-path=/data \
          --mem-queue-size=0

would love any help on how to investigate it. nothing changed other then upgrading from 1.1.0 to 1.3.0.
we are running NSQ on Kubernetes successfully for the past few years.