ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out

An RDMA queue pair could not transition to Ready-To-Receive, so the connection never established. On RoCE this is nearly always a GID index or MTU mismatch between the two endpoints; on InfiniBand it points at the subnet manager not having configured the path. This entry explains how to confirm the cause, apply the fix, and separate it from adjacent infiniband failures.

The literal signature is "ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out". It is a communication failure associated with InfiniBand, RoCE, and RDMA queue pairs. The line identifies the failing operation or subsystem, while the surrounding evidence decides whether it is the initiating fault or a downstream symptom.

An RDMA queue pair could not transition to Ready-To-Receive, so the connection never established. On RoCE this is nearly always a GID index or MTU mismatch between the two endpoints; on InfiniBand it points at the subnet manager not having configured the path. The failure becomes visible at this call site because the operation first requires the missing resource, valid state, healthy peer, or correct result. Earlier log lines and a known-good control carry more causal value than the final wrapper exception.

• The workload stops or loses forward progress after emitting "ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out".
• A retry on the same configuration reproduces the failure because the causal state has not changed.
• The outer framework exception can hide the rank, node, allocation, or dependency that failed first.
• Increasing timeouts or reducing workload size can suppress the symptom without correcting the cause.

• InfiniBand, RoCE, and RDMA queue pairs
• production-shaped multi-accelerator workloads
• containerized and bare-metal deployments of the same stack

Apply the following checklist to a small reproduction: each box below is a positive signal that you are looking at this exact failure rather than a sibling in the same taxonomy.

• ✓Find the first occurrence of "ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out" and preserve at least 100 lines before it.
• ✓Identify which rank, node, device, or process emitted the first related warning.
• ✓ibstat / ibv_devinfo, the port must be ACTIVE with a valid LID (InfiniBand) or a valid GID (RoCE). On IB, check that a subnet manager is running (sminfo). Test the raw path with ib_write_bw between the two hosts before blaming the application.
• ✓Repeat the same input after the targeted change and require the signature to disappear.
• ✓Resume from job start only after the control passes.

ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out

Timestamps and exact values vary across runs, but the pattern. An info-level start, an early WARN, an ERROR carrying the symptom. Is the actual fingerprint you should alert on. The Denpex platform flags this combination automatically.

on RoCE, confirm both endpoints use the same GID index and RoCE version, show_gids on each host, and set NCCL_IB_GID_INDEX to match. Confirm MTU agrees on both sides. This changes the condition in the causal diagnosis instead of hiding the outer exception. The repeated control proves ownership before recovery from job start.

# Preserve evidence before restarting
NCCL_DEBUG=INFO NCCL_DEBUG_SUBSYS=INIT,NET,COLL torchrun train.py
ibstat
ethtool -S <interface>

# Find the exact signature in the complete log
rg -n -F -- "ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out" <log-file>

Adapt the snippet to your framework. The same pattern holds for PyTorch Lightning, Hugging Face Trainer, DeepSpeed, Megatron-LM, and vLLM training wrappers. Where the wrapper exposes a config flag (for examplelr_scheduler_type in Trainer), prefer the flag over the imperative API to keep the schedule declarative and reproducible.

Select the traceback or open the failed terminal, then run Denpex locally to see the initiating rank, collateral failures, exact fix, and verification command without uploading the log.

• An RDMA queue pair could not transition to Ready-To-Receive, so the connection never established. On RoCE this is nearly always a GID index or MTU mismatch between the two endpoints; on InfiniBand it points at the subnet manager not having configured the path.
• The decisive evidence is the first log line that precedes "ibv_rc_pingpong: Failed to modify QP to RTR errno 110 Connection timed out" and differs from a healthy run.
• A nearby failure remains a competing hypothesis until a control separates configuration, capacity, transport, and hardware causes.