segmentation fault on fetchOffsets for unseen groups

[MMM-J]

Environment Information

• OS: Debian
• Node Version: v22.16.0
• confluent-kafka-javascript version: 1.4.1

Steps to Reproduce
The library crashes with a segmentation fault and without an informative error when calling:

const offsets = await admin.fetchOffsets({
    groupId: groupId,
});

with groupId not being used by consumers yet (or the consumer group not having committed any offsets yet)

[milindl]

Thanks for the report. I could reproduce this intermittently (for me, it happens the first time I talk to a cluster with a group id which is not existing), but not after that. The stack trace makes me suspect the issue is within the underlying librdkafka, not the javascript addon code.

Stack trace attached for further info to help while we debug this:

0x00007ffff4af5508 in rd_kafka_cgrp_op (rkcg=0x0, rktp=0x0, replyq=..., type=RD_KAFKA_OP_COORD_QUERY, err=RD_KAFKA_RESP_ERR_NOT_COORDINATOR) at rdkafka_cgrp.c:6646
6646            rd_kafka_q_enq(rkcg->rkcg_ops, rko);
(gdb) bt
#0  0x00007ffff4af5508 in rd_kafka_cgrp_op (rkcg=0x0, rktp=0x0, replyq=..., type=RD_KAFKA_OP_COORD_QUERY, err=RD_KAFKA_RESP_ERR_NOT_COORDINATOR) at rdkafka_cgrp.c:6646
#1  0x00007ffff4ab4ff7 in rd_kafka_handle_OffsetFetch (rk=0x7ffff4f86e20, rkb=0x7fffdc026530, err=RD_KAFKA_RESP_ERR_NOT_COORDINATOR, rkbuf=0x7fffcc00b2f0, request=0x0, offsets=0x7fffefa664e0,
    update_toppar=0 '\000', add_part=1 '\001', allow_retry=0 '\000') at rdkafka_request.c:1370
#2  0x00007ffff4bbc370 in rd_kafka_ListConsumerGroupOffsetsResponse_parse (rko_req=0x7fffd400a330, rko_resultp=0x7fffefa66538, reply=0x7fffcc00b2f0, errstr=0x7fffefa66540 "", errstr_size=512)
    at rdkafka_admin.c:7114
#3  0x00007ffff4b8a1d3 in rd_kafka_admin_response_parse (rko=0x7fffd400a330) at rdkafka_admin.c:943
#4  0x00007ffff4b8ae85 in rd_kafka_admin_worker (rk=0x7ffff4f86e20, rkq=0x0, rko=0x7fffd400a330) at rdkafka_admin.c:1251
#5  0x00007ffff4b8a0e1 in rd_kafka_admin_handle_response (rk=0x7ffff4f86e20, rkb=0x7fffdc026530, err=RD_KAFKA_RESP_ERR_NO_ERROR, reply=0x7fffcc00b2f0, request=0x7fffdc0604a0, opaque=0x7fffd4009eb0)
    at rdkafka_admin.c:919
#6  0x00007ffff4a94112 in rd_kafka_buf_callback (rk=0x7ffff4f86e20, rkb=0x7fffdc026530, err=RD_KAFKA_RESP_ERR_NO_ERROR, response=0x7fffcc00b2f0, request=0x7fffdc0604a0) at rdkafka_buf.c:512
#7  0x00007ffff4a93ebe in rd_kafka_buf_handle_op (rko=0x7fffcc00ac10, err=RD_KAFKA_RESP_ERR_NO_ERROR) at rdkafka_buf.c:455
#8  0x00007ffff4a9d28d in rd_kafka_op_handle_std (rk=0x7ffff4f86e20, rkq=0x7fffefa66c30, rko=0x7fffcc00ac10, cb_type=1) at rdkafka_op.c:912
#9  0x00007ffff4a9d33d in rd_kafka_op_handle (rk=0x7ffff4f86e20, rkq=0x7fffefa66c30, rko=0x7fffcc00ac10, cb_type=RD_KAFKA_Q_CB_CALLBACK, opaque=0x7ffff4f86e20, callback=0x7ffff4a47f95 <rd_kafka_poll_cb>)
    at rdkafka_op.c:952
#10 0x00007ffff4a982fc in rd_kafka_q_serve0 (rkq=0x53737e0, timeout_ms=597, max_cnt=0, cb_type=RD_KAFKA_Q_CB_CALLBACK, callback=0x0, opaque=0x0, is_consume_call=0 '\000') at rdkafka_queue.c:627
#11 0x00007ffff4a98458 in rd_kafka_q_serve (rkq=0x53737e0, timeout_ms=597, max_cnt=0, cb_type=RD_KAFKA_Q_CB_CALLBACK, callback=0x0, opaque=0x0) at rdkafka_queue.c:658
#12 0x00007ffff4a43d16 in rd_kafka_thread_main (arg=0x7ffff4f86e20) at rdkafka.c:2263
#13 0x00007ffff789c87b in start_thread (arg=<optimized out>) at ./nptl/pthread_create.c:444
#14 0x00007ffff7929c6c in clone3 () at ../sysdeps/unix/sysv/linux/x86_64/clone3.S:78