rd_tmpabuf_alloc0: rd_kafka_topic_info_new_with_rack:1860?

[IOnlyFetchBranches]

Hello, we have a fairly big (61 project) net48 (ugh) solution, that recently developed the following.

rd_tmpabuf_alloc0: rd_kafka_topic_info_new_with_rack:1860: requested size 16 + 24 > 36
rd_tmpabuf_alloc0: rd_kafka_topic_info_new_with_rack:1860: requested size 16 + 24 > 36
The program '[233052] CTR.exe' has exited with code -1073741819 (0xc0000005) 'Access violation'.
The program '[233052] CTR.exe: Program Trace' has exited with code 0 (0x0).

I know that System.Memory got updated, however this appears to be coming from libdrkafka itself ;(
We are using CHR.Avro v7 right now, but we got the same issue with v9 (latest) as well.

I don't expect anyone here to know but perhaps this rings some bells to help me poke around more with what may be causing it to not be able to allocate the buffer. A little rubber ducking, this all started after a seemingly unrelated package update, since it is such a large solution i fear one of the dependencies may be causing it. Just figured i'd rubber duck a bit on here if anyones seen something like this before ;)

Thanks so much!

[dstelljes]

Have you been able to reproduce in an empty 4.8 solution? If you’re using Confluent Platform, you can lean on your account reps as well; we recently uncovered an issue that only occurred against Confluent Platform clusters (confluentinc/librdkafka#4253) and our TAM was able to escalate internally once we had a reproduction.